Rozprawy doktorskie na temat „Perceptual Quality Assessment”

Subjective video quality is a subjective characteristic of video quality. It is concerned with how a video is perceived by the viewer and designates his or her opinion on the particular video sequence. Subjective video quality tests are quite expensive in terms of time (preparation and running) and human resources. The main objectives of this testing is how the human observes the video quality since they are the ultimate end user. There are many ways of testing the quality of the videos. We have used ITU-T Recommendation P.910.
In our research work, we have designed the tool that can be used to conduct a mass-scale level survey or subjective tests. ACR is the only method used to carry out the subjective video assessment. The test is very useful in the context of a video streaming quality. The survey can be used in various countries and sectors with low internet speeds to determine the kind of video or the compression technique, bit rate, or format that gives the best quality.
0700627491, 0760935352

Thesis (Ph. D.)--University of California, San Diego, 2007.
Title from first page of PDF file (viewed Mar. 14, 2007). Available via ProQuest Digital Dissertations. Vita. Includes bibliographical references (p. 149-156).

This thesis discusses quality assessment of audio communications systems, in particular telephone networks. A new technique for time-delay estimation based on a smoothed weighted histogram of frame-by-frame delays is presented. This has low complexity and is found to be more robust to non-linear distortions typical of telephone networks. This technique is further extended to identify piecewise constant delay, enabling models to be used for assessing packet-based transmission such as voice over IP, where delay may change several times during a measurement. It is shown that equalisation improves the accuracy of perceptual models for measurements that may include analogue or acoustic components. Linear transfer function estimation is found to be unreliable due to non-linear distortions. Spectral difference and phaseless cross-spectrum estimation methods for identifying and equalising the linear transfer function are implemented for this application, operating in the filter-bank and short-term Fourier spectrum domains. This thesis provides the first detailed examination of the process of selecting and mapping multiple objective perceptual distortion parameters to estimated subjective quality. The systematic variation of subjective opinion between tests is examined and addressed using a new method of monotonic polynomial regression. The effect on conventional regression techniques, and a new joint optimisation process, are considered.

The ever-increasing demand for good communications/toll quality speech has created a renewed interest into the perceptual impact of rate compression. Two general areas are investigated in this work, namely speech quality assessment and speech coding. In the field of speech quality assessment, a model is developed which simulates the processing stages of the peripheral auditory system. At the output of the model a "running" auditory spectrum is obtained. This represents the auditory (spectral) equivalent of any acoustic sound such as speech. Auditory spectra from coded speech segments serve as inputs to a second model. This model simulates the information centre in the brain which performs the speech quality assessment.

2007/2008
The intensity of natural light can span over 10 orders of magnitude from starlight to direct sunlight. Even in a single scene, the luminance of the bright areas can be thousands or millions of times greater than the luminance in the dark areas; the ratio between the maximum and the minimum luminance values is commonly known as dynamic range or contrast. The human visual system is able to operate in an extremely wide range of luminance conditions without saturation and at the same time it can perceive fine details which involve small luminance differences. Our eyes achieve this ability by modulating their response as a function of the local mean luminance with a process known as local adaptation. In particular, the visual sensation is not linked to the absolute luminance, but rather to its spatial and temporal variation. One consequence of the local adaptation capability of the eye is that the objects in a scene maintain their appearance even if the light source illuminating the scene changes significantly. On the other hand, the technologies used for the acquisition and reproduction of digital images are able to handle correctly a significantly smaller luminance range of 2 to 3 orders of magnitude at most. Therefore, a high dynamic range (HDR) image poses several challenges and requires the use of appropriate techniques. These elementary observations define the context in which the entire research work described in this Thesis has been performed. As indicated below, different fields have been considered; they range from the acquisition of HDR images to their display, from visual quality evaluation to medical applications, and include some developments on a recently proposed class of display equipment. An HDR image can be captured by taking multiple photographs with different exposure times or by using high dynamic range sensors; moreover, synthetic HDR images can be generated with computer graphics by means of physically-based algorithms which often involve advanced lighting simulations. An HDR image, although acquired correctly, can not be displayed on a conventional monitor. The white level of most devices is limited to a few hundred cd/m² by technological constraints, primarily linked to the power consumption and heat dissipation; the black level also has a non negligible luminance, in particular for devices based on the liquid crystal technology. However, thanks to the aforementioned properties of the human visual system, an exact reproduction of the luminance in the original scene is not strictly necessary in order to produce a similar sensation in the observer. For this purpose, dynamic range reduction algorithms have been developed which attenuate the large luminance variations in an image while preserving as far as possible the fine details. The most simple dynamic range reduction algorithms map each pixel individually with the same nonlinear function commonly known as tone mapping curve. One operator we propose, based on a modified logarithmic function, has a low computational cost and contains one single user-adjustable parameter. However, the methods belonging to this category can reduce the visibility of the details in some portions of the image. More advanced methods also take into account the pixel neighborhood. This approach can achieve a better preservation of the details, but the loss of one-to-one mapping from input luminances to display values can lead to the formation of gradient reversal effects, which typically appear as halos around the object boundaries. Different solutions to this problem have been attempted. One method we introduce is able to avoid the formation of halos and intrinsically prevents any clipping of the output display values. The method is formulated as a constrained optimization problem, which is solved efficiently by means of appropriate numerical methods. In specific applications, such as the medical one, the use of dynamic range reduction algorithms is discouraged because any artifacts introduced by the processing can lead to an incorrect diagnosis. In particular, a one-to-one mapping from the physical data (for instance, a tissue density in radiographic techniques) to the display value is often an essential requirement. For this purpose, high dynamic range displays, capable of reproducing images with a wide luminance range and possibly a higher bit depth, are under active development. Dual layer LCD displays, for instance, use two liquid crystal panels stacked one on top of the other over an enhanced backlight unit in order to achieve a dynamic range of 4 ÷ 5 orders of magnitude. The grayscale reproduction accuracy is also increased, although a “bit depth” can not be defined unambiguously because the luminance levels obtained by the combination of the two panels are partially overlapped and unevenly spaced. A dual layer LCD display, however, requires the use of complex splitting algorithms in order to generate the two images which drive the two liquid crystal panels. A splitting algorithm should compensate multiple sources of error, including the parallax introduced by the viewing angle, the gray-level clipping introduced by the limited dynamic range of the panels, the visibility of the reconstruction error, and glare effects introduced by an unwanted light scattering between the two panels. For these reasons, complex constrained optimization techniques are necessary. We propose an objective function which incorporates all the desired constraints and requirements and can be minimized efficiently by means of appropriate techniques based on multigrid methods. The quality assessment of high dynamic range images requires the development of appropriate techniques. By their own nature, dynamic range reduction algorithms change the luminance values of an image significantly and make most image fidelity metrics inapplicable. Some particular aspects of the methods can be quantified by means of appropriate operators; for instance, we introduce an expression which describes the detail attenuation introduced by a tone mapping curve. In general, a subjective quality assessment is preferably performed by means of appropriate psychophysical experiments. We conducted a set of experiments, targeted specifically at measuring the level of agreement between different users when adjusting the parameter of the modified logarithmic mapping method we propose. The experimental results show a strong correlation between the user-adjusted parameter and the image statistics, and suggest a simple technique for the automatic adjustment of this parameter. On the other hand, the quality assessment in the medical field is preferably performed by means of objective methods. In particular, task-based quality measures evaluate by means of appropriate observer studies the clinical validity of the image used to perform a specific diagnostic task. We conducted a set of observer studies following this approach, targeted specifically at measuring the clinical benefit introduced by a high dynamic range display based on the dual layer LCD technology over a conventional display with a low dynamic range and 8-bit quantization. Observer studies are often time consuming and difficult to organize; in order to increase the number of tests, the human observers can be partially replaced by appropriate software applications, known as model observers or computational observers, which simulate the diagnostic task by means of statistical classification techniques. This thesis is structured as follows. Chapter 1 contains a brief background of concepts related to the physiology of human vision and to the electronic reproduction of images. The description we make is by no means complete and is only intended to introduce some concepts which will be extensively used in the following. Chapter 2 describes the technique of high dynamic range image acquisition by means of multiple exposures. In Chapter 3 we introduce the dynamic range reduction algorithms, providing an overview of the state of the art and proposing some improvements and novel techniques. In Chapter 4 we address the topic of quality assessment in dynamic range reduction algorithms; in particular, we introduce an operator which describes the detail attenuation introduced by tone mapping curves and describe a set of psychophysical experiments we conducted for the adjustment of the parameter in the modified logarithmic mapping method we propose. In Chapter 5 we move to the topic of medical images and describe the techniques used to map the density data of radiographic images to display luminances. We point out some limitations of the current technical recommendation and propose an improvement. In Chapter 6 we describe in detail the dual layer LCD prototype and propose different splitting algorithms for the generation of the two images which drive the two liquid crystal panels. In Chapter 7 we propose one possible technique for the estimation of the equivalent bit depth of a dual layer LCD display, based on a statistical analysis of the quantization noise. Finally, in Chapter 8 we address the topic of objective quality assessment in medical images and describe a set of observer studies we conducted in order to quantify the clinical benefit introduced by a high dynamic range display. No general conclusions are offered; the breadth of the subjects has suggested to draw more focused comments at the end of the individual chapters.
XXI Ciclo
1982

Hospital and clinical environments are rapidly moving toward the digital capture, processing, storage, and transmission of medical images. X-ray cardio-angiograms are used to observe coronary blood flow, diagnose arterial disease and perform coronary angioplasty or bypass surgery. The digital storage and transmission of these cardiovascular images has significant potential to improve patient care. For example, digital images enable electronic archiving, network transmission and useful manipulation of diagnostic information such as image enhancement. The efficient compression of medical images is tremendously important for economical storage and fast transmission, since digitised medical images must be of high-quality, requiring high-resolution and having a large volume in general. The use of lossily compressed images has created a need for the development of objective quality assessment metrics I measuring perceived subjective opinions by viewers for optimal compression rate/distortion trade-off. Quality assessment metrics, based on models of the human visual system, have more accurately predicted perceived quality than traditional error-based objective quality metrics. This thesis presents a proposed Multi-stage Perceptual Quality Assessment (MPQA) model for compressed images. The motivation for the development of a perceptual quality assessment is to measure (in)visible physical differences between original and processed images. MPQA produces visible distortion maps and quantitative error measured informed by considerations of the human visual system. Original and decompressed images are decomposed into different spatial frequency bands and orientations modelling the human cortex. Contrast errors are calculated for each frequency and orientation, and masked as a function of contrast sensitivity and background uncertainty. Spatially masked contrast error measurements are made across frequency bands and orientations to produce a single Perceptual Distortion Visibility Map (PDVM). A Perceptual Quality Rating (PQR) is calculated from the PDVM and transformed into a one to five scale for direct comparison with the Mean Opinion Score (MOS), generally used in subjective rating. For medical applications, acceptable decompressed medical images might be those which are perceptually pleasing, contain no visible artefacts and have no loss in diagnostic content. To investigate this problem, clinical tests identifying diagnostically acceptable image reconstructions is performed and demonstrates that the proposed perceptual quality rating method has better agreement with observers' responses than objective error measurement methods. The vision models presented in the thesis are also implemented in the thresholding and quantisation stages of a compression algorithm. An HVS-informed perceptual thresholding and quantisation method is also shown to produce improved compression ratio performance with less visible distortions.

A dominant source of Internet traffic, today, is constituted of compressed images. In modern multimedia communications, image compression plays an important role. Some of the image compression standards set by the Joint Photographic Expert Group (JPEG) include JPEG and JPEG2000. The expert group came up with the JPEG image compression standard so that still pictures could be compressed to be sent over an e-mail, be displayed on a webpage, and make high-resolution digital photography possible. This standard was originally based on a mathematical method, used to convert a sequence of data to the frequency domain, called the Discrete Cosine Transform (DCT). In the year 2000, however, a new standard was proposed by the expert group which came to be known as JPEG2000. The difference between the two is that the latter is capable of providing better compression efficiency. There is also a downside to this new format introduced. The computation required for achieving the same sort of compression efficiency as one would get with the original JPEG format is higher. JPEG is a lossy compression standard which can throw away some less important information without causing any noticeable perception differences. Whereas, in lossless compression, the primary purpose is to reduce the number of bits required to represent the original image samples without any loss of information. The areas of application of the JPEG image compression standard include the Internet, digital cameras, printing, and scanning peripherals. In this thesis work, a simulator kind of functionality setup is needed for conducting the objective quality assessment. An image is given as an input to our wireless communication system and its data size is varied (e.g. 5%, 10%, 15%, etc) and a Signal-to-Noise Ratio (SNR) value is given as input, for JPEG2000 compression. Then, this compressed image is passed through a JPEG encoder and then transmitted over a Rayleigh fading channel. The corresponding image obtained after having applied these constraints on the original image is then decoded at the receiver and inverse discrete wavelet transform (IDWT) is applied to inverse the JPEG 2000 compression. Quantization is done for the coefficients which are scalar-quantized to reduce the number of bits to represent them, without the loss of quality of the image. Then the final image is displayed on the screen. The original input image is co-passed with the images of varying data size for an SNR value at the receiver after decoding. In particular, objective perceptual quality assessment through Structural Similarity (SSIM) index using MATLAB is provided.

Thesis (Ph. D.)--Harvard-MIT Division of Health Sciences and Technology, 2009.
Cataloged from PDF version of thesis.
Includes bibliographical references (p. 105-109).
Few empirically-derived guidelines exist for optimizing the assessment of vocal function in children with voice disorders. The goal of this investigation was to identify a minimal set of speech tasks and associated acoustic analysis methods that are most salient in characterizing the impact of vocal nodules on vocal function in children. Hence, a pediatric assessment protocol was developed based on the standardized Consensus Auditory Perceptual Evaluation of Voice (CAPE-V) used to evaluate adult voices. Adult and pediatric versions of the CAPE-V protocols were used to gather recordings of vowels and sentences from adult females and children (4-6 and 8-10 year olds) with normal voices and vocal nodules, and these recordings were subjected to perceptual and acoustic analyses. Results showed that perceptual ratings for breathiness best characterized the presence of nodules in children's voices, and ratings for the production of sentences best differentiated normal voices and voices with nodules for both children and adults. Selected voice quality-related acoustic algorithms designed to quantitatively evaluate acoustic measures of vowels and sentences, were modified to be pitch-independent for use in analyzing children's voices. Synthesized vowels for children and adults were used to validate the modified algorithms by systematically assessing the effects of manipulating the periodicity and spectral characteristics of the synthesizer's voicing source.
(cont.) In applying the validated algorithms to the recordings of subjects with normal voices and vocal nodules, the acoustic measure tended to differentiate normal voices and voices with nodules in children and adults, and some displayed significant correlations with the perceptual attributes of overall severity of dysphonia, roughness, and/or breathiness. None of the acoustic measures correlated significantly with the perceptual attribute of strain. Limitations in the strength of the correlations between acoustic measures and perceptual attributes were attributed to factors that can be addressed in future investigations, which can now utilize the algorithms that were developed in this investigation for children's voices. Preliminary recommendations are made for the clinical assessment of pediatric voice disorders.
by Asako Masaki.
Ph.D.

Submitted by Isaac Francisco de Souza Dias (isaac.souzadias@ufpe.br) on 2016-07-13T18:59:10Z No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Shifted Gradient Similarity - A perceptual video quality assessment index for adaptive streaming encoding.pdf: 5625470 bytes, checksum: 8ec1d179ec4cca056eb66609ba5791a0 (MD5)
Made available in DSpace on 2016-07-13T18:59:10Z (GMT). No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Shifted Gradient Similarity - A perceptual video quality assessment index for adaptive streaming encoding.pdf: 5625470 bytes, checksum: 8ec1d179ec4cca056eb66609ba5791a0 (MD5) Previous issue date: 2016-03-04
Adaptive video streaming has become prominent due to the rising diversity of Web-enabled personal devices and the popularity of social networks. Common limitations in Internet bandwidth, decoding speed and battery power available in such devices challenge the efficiency of content encoders to preserve visual quality at reduced data rates over a wide range of display resolutions, typically compressing to lower than 1% of the massive raw data rate. Furthermore, the human visual system does not uniformly perceive losses of spatial and temporal information, so a simple physical objective model such as the mean squared error does not correlate well with perceptual quality. Objective assessment and prediction of perceptual quality of visual content has greatly improved in the past decade, but remains an open problem. Among the most relevant psychovisual quality metrics are the many versions of the Structural Similarity (SSIM) index. In this work, several of the most efficient SSIM-based metrics, such as the Multi-Scale Fast SSIM and the Gradient Magnitude Similarity Deviation (GMSD), are decomposed into their component techniques and reassembled in order to measure and understand the contribution of each technique and to develop improvements in quality and efficiency. The metrics are applied to the LIVE Mobile Video Quality and TID2008 databases and the results are correlated to the subjective data included in the databases in the form of mean opinion scores (MOS), so each metric’s degree of correlation indicates its ability to predict perceptual quality. Additionally, the metrics’ applicability to the recent, relevant psychovisal rate-distortion optimization (Psy-RDO) implementation in the x264 encoder, which currently lacks an ideal objective assessment metric, is investigated as well. The “Shifted Gradient Similarity” (SG-Sim) index is proposed with an improved feature enhancement by avoiding a common unintended loss of analysis information in SSIM-based indexes, and achieving considerably higher MOS correlation than the existing metrics investigated in this work. More efficient spatial pooling filters are proposed, as well: the decomposed 1-D integer Gaussian filter limited to two standard deviations, and the downsampling Box filter based on the integral image, which retain respectively 99% and 98% equivalence and achieve speed gains of respectively 68% and 382%. In addition, the downsampling filter also enables broader scalability, particularly for Ultra High Definition content, and defines the “Fast SG-Sim” index version. Furthermore, SG-Sim is found to improve correlation with Psy-RDO, as an ideal encoding quality metric for x264. Finally, the algorithms and experiments used in this work are implemented in the “Video Quality Assessment in Java” (jVQA) software, based on the AviSynth and FFmpeg platforms, and designed for customization and extensibility, supporting 4K Ultra-HD content and available as free, open source code.
Cada vez mais serviços de streaming de vídeo estão migrando para o modelo adaptativo, devido à crescente diversidade de dispositivos pessoais conectados à Web e à popularidade das redes sociais. Limitações comuns na largura de banda de Internet, velocidade de decodificação e potência de baterias disponíveis em tais dispositivos desafiam a eficiência dos codificadores de conteúdo para preservar a qualidade visual em taxas de dados reduzidas e abrangendo uma ampla gama de resoluções de tela, tipicamente comprimindo para menos de 1% da massiva taxa de dados bruta. Ademais, o sistema visual humano não percebe uniformemente as perdas de informação espacial e temporal, então um modelo objetivo físico simples como a média do erro quadrático não se correlaciona bem com qualidade perceptível. Técnicas de avaliação e predição objetiva de qualidade perceptível de conteúdo visual se aprimoraram amplamente na última década, mas o problema permanece em aberto. Dentre as métricas de qualidade psicovisual mais relevantes estão muitas versões do índice de similaridade estrutural (Structural Similarity — SSIM). No presente trabalho, várias das mais eficientes métricas baseadas em SSIM, como o Multi-Scale Fast SSIM e o Gradient Magnitude Similarity Deviation (GMSD), são decompostas em suas técnicas-componentes e recombinadas para se obter medidas e entendimento sobre a contribuição de cada técnica e se desenvolver aprimoramentos à sua qualidade e eficiência. Tais métricas são aplicadas às bases de dados LIVE Mobile Video Quality e TID2008 e os resultados são correlacionados aos dados subjetivos incluídos naquelas bases na forma de escores de opinião subjetiva (mean opinion score — MOS), de modo que o grau de correlação de cada métrica indique sua capacidade de predizer qualidade perceptível. Investiga-se, ainda, a aplicabilidade das métricas à recente e relevante implementação de otimização psicovisual de distorção por taxa (psychovisual rate-distortion optimization — Psy-RDO) do codificador x264, ao qual atualmente falta uma métrica de avaliação objetiva ideal. O índice “Shifted Gradient Similarity” (SG-Sim) é proposto com uma técnica aprimorada de realce de imagem que evita uma perda não-pretendida de informação de análise, comum em índices baseados em SSIM, assim alcançando correlação consideravelmente maior com MOS comparado às métricas existentes investigadas neste trabalho. Também são propostos filtros de consolidação espacial mais eficientes: o filtro gaussiano de inteiros 1-D decomposto e limitado a dois desvios padrão e o filtro “box” subamostrado baseado na imagem integral, os quais retém, respectivamente, 99% e 98% de equivalência e obtém ganhos de velocidade de, respectivamente, 68% e 382%. O filtro subamostrado também promove escalabilidade, especialmente para conteúdo de ultra-alta definição, e define a versão do índice “Fast SG-Sim”. Ademais, verifica-se que o SG-Sim aumenta a correlação com Psy-RDO, indicando-se uma métrica de qualidade de codificação ideal para o x264. Finalmente, os algoritmos e experimentos usados neste trabalho estão implementados no software “Video Quality Assessment in Java” (jVQA), baseado nas plataformas AviSynth e FFmpeg e que é projetado para personalização e extensibilidade, suportando conteúdo ultra-alta definição “4K” e disponibilizado como código-fonte aberto e livre.

Measurement of visual quality is crucial for various image and video processing applications.

The goal of objective image quality assessment is to introduce a computational quality metric that can predict image or video quality. Many methods have been proposed in the past decades. Traditionally, measurements convert the spatial data into some other feature domains, such as the Fourier domain, and detect the similarity, such as mean square distance or Minkowsky distance, between the test data and the reference or perfect data, however only limited success has been achieved. None of the complicated metrics show any great advantage over other existing metrics.

The common idea shared among many proposed objective quality metrics is that human visual error sensitivities vary in different spatial and temporal frequency and directional channels. In this thesis, image quality assessment is approached by proposing a novel framework to compute the lost information in each channel not the similarities as used in previous methods. Based on natural scene statistics and several image models, an information theoretic framework is designed to compute the perceptual information contained in images and evaluate image quality in the form of entropy.

The thesis is organized as follows. Chapter I give a general introduction about previous work in this research area and a brief description of the human visual system. In Chapter II statistical models for natural scenes are reviewed. Chapter III proposes the core ideas about the computation of the perceptual information contained in the images. In Chapter IV, information theoretic criteria for image quality assessment are defined. Chapter V presents the simulation results in detail. In the last chapter, future direction and improvements of this research are discussed.

As there is a rapid growth in the field of wireless communications, the demand for various multimedia services is also increasing. The data that is being transmitted suffers from distortions through source encoding and transmission over errorprone channels. Due to these errors, the quality of the content is degraded. There is a need for service providers to provide certain Quality of Experience (QoE) to the end user. Several methods are being developed by network providers for better QoE.The human tendency mainly focuses on distortions in the Region of Interest(ROI) which are perceived to be more annoying compared to the Background(BG). With this as a base, the main aim of this thesis is to get an accurate prediction quality metric to measure the quality of the image over ROI and the BG independently. Reduced Reference Image Quality Assessment (RRIQA), a reduced reference image quality assessment metric, is chosen for this purpose. In this method, only partial information about the reference image is available to assess the quality. The quality metric is measured independently over ROI and BG. Finally the metric estimated over ROI and BG are pooled together to get aROI aware metric to predict the Mean Opinion Score (MOS) of the image.In this thesis, an ROI aware quality metric is used to measure the quality of distorted images that are generated using a wireless channel. The MOS of distorted images are obtained. Finally, the obtained MOS are validated with the MOS obtained from a database [1].It is observed that the proposed image quality assessment method provides better results compared to the traditional approach. It also gives a better performance over a wide variety of distortions. The obtained results show that the impairments in ROI are perceived to be more annoying when compared to the BG.

Perceptual assessment of voice in people with surgical voice restoration (SVR) is essential to evaluate surgical and other interventions aimed at delivering optimal voice quality. Currently there are no tools to measure this that do not have issues of validity and reliability. This work describes the development and trialling of investigatory versions of three scales to address this situation: a) the Sunderland Tracheoesophageal Perceptual Scale (SToPS) for professional raters, b) the Naïve Rater Scale for non-specialist raters and c) the Patient and Carer Scale. In the final testing of the pilot version 55 speakers using tracheoesophageal voice were evaluated by twelve Speech and Language Therapists (SLT’s) and ten Ear, Nose and Throat (ENT) surgeons, divided into experienced or not at assessing voice. Ten naïve raters assessed the voice stimuli within a test-retest design. Forty tracheoesophageal speakers and thirty-seven carers attended an interview to rate their own or their relative’s voice. Inter rater agreement was then calculated between SLT, ENT, naïve, patient and carer groups with weighted kappa co-efficients Strength of agreement values (Landis and Koch 1977) were compared to profession and expertise. Expert SLT’s achieved “good” agreement for nine of fourteen parameters. Naïve judges attained “good” levels of inter and intra-rater agreement for the parameters Overall Grade and Social Acceptability. The greatest inter group consensus was for patients and carers, with “good” agreement for Intelligibility, Volume and Wetness. The only other “good” agreement was between naïve/ENT and naïve/ SLT groups for Overall Grade. The scales are ready for clinical use with the proviso that future work will determine whether it is possible to enhance agreement so less experienced judges can achieve “good” levels of agreement for more parameters and examine which perceptual parameters might be more prominent or vital for outcomes for different groups.

Ce travail vise à mieux évaluer la qualité perceptuelle des images contenant des distorsions structurelles et géométriques notamment dans le contexte de médias immersifs. Nous proposons et explorons un cadre algorithmique hiérarchique de la perception visuelle. Inspiré par le système visuel humain, nous investiguons plusieurs niveaux de représentations des images : bas niveau (caractéristiques élémentaires comme les segments), niveau intermédiaire (motif complexe, encodage de contours), haut niveau (abstraction et reconnaissance des données visuelles). La première partie du manuscrit traite des représentations bas niveau pour la structure et texture. U n modèle basé filtre bilatéral est d’abord introduit pour qualifier les rôles respectifs de l’information texturale et structurelle dans diverses tâches d’évaluation (utilité, qualité. . . ). Une mesure de qualité d’image/vidéo est proposée pour quantifier les déformations de structure spatiales et temporelles perçues en utilisant une métrique dite élastique. La seconde partie du mémoire explore les représentations de niveaux intermédiaires. Un modèle basé « schetch token » et un autre basé sur codage d’un arbre de contexte sont présentés pour évaluer la qualité perçue. La troisième partie traite des représentations haut niveau. Deux approches d’apprentissage machine sont proposées pour apprendre ces représentations : une basée sur un technique de convolutional sparse coding, l’autre sur des réseaux profonds de type generative adversarial network. Au long du manuscrit, plusieurs expériences sont menées sur différentes bases de données pour plusieurs applications (FTV, visualisation multi-vues, images panoramiques 360. . . ) ainsi que des études utilisateurs
This work aims to better evaluate the perceptual quality of image/video that contains structural and geometric related distortions in the context of immersive multimedia. We propose and explore a hierarchical framework of visual perception for image/video. Inspired by representation mechanism of the visual system, low-level (elementary visual features, e.g. edges), mid-level (intermediate visual patterns, e.g. codebook of edges), and higher-level (abstraction of visual input, e.g. category of distorted edges) image/video representations are investigated for quality assessment. The first part of this thesis addresses the low-level structure and texture related representations. A bilateral filter-based model is first introduced to qualify the respective role of structure and texture information in various assessment tasks (utility, quality . . . ). An image quality/video quality measure is proposed to quantify structure deformation spatially and temporally using new elastic metric. The second part explores mid-level structure related representations. A sketch-token based model and a context tree based model are presented in this part for the image and video quality evaluation. The third part explores higher-level structure related representations. Two machine learning approaches are proposed to learn higher-level representation: a convolutional sparse coding based and a generative adversarial network. Along the thesis, experiments an user studies have been conducted on different databases for different applications where special structure related distortions are observed (FTV, multi-view rendering, omni directional imaging . . . )

A avaliação de qualidade de segmentação de vídeos tem se mostrado um problema pouco investigado no meio científico. Apesar disso, estudos recentes na área resultaram em algumas métricas que têm como finalidade avaliar objetivamente a qualidade da segmentação produzida pelos algoritmos. Tais métricas consideram as diferentes formas em que os erros ocorrem (fatores perceptuais) e seus parâmetros são ajustados de acordo com a aplicação em que se pretende utilizar os vídeos segmentados. Neste trabalho apresentam-se: i) uma avaliação da métrica que representa o estado-da-arte, demonstrando que seu desempenho varia de acordo com o algoritmo; ii) um método subjetivo para avaliação de qualidade de segmentação; e iii) uma nova métrica perceptual objetiva, derivada do método subjetivo aqui proposto, capaz de encontrar o melhor ajuste dos parâmetros de dois algoritmos de segmentação encontrados na literatura, quando os vídeos por eles segmentados são utilizados na composição de cenas em ambientes de Teleconferência Imersiva.
Assessment of video segmentation quality is a problem seldom investigated by the scientific community. Nevertheless, recent studies presented some objective metrics to evaluate algorithms. Such metrics consider different ways in which segmentation errors occur (perceptual factors) and its parameters are adjusted according to the application for which the segmented frames are intended. In this work: i) we demonstrate empirically that the performance of existing metrics changes according to the segmentation algorithm; ii) we developed a subjective method to evaluate segmentation quality; and iii) we contribute with a new objective metric derived on the basis of experiments from subjective method in order to adjust the parameters of two bilayer segmentation algorithms found in the literature when these algorithms are used for compose scenes in Immersive Teleconference environments.

Call quality has been crucial from the inception of telecommunication networks. Operators need to monitor call quality from the end-user's perspective, in order to retain subscribers and reduce subscriber 'churn'. Operators worry not only about call quality and interconnect revenue loss, but also about network connectivity issues in areas where mobile network gateways are prevalent. Bandwidth quality as experienced by the end-user is equally important in helping operators to reduce churn. The parameters that network operators use to improve call quality are mainly from the end-user's perspective. These parameters are usually ASR (answer seizure ratio), PDD (postdial delay), NER (network efficiency ratio), the number of calls for which these parameters have been analyzed and successful calls. Operators use these parameters to evaluate and optimize the network to meet their quality requirements. Analysis of speech quality is a major arena for research. Traditionally, users' perception of speech quality has been measured offline using subjective listening tests. Such tests are, however, slow, tedious and costly. An alternative method is therefore needed; one that can be automatically computed on the subscriber's handset, be available to the operator as well as to subscribers and, at the same time, provide results that are comparable with conventional subjective scores. QMeter® 'a set of tools for signal and bandwidth measurement that have been developed bearing in mind all the parameters that influence call and bandwidth quality experienced by the end-user' addresses these issues and, additionally, facilitates dynamic tariff propositions which enhance the credibility of the operator. This research focuses on call quality parameters from the end-user's perspective. The call parameters used in the research are signal strength, successful call rate, normal drop call rate, and hand-over drop rate. Signal strength is measured for every five milliseconds of an active call and average signal strength is calculated for each successful call. The successful call rate, normal drop rate and hand-over drop rate are used to achieve a measurement of the overall call quality. Call quality with respect to bundles of 10 calls is proposed. An attempt is made to visualize these parameters for better understanding of where the quality is bad, good and excellent. This will help operators, as well as user groups, to measure quality and coverage. Operators boast about their bandwidth but in reality, to know the locations where speed has to be improved, they need a tool that can effectively measure speed from the end-user's perspective. BM (bandwidth meter), a tool developed as a part of this research, measures the average speed of data sessions and stores the information for analysis at different locations. To address issues of quality in the subscriber segment, this research proposes the varying of tariffs based on call and bandwidth quality. Call charging based on call quality as perceived by the end-user is proposed, both to satisfy subscribers and help operators to improve customer satisfaction and increase average revenue per user. Tariff redemption procedures are put forward for bundles of 10 calls and 10 data sessions. In addition to the varying of tariffs, quality escalation processes are proposed. Deploying such tools on selected or random samples of users will result in substantial improvement in user loyalty which, in turn, will bring operational and economic advantages.

Le domaine de recherche dans l'évaluation objective de la qualité des images couleur a connu un regain d'intérêt ces dernières années. Les travaux sont essentiellement dictés par l'avènement des images numérique et par les nouveaux besoins en codage d'images (compression, transmission, restauration, indexation,…). Jusqu'à présent la meilleure évaluation reste visuelle (donc subjective) soit par des techniques psychophysiques soit par évaluation experte. Donc, il est utile, voire nécessaire, de mettre en place des critères et des mesures objectifs qui produisent automatiquement des notes de qualité se rapprochant le plus possible des notes de qualité données par l'évaluation subjective. Nous proposons, tout d'abort, une nouvelle métrique avec référence d'évaluation de la qualité des images couleur, nommée Delta E globale, se base sur l'aspect couleur et intègre les caractéristiques du système visuel humain (SVH). Les performances ont été mesurées dans deux domaines d'application la compression et la restauration. Les expérimentations réalisées montrent une corrélation importante entre les résultats obtenus et l'appréciation subjective. Ensuite, nous proposons une nouvelle approche d'évaluation sans référence de la qualité des images couleur en se basant sur les réseaux de neurones : compte tenu du caractère multidimensionnel de la qualité d'images, une quantification de la qualité a été proposée en se basant sur un ensemble d'attributs formant le descripteur PN (Précision, Naturalité). La précision traduit la netteté et la clarté. Quant à la naturalité, elle traduit la luminosité et la couleur. Pour modéliser le critère de la couleur, trois métriques sans référence ont été définies afin de détecter la couleur dominante dans l'image, la proportion de cette couleur et sa dispersion spatiale. Cette approche se base sur les réseaux de neurones afin d'imiter la perception du SVH. Deux variantes de cette approche ont été expérimentées (directe et progressive). Les résultats obtenus ont montré la performance de la variante progressive par rapport à la variante directe. L'application de l'approche proposée dans deux domaines : dans le contexte de la restauration, cette approche a servi comme un critère d'arrêt automatique pour les algorithmes de restauration. De plus, nous l'avons utilisé au sein d'un système d'estimation de la qualité d'images afin de détecter automatiquement le type de dégradation contenu dans une image. Dans le contexte de l'indexation et de la recherche d'images, l'approche proposée a servi d'introduire la qualité des images de la base comme index. Les résultats expérimentaux ont montré l'amélioration des performances du système de recherche d'images par le contenu en utilisant l'index qualité ou en réalisant un raffinement des résultats avec le critère de qualité
The research area in the objective quality assessment of the color images has been a renewed interest in recent years. The work is primarily driven by the advent of digital pictures and additional needs in image coding (compression, transmission, recovery, indexing,...). So far the best evaluation is visual (hence subjective) or by psychophysical techniques or by expert evaluation. Therefore, it is useful, even necessary, to establish criteria and objectives that automatically measures quality scores closest possible quality scores given by the subjective evaluation. We propose, firstly, a new full reference metric to assess the quality of color images, called overall Delta E, based on color appearance and incorporates the features of the human visual system (HVS). Performance was measured in two areas of application compression and restoration. The experiments carried out show a significant correlation between the results and subjective assessment.Then, we propose a new no reference quality assessmenent color images approach based on neural networks: given the multidimensional nature of image quality, a quantification of quality has been proposed, based on a set of attributes forming the descriptor UN (Utility, Naturalness). Accuracy reflects the sharpness and clarity. As for naturality, it reflects the brightness and color. To model the criterion of color, three no reference metrics were defined to detect the dominant color in the image, the proportion of that color and its spatial dispersion. This approach is based on neural networks to mimic the HVS perception. Two variants of this approach have been tried (direct and progressive). The results showed the performance of the progressive variant compared to the direct variant. The application of the proposed approach in two areas: in the context of restoration, this approach has served as a stopping criterion for automatic restoration algorithms. In addition, we have used in a system for estimating the quality of images to automatically detect the type of content in an image degradation. In the context of indexing and image retrieval, the proposed approach was used to introduce the quality of images in the database as an index. The experimental results showed the improvement of system performance image search by content by using the index or by making a quality refinement results with the quality criterion

碩士
國立臺北科技大學
電機工程系研究所
101
Image quality assessment is to measure the visual difference between two images. In order to make evaluation result in line with the visual quality perceived by the human, it is necessary to make use of the characteristic of the human visual system (HVS) into image quality assessment methods. Most of the literature works focus on how to more effectively compare two images of local information, but ignore the fact that they should have unequal sum-up weighting to the total, as the human vision has different sensitivity in regions with different contents and distortion. This work proposes an image assessment methods based on VIF. At the first, we extract the features of visual sensitivity in images with the Haar wavelet transform and log-Gabor filter, and detect the salient object region with the Laplacian filter, and calculate the distortion region with SSIM. For log-Gabor filter, we apply weighting to the log-Gabor frequency band based on the contrast sensitivity function (CSF). Since VIF assessment is based on the information comparison and not effective in luminance distortion, we take into account the luminance component of SSIM in the image information calculation, to compensate for the weakness of VIF in brightness distortion. The experimental results on image database show that our overall performance and efficiency outperforms the general image quality assessment methods.

With the rapid development of network visual communication technologies, digital video has become ubiquitous and indispensable in our everyday lives. Video acquisition, communication, and processing systems introduce various types of distortions, which may have major impact on perceived video quality by human observers. Effective and efficient objective video quality assessment (VQA) methods that can predict perceptual video quality are highly desirable in modern visual communication systems for performance evaluation, quality control and resource allocation purposes. Moreover, perceptual VQA measures may also be employed to optimize a wide variety of video processing algorithms and systems for best perceptual quality. This thesis exploits several novel ideas in the areas of video quality assessment and enhancement. Firstly, by considering a video signal as a 3D volume image, we propose a 3D structural similarity (SSIM) based full-reference (FR) VQA approach, which also incorporates local information content and local distortion-based pooling methods. Secondly, a reduced-reference (RR) VQA scheme is developed by tracing the evolvement of local phase structures over time in the complex wavelet domain. Furthermore, we propose a quality-aware video system which combines spatial and temporal quality measures with a robust video watermarking technique, such that RR-VQA can be performed without transmitting RR features via an ancillary lossless channel. Finally, a novel strategy for enhancing video denoising algorithms, namely poly-view fusion, is developed by examining a video sequence as a 3D volume image from multiple (front, side, top) views. This leads to significant and consistent gain in terms of both peak signal-to-noise ratio (PSNR) and SSIM performance, especially at high noise levels.

博士
國立雲林科技大學
設計學研究所
104
Three studies were conducted to explore the eye movement responses of subjects assessing perceived image quality. The main purposes of this study were (1) to explore the relationship between the physical attributes of images and the perceived image quality, (2) to explore and analyze eye movement response in a perceived image-quality assessment task, and (3) to determine a fixation map for a perceived image-quality assessment task. The methods and results of these three studies are described as follows. 　　In Study I, 35 images were used. Data on seven observers’ visual assessments of the contrast, saturation, and sharpness of images were collected for analysis. The results showed that, when the image contrast or saturation was increased, the image quality was perceived to be more visually pleasing. By contrast, when the sharpness of an image was increased or decreased, the image quality was perceived to be the least pleasing. The results of a canonical correlation analysis revealed that physical attributes explained 57.94% of the perceived image quality. 　　In Study II, 30 observers assessed 11 images, and their eye movement responses were analyzed to determine how the responses affected their perceptions of the image quality. The results showed that their subjective assessment of perceived image color quality exhibited a highly positive correlation with global perceived image quality. Additionally, the subjective assessment of perceived image color quality increased with the image color gamut. Eye movement data for the perceived image-color quality assessment correlated positively with the eye movement data for the perceived image-quality assessment. In observers who spent less time assessing the images, the information carried by their eye movement facilitated their predicting perceived image-color quality assessments more accurately and consistently compared with their predictions of perceived image-quality assessments. At a higher amplitude of saccades, the information carried by the eye movement assessment facilitated predictions of the perceived image quality. 　　In Study III, two visual assessment experiments were conducted. In total, 45 observers (30 in Experiment 1, 15 in Experiment 2) assessed 17 images (11 in Experiment 1, 6 in Experiment 2), and their eye movements were analyzed. The results showed that the CIEL*a*b color difference formula could facilitate easily understanding and computing the differences in a high number of fixation maps. The results from the two experiments were consistent, indicating that eye-tracking data are robust for predicting image quality. Observers tend to focus on either human faces or animal eyes. This study also shows that observers tend to ignore blue skies, grass, and foliage in landscape images. The central region attracted more attention than did the background or other objects in the images studied. All results have been applied to TTLA’s projects, which included developing and designing integrated image-processors for digital liquid crystal display televisions.

碩士
國立交通大學
電信工程系所
96
In this study, a joint spectro-temporal auditory model was utilized to assess speech quality objectively. In this model, the first stage is to mimic early cochlear functions of the spectrum estimation and the second stage is to mimic cortical functions of the multi-dimensional spectrum analysis. The goal of this study is to predict subjective mean opinion score (MOS). Objective speech quality assessment can be done by two methods：intrusive and non-intrusive. In this study, firstly, we observe and analyze patterns of the clean speech, the noisy speech with different background noise, and the degraded speech through different codecs at two auditory stages. Secondly, we will derive an objective estimate of the MOS from data-driven perceptual parameters which are believed to reflect people’s judgment on speech quality. Four perceptual parameters considered are intelligibility, naturalness, and pitch distortion. Finally, we use multiple regression analysis to combine the relationship between speech quality and these perceptual parameters, and then obtain our predicted MOS. We then demonstrate the MOS can be characterized quickly and reliably by these three perceptual features.

視覺信號，包括圖像，視頻等，在采集，壓縮，存儲，傳輸，重新生成的過程中都會被各種各樣的噪聲所影響，因此他們的主觀質量也就會降低。所以，主觀視覺質量在現今的視覺信號處理跟通訊系統中起到了很大的作用。這篇畢業論文主要討論質量評價的算法設計，以及這些衡量標準在視覺信號處理上的應用。這篇論文的工作主要包括以下五個方面。
第一部分主要集中在具有完全套考原始圖像的圖像質量評價。首先我們研究人類視覺系統的特征。具體說來，視覺在結構化失真上面的水平特性和顯著特征會被建模然后應用到結構相似度(SSIM)這個衡量標準上。實驗顯示我們的方法明顯的提高了衡量標準典主觀評價的相似度。由這個質量衡量標準的啟發，我們設計了一個主觀圖像壓縮的方法。其中我們提出了一個自適應的塊大小的超分辨率算法指導的下采樣的算法。實驗結果證明提出的圖像壓縮算法無論在主觀還是在客觀層面都構建了高質量的圖像。
第二個部分的工作主要討論具有完全參考原始視頻的視頻質量評價。考慮到人類視覺系統的特征，比如時空域的對此敏感函數，眼球的移動，紋理的遮掩特性，空間域的一致性，時間域的協調性，不同塊變換的特性，我們設計了一個自適應塊大小的失真閾值的模型。實驗證明，我們提出的失真閾值模型能夠更精確的描迷人類視覺系統的特性。基于這個自適應塊大小的失真閾值模型，我們設計了一個簡單的主觀質量評價標準。在公共的圓像以及視頻的主觀數據庫上的測試結果證明了這個簡單的評價標準的有效性。因此，我們把這個簡單的質量標準應用于視頻編碼系統中。它可以在同樣的碼率下提供更高主觀質量的視頻。
第三部分我們討論具有部分參考信息的圖像質量評價。我們通過描迷重組后的離散余弦變換域的系數的統計分布來衡量圖像的主觀質量。提出的評價標準發掘了相鄰的離散余弦系數的相同統計特性，相鄰的重組離散余弦系數的互信息，以及圖像的能量在不同頻率下的分布。實驗結果證明我們提出的質量標準河以超越其他的具有部分參考信息的質量評價標準，甚至還超過了具有完全參考信息的質量評價標準。而且，提取的特征很容易被編碼以及隱藏到圖像中以便于在圖像通訊中進行質量監控。
第四部分我們討論具有部分參考信息的視頻質量評價。我們提取的特征可以很好的描迷空間域的信息失，和時間域的相鄰兩幀間的直方圖的統計特性。在視頻主觀質量的數據庫上的實驗結果，也證明了提出的方法河以超越其他代表性的視頻質量評價標準，甚至是具有完全參考信息的質量評價標準， 譬如PSNR以及SSIM 。我們的方法只需要很少的特征來描迷每一幀視頻圖像。對于每一幀圖像，一個特征用于描迷空間域的特點，另外三個特征用于描述時間域的特點。考慮到計算的復雜度以及壓縮特征所需要的碼率，提出的方法河以很簡單的在視頻的傳輸過程中監控視頻的質量。
之前的四部分提到的主觀質量評價標準主要集中在傳統的失真上面， 譬如JPEG 圖像壓縮， H.264視頻壓縮。在最后一部分，我們討論在圖像跟視頻的retargeting過程中的失真。現如今，隨著消費者電子的發展，視覺信號需要在不同分辨率的顯示設備上進行通訊交互。因此， retargeting的算法把同一個原始圖像適應于不同的分辨率的顯示設備。這樣的過程就會引入圖像的失真。我們研究了對于retargeting圖像主觀質量的測試者的分數，從三個方面進行討論測試者對于retargeting圖像失真的反應.圖像retargeting的尺度，圖像retargeting的算法，原始圖像的內容特性。通過大量的主觀實驗測試，我們構建了一個關于圖像retargeting的主觀數據庫。基于這個主觀數據庫，我們評價以及分析了幾個具有代表性的質量評價標準。
Visual signals, including images, videos, etc., are affected by a wide variety of distortions during acquisition, compression, storage, processing, transmission, and reproduction processes, which result in perceptual quality degradation. As a result, perceptual quality assessment plays a very important role in today's visual signal processing and communication systems. In this thesis, quality assessment algorithms for evaluating the visual signal perceptual quality, as well as the applications on visual signal processing and communications, are investigated. The work consists of five parts as briefly summarized below.
The first part focuses on the full-reference (FR) image quality assessment. The properties of the human visual system (HVS) are firstly investigated. Specifically, the visual horizontal effect (HE) and saliency properties over the structural distortions are modelled and incorporated into the structure similarity index (SSIM). Experimental results show significantly improved performance in matching the subjective ratings. Inspired by the developed FR image metric, a perceptual image compression scheme is developed, where the adaptive block-based super-resolution directed down-sampling is proposed. Experimental results demonstrated that the proposed image compression scheme can produce higher quality images in terms of both objective and subjective qualities, compared with the existing methods.
The second part concerns the FR video quality assessment. The adaptive block-size transform (ABT) based just-noticeable difference (JND) for visual signals is investigated by considering the HVS characteristics, e.g., spatio-temporal contrast sensitivity function (CSF), eye movement, texture masking, spatial coherence, temporal consistency, properties of different block-size transforms, etc. It is verified that the developed ABT based JND can more accurately depict the HVS property, compared with the state-of-the-art JND models. The ABT based JND is thereby utilized to develop a simple perceptual quality metric for visual signals. Validations on the image and video subjective quality databases proved its effectiveness. As a result, the developed perceptual quality metric is employed for perceptual video coding, which can deliver video sequences of higher perceptual quality at the same bit-rates.
The third part discusses the reduced-reference (RR) image quality assessment, which is developed by statistically modelling the coe cient distribution in the reorganized discrete cosine transform (RDCT) domain. The proposed RR metric exploits the identical statistical nature of the adjacent DCT coefficients, the mutual information (MI) relationship between adjacent RDCT coefficients, and the image energy distribution among different frequency components. Experimental results demonstrate that the proposed metric outperforms the representative RR image quality metrics, and even the FR quality metric, i.e., peak signal to noise ratio (PSNR). Furthermore, the extracted RR features can be easily encoded and embedded into the distorted images for quality monitoring during image communications.
The fourth part investigates the RR video quality assessment. The RR features are extracted to exploit the spatial information loss and the temporal statistical characteristics of the inter-frame histogram. Evaluations on the video subjective quality databases demonstrate that the proposed method outperforms the representative RR video quality metrics, and even the FR metrics, such as PSNR, SSIM in matching the subjective ratings. Furthermore, only a small number of RR features is required to represent the original video sequence (each frame requires only 1 and 3 parameters to depict the spatial and temporal characteristics, respectively). By considering the computational complexity and the bit-rates for extracting and representing the RR features, the proposed RR quality metric can be utilized for quality monitoring during video transmissions, where the RR features for perceptual quality analysis can be easily embedded into the videos or transmitted through an ancillary data channel.
The aforementioned perceptual quality metrics focus on the traditional distortions, such as JPEG image compression noise, H.264 video compression noise, and so on. In the last part, we investigate the distortions introduced during the image and video retargeting process. Nowadays, with the development of the consumer electronics, more and more visual signals have to communicate between different display devices of different resolutions. The retargeting algorithm is employed to adapt a source image of one resolution to be displayed in a device of a different resolution, which may introduce distortions during the retargeting process. We investigate the subjective responses on the perceptual qualities of the retargeted images, and discuss the subjective results from three perspectives, i.e., retargeting scales, retargeting methods, and source image content attributes. An image retargeting subjective quality database is built by performing a large-scale subjective study of image retargeting quality on a collection of retargeted images. Based on the built database, several representative quality metrics for retargeted images are evaluated and discussed.
Detailed summary in vernacular field only.
Detailed summary in vernacular field only.
Detailed summary in vernacular field only.
Detailed summary in vernacular field only.
Detailed summary in vernacular field only.
Detailed summary in vernacular field only.
Ma, Lin.
"December 2012."
Thesis (Ph.D.)--Chinese University of Hong Kong, 2013.
Includes bibliographical references (leaves 185-197).
Abstract also in Chinese.
Dedication --- p.ii
Acknowledgments --- p.iii
Abstract --- p.viii
Publications --- p.xi
Nomenclature --- p.xvii
Contents --- p.xxiv
List of Figures --- p.xxviii
List of Tables --- p.xxx
Chapter 1 --- Introduction --- p.1
Chapter 1.1 --- Motivation and Objectives --- p.1
Chapter 1.2 --- Subjective Perceptual Quality Assessment --- p.5
Chapter 1.3 --- Objective Perceptual Quality Assessment --- p.10
Chapter 1.3.1 --- Visual Modelling Approach --- p.10
Chapter 1.3.2 --- Engineering Modelling Approach --- p.15
Chapter 1.3.3 --- Perceptual Subjective Quality Databases --- p.19
Chapter 1.3.4 --- Performance Evaluation --- p.21
Chapter 1.4 --- Thesis Contributions --- p.22
Chapter 1.5 --- Organization of the Thesis --- p.24
Chapter I --- Full Reference Quality Assessment --- p.26
Chapter 2 --- Full Reference Image Quality Assessment --- p.27
Chapter 2.1 --- Visual Horizontal Effect for Image Quality Assessment --- p.27
Chapter 2.1.1 --- Introduction --- p.27
Chapter 2.1.2 --- Proposed Image Quality Assessment Framework --- p.28
Chapter 2.1.3 --- Experimental Results --- p.34
Chapter 2.1.4 --- Conclusion --- p.36
Chapter 2.2 --- Image Compression via Adaptive Block-Based Super-Resolution Directed Down-Sampling --- p.37
Chapter 2.2.1 --- Introduction --- p.37
Chapter 2.2.2 --- The Proposed Image Compression Framework --- p.38
Chapter 2.2.3 --- Experimental Results --- p.42
Chapter 2.2.4 --- Conclusion --- p.45
Chapter 3 --- Full Reference Video Quality Assessment --- p.46
Chapter 3.1 --- Adaptive Block-size Transform based Just-Noticeable Dfference Model for Visual Signals --- p.46
Chapter 3.1.1 --- Introduction --- p.46
Chapter 3.1.2 --- JND Model based on Transforms of Different Block Sizes --- p.48
Chapter 3.1.3 --- Selection Strategy Between Transforms of Different Block Sizes --- p.53
Chapter 3.1.4 --- JND Model Evaluation --- p.56
Chapter 3.1.5 --- Conclusion --- p.60
Chapter 3.2 --- Perceptual Quality Assessment --- p.60
Chapter 3.2.1 --- Experimental Results --- p.62
Chapter 3.2.2 --- Conclusion --- p.64
Chapter 3.3 --- Motion Trajectory Based Visual Saliency for Video Quality Assessment --- p.65
Chapter 3.3.1 --- Motion Trajectory based Visual Saliency for VQA --- p.66
Chapter 3.3.2 --- New Quaternion Representation (QR) for Each frame --- p.66
Chapter 3.3.3 --- Saliency Map Construction by QR --- p.67
Chapter 3.3.4 --- Incorporating Visual Saliency with VQAs --- p.68
Chapter 3.3.5 --- Experimental Results --- p.69
Chapter 3.3.6 --- Conclusion --- p.72
Chapter 3.4 --- Perceptual Video Coding --- p.72
Chapter 3.4.1 --- Experimental Results --- p.75
Chapter 3.4.2 --- Conclusion --- p.76
Chapter II --- Reduced Reference Quality Assessment --- p.77
Chapter 4 --- Reduced Reference Image Quality Assessment --- p.78
Chapter 4.1 --- Introduction --- p.78
Chapter 4.2 --- Reorganization Strategy of DCT Coefficients --- p.81
Chapter 4.3 --- Relationship Analysis of Intra and Inter RDCT subbands --- p.83
Chapter 4.4 --- Reduced Reference Feature Extraction in Sender Side --- p.88
Chapter 4.4.1 --- Intra RDCT Subband Modeling --- p.89
Chapter 4.4.2 --- Inter RDCT Subband Modeling --- p.91
Chapter 4.4.3 --- Image Frequency Feature --- p.92
Chapter 4.5 --- Perceptual Quality Analysis in the Receiver Side --- p.95
Chapter 4.5.1 --- Intra RDCT Feature Difference Analysis --- p.95
Chapter 4.5.2 --- Inter RDCT Feature Difference Analysis --- p.96
Chapter 4.5.3 --- Image Frequency Feature Difference Analysis --- p.96
Chapter 4.6 --- Experimental Results --- p.98
Chapter 4.6.1 --- Efficiency of the DCT Reorganization Strategy --- p.98
Chapter 4.6.2 --- Performance of the Proposed RR IQA --- p.99
Chapter 4.6.3 --- Performance of the Proposed RR IQA over Each Individual Distortion Type --- p.105
Chapter 4.6.4 --- Statistical Significance --- p.107
Chapter 4.6.5 --- Performance Analysis of Each Component --- p.109
Chapter 4.7 --- Conclusion --- p.111
Chapter 5 --- Reduced Reference Video Quality Assessment --- p.113
Chapter 5.1 --- Introduction --- p.113
Chapter 5.2 --- Proposed Reduced Reference Video Quality Metric --- p.114
Chapter 5.2.1 --- Reduced Reference Feature Extraction from Spatial Perspective --- p.116
Chapter 5.2.2 --- Reduced Reference Feature Extraction from Temporal Perspective --- p.118
Chapter 5.2.3 --- Visual Quality Analysis in Receiver Side --- p.121
Chapter 5.3 --- Experimental Results --- p.123
Chapter 5.3.1 --- Consistency Test of the Proposed RR VQA over Compressed Video Sequences --- p.124
Chapter 5.3.2 --- Consistency Test of the Proposed RR VQA over Video Sequences with Simulated Distortions --- p.126
Chapter 5.3.3 --- Performance Evaluation of the Proposed RR VQA on Compressed Video Sequences --- p.129
Chapter 5.3.4 --- Performance Evaluation of the Proposed RR VQA on Video Sequences Containing Transmission Distortions --- p.133
Chapter 5.3.5 --- Performance Analysis of Each Component --- p.135
Chapter 5.4 --- Conclusion --- p.137
Chapter III --- Retargeted Visual Signal Quality Assessment --- p.138
Chapter 6 --- Image Retargeting Perceptual Quality Assessment --- p.139
Chapter 6.1 --- Introduction --- p.139
Chapter 6.2 --- Preparation of Database Building --- p.142
Chapter 6.2.1 --- Source Image --- p.142
Chapter 6.2.2 --- Retargeting Methods --- p.143
Chapter 6.2.3 --- Subjective Testing --- p.146
Chapter 6.3 --- Data Processing and Analysis for the Database --- p.150
Chapter 6.3.1 --- Processing of Subjective Ratings --- p.150
Chapter 6.3.2 --- Analysis and Discussion of the Subjective Ratings --- p.153
Chapter 6.4 --- Objective Quality Metric for Retargeted Images --- p.162
Chapter 6.4.1 --- Quality Metric Performances on the Constructed Image Retargeting Database --- p.162
Chapter 6.4.2 --- Subjective Analysis of the Shape Distortion and Content Information Loss --- p.165
Chapter 6.4.3 --- Discussion --- p.167
Chapter 6.5 --- Conclusion --- p.169
Chapter 7 --- Conclusions --- p.170
Chapter 7.1 --- Conclusion --- p.170
Chapter 7.2 --- Future Work --- p.173
Chapter A --- Attributes of the Source Image --- p.176
Chapter B --- Retargeted Image Name and the Corresponding Number --- p.179
Chapter C --- Source Image Name and the Corresponding Number --- p.183
Bibliography --- p.185

碩士
國立臺北科技大學
電機工程研究所
105
Image Quality Assessment (IQA) aims to predict the quality of distorted images when viewed by human observers. Besides quantifying the distortions of an image, characteristics of Human Visual System (HVS) should also be taken into account so that predictions can be in line with human judgements. Existing IQA metrics mostly focus on designing an accurate local quality estimation; whereas the global quality score for a distorted image is often obtained by averaging local quality scores, overlooking the individual importance of local regions. Moreover, most IQA metrics can only assess grayscale images, hence chrominance data is usually omitted by assessment processes. This work aims to improve the performance of VIF by incorporating chrominance channels into the algorithm. Perceptual color difference is measured by a novel superpixel-based CIEDE2000 extension. Color difference and visual saliency are used as pooling weights to determine the different importance of each local quality estimation of an image. Extensive experiments performed on six benchmark IQA databases show that the proposed NEWSS-VIF achieves promising results for estimating perceptual quality of color images.

The capability of hand-held devices to acquire high-definition visual content has led to a tremendous increase in the number of images and videos captured daily. However, camera hardware and pipelines are not perfect and lead to multiple distortions in the captured content. This makes quality assessment (QA) imperative to advance the qualitative capability of different devices and the pipelines used. More particularly, the aim of perceptual quality assessment is to quantitatively analyze the perceptual quality of the captured content with respect to the distortions observed by the human visual system. This thesis focuses on two aspects of perceptual quality assessment. Firstly, we focus on the subjective and objective quality assessment of low-light restored images. Then we consider the problem of unsupervised quality assessment methods for authentically distorted images. The quality assessment of restored low-light images is an important tool for benchmarking and improving low-light restoration (LLR) algorithms. While several LLR algorithms exist, the subjective perception of the restored images has been much less studied. Challenges in capturing aligned low-light and well-lit image pairs and collecting a large number of human opinion scores of quality for training warrant the design of unsupervised (or opinion unaware) no-reference (NR) QA methods. In this part, we study the subjective perception of low-light restored images and their unsupervised NR QA. Our contributions are two-fold. We first create a dataset of restored low-light images using various LLR methods, conduct a subjective QA study, and benchmark the performance of existing QA methods. The lack of good perceptual quality metrics designed explicitly for the low-light scenario is an important limitation in advancing the design of restoration methods. To tackle this, we present a self-supervised contrastive learning technique to extract distortion-aware features from the restored low-light images. We show that these features can be effectively used to build an opinion unaware image quality analyzer. Detailed experiments reveal that our unsupervised NR QA model achieves state-of-the-art performance among all such quality measures for low-light restored images. The quality assessment of camera captured authentically distorted images is challenging due to the lack of a reference. While there is a plethora of supervised no reference image QA algorithms, there is a need to study unsupervised or opinion unaware algorithms based on their superior generalization performance. We explore self-supervised learning (SSL) for the feature design on authentically distorted images to predict quality without training on human labels. While SSL on synthetic distortions has recently shown promise, there is a need to enrich the feature learning on authentic distortions. We propose a novel two-stage learning approach on synthetic and authentically distorted images with different learning methodologies. We perform contrastive learning with positives and negatives that vary with quality on synthetic data to capture quality features. While learning on authentically distorted images, we only consider positives due to the difficulty in obtaining negatives that vary in quality alone. We employ the SimSiam framework to enrich features by fine-tuning on authentically distorted images. We show that the self-supervised features we learn can be used to make perceptually consistent image quality predictions on authentically distorted images without training on any human opinion scores. We achieve state-of-the-art performance on multiple authentically distorted datasets without training on them.

碩士
國立暨南國際大學
電機工程學系
102
In recently years, three-dimensional (3D) video is very popular. Most existing 3D video quality metrics uses traditional 2D image quality assessment (IQA) to judge the quality of 3D video. However, 2D IQA does not sufficient render human judgment for 3D video, since 3D image have some different characteristics with 2D video. Therefore, this thesis proposes a new 3D video quality assessment based on 3D visual perceptual for texture and depth image to measure the quality of 3D videos. As observe in this research, the 3D distortions can be classified to depth distortions and texture distortions. In the aspect of 3D videos, depth information is the main factor of effecting 3D watching experience, if depth information is distorted will make 3D watching discomfort. The texture distortions include ghost effect; contour artifacts; compressed texture quality decadence. In addition, the distortions of texture and depth image are also important like compressed distortion, after compressed the quality of texture and depth map will be lowed. In this thesis, the proposed assessment metric uses the edge information to evaluate the ghost effect and contour artifacts. For compressed texture quality decadence, this thesis uses 2D quality metric to evaluate the quality and consider the effect of object shift. Since the characteristics of depth map are different with color image. Therefore, this thesis chooses 2D quality metric IWSSIM to evaluate the quality in compressed depth map. The experimental results show that the proposed metric compare to 2D Quality metrics and other reference metrics is more suitable to the human perception, as demonstrated by Pearson correlation coefficient (PLCC) obtained from subjective score and objective score. The proposed 3D video quality assessment is based 3D visual perceptual. It makes the quality assessment result more approach the 3D experience of viewer.

abstract: The quality of real-world visual content is typically impaired by many factors including image noise and blur. Detecting and analyzing these impairments are important steps for multiple computer vision tasks. This work focuses on perceptual-based locally adaptive noise and blur detection and their application to image restoration. In the context of noise detection, this work proposes perceptual-based full-reference and no-reference objective image quality metrics by integrating perceptually weighted local noise into a probability summation model. Results are reported on both the LIVE and TID2008 databases. The proposed metrics achieve consistently a good performance across noise types and across databases as compared to many of the best very recent quality metrics. The proposed metrics are able to predict with high accuracy the relative amount of perceived noise in images of different content. In the context of blur detection, existing approaches are either computationally costly or cannot perform reliably when dealing with the spatially-varying nature of the defocus blur. In addition, many existing approaches do not take human perception into account. This work proposes a blur detection algorithm that is capable of detecting and quantifying the level of spatially-varying blur by integrating directional edge spread calculation, probability of blur detection and local probability summation. The proposed method generates a blur map indicating the relative amount of perceived local blurriness. In order to detect the flat/near flat regions that do not contribute to perceivable blur, a perceptual model based on the Just Noticeable Difference (JND) is further integrated in the proposed blur detection algorithm to generate perceptually significant blur maps. We compare our proposed method with six other state-of-the-art blur detection methods. Experimental results show that the proposed method performs the best both visually and quantitatively. This work further investigates the application of the proposed blur detection methods to image deblurring. Two selective perceptual-based image deblurring frameworks are proposed, to improve the image deblurring results and to reduce the restoration artifacts. In addition, an edge-enhanced super resolution algorithm is proposed, and is shown to achieve better reconstructed results for the edge regions.
Dissertation/Thesis
Doctoral Dissertation Electrical Engineering 2016

Yes
This work presents a methodology to optimize the selection of multiple parameter levels of an image acquisition, degradation, or post-processing process applied to stimuli intended to be used in a subjective image or video quality assessment (QA) study. It is known that processing parameters (e.g. compression bit-rate) or techni- cal quality measures (e.g. peak signal-to-noise ratio, PSNR) are often non-linearly related to human quality judgment, and the model of either relationship may not be known in advance. Using these approaches to select parameter levels may lead to an inaccurate estimate of the relationship between the parameter and subjective quality judgments – the system’s quality model. To overcome this, we propose a method for modeling the rela- tionship between parameter levels and perceived quality distances using a paired comparison parameter selection procedure in which subjects judge the perceived similarity in quality. Our goal is to enable the selection of evenly sampled parameter levels within the considered quality range for use in a subjective QA study. This approach is tested on two applications: (1) selection of compression levels for laparoscopic surgery video QA study, and (2) selection of dose levels for an interventional X-ray QA study. Subjective scores, obtained from the follow-up single stimulus QA experiments conducted with expert subjects who evaluated the selected bit-rates and dose levels, were roughly equidistant in the perceptual quality space - as intended. These results suggest that a similarity judgment task can help select parameter values corresponding to desired subjective quality levels.
Parts of this work were performed within the Telesurgery project (co-funded by iMinds, a digital research institute founded by the Flemish Government; project partners are Unilabs Teleradiology, SDNsquare and Barco, with project support from IWT) and the PANORAMA project (co-funded by grants from Belgium, Italy, France, the Netherlands, the United Kingdom, and the ENIAC Joint Undertaking).

Tone mapping operators aim to compress high dynamic range (HDR) images to low dynamic range ones so as to visualize HDR images on standard displays. Most existing works were demonstrated on specific examples without being thoroughly tested on well-established and subject-validated image quality assessment models. A recent tone mapped image quality index (TMQI) made the first attempt on objective quality assessment of tone mapped images. TMQI consists of two fundamental building blocks: structural fidelity and statistical naturalness. In this thesis, we propose an enhanced tone mapped image quality index (eTMQI) by 1) constructing an improved nonlinear mapping function to better account for the local contrast visibility of HDR images and 2) developing an image dependent statistical naturalness model to quantify the unnaturalness of tone mapped images based on a subjective study. Experiments show that the modified structural fidelity and statistical naturalness terms in eTMQI better correlate with subjective quality evaluations. Furthermore, we propose an iterative optimization algorithm for tone mapping. The advantages of this algorithm are twofold: 1) eTMQI and TMQI can be compared in a more straightforward way; 2) better quality tone mapped images can be automatically generated by using eTMQI as the optimization goal. Numerical and subjective experiments demonstrate that eTMQI is a superior objective quality assessment metric for tone mapped images and consistently outperforms TMQI.

Objective Image and Video Quality Assessment (I/VQA) measures predict image/video quality as perceived by human beings - the ultimate consumers of visual data. Existing research in the area is mainly limited to benchmarking and monitoring of visual data. The use of I/VQA measures in the design and optimization of image/video processing algorithms and systems is more desirable, challenging and fruitful but has not been well explored. Among the recently proposed objective I/VQA approaches, the structural similarity (SSIM) index and its variants have emerged as promising measures that show superior performance as compared to the widely used mean squared error (MSE) and are computationally simple compared with other state-of-the-art perceptual quality measures. In addition, SSIM has a number of desirable mathematical properties for optimization tasks. The goal of this research is to break the tradition of using MSE as the optimization criterion for image and video processing algorithms. We tackle several important problems in visual communication applications by exploiting SSIM-inspired design and optimization to achieve significantly better performance. Firstly, the original SSIM is a Full-Reference IQA (FR-IQA) measure that requires access to the original reference image, making it impractical in many visual communication applications. We propose a general purpose Reduced-Reference IQA (RR-IQA) method that can estimate SSIM with high accuracy with the help of a small number of RR features extracted from the original image. Furthermore, we introduce and demonstrate the novel idea of partially repairing an image using RR features. Secondly, image processing algorithms such as image de-noising and image super-resolution are required at various stages of visual communication systems, starting from image acquisition to image display at the receiver. We incorporate SSIM into the framework of sparse signal representation and non-local means methods and demonstrate improved performance in image de-noising and super-resolution. Thirdly, we incorporate SSIM into the framework of perceptual video compression. We propose an SSIM-based rate-distortion optimization scheme and an SSIM-inspired divisive optimization method that transforms the DCT domain frame residuals to a perceptually uniform space. Both approaches demonstrate the potential to largely improve the rate-distortion performance of state-of-the-art video codecs. Finally, in real-world visual communications, it is a common experience that end-users receive video with significantly time-varying quality due to the variations in video content/complexity, codec configuration, and network conditions. How human visual quality of experience (QoE) changes with such time-varying video quality is not yet well-understood. We propose a quality adaptation model that is asymmetrically tuned to increasing and decreasing quality. The model improves upon the direct SSIM approach in predicting subjective perceptual experience of time-varying video quality.

Since its introduction in 2004, the Structural Similarity (SSIM) index has gained widespread popularity as an image quality assessment measure. SSIM is currently recognized to be one of the most powerful methods of assessing the visual closeness of images. That being said, the Mean Squared Error (MSE), which performs very poorly from a perceptual point of view, still remains the most common optimization criterion in image processing applications because of its relative simplicity along with a number of other properties that are deemed important. In this thesis, some necessary tools to assist in the design of SSIM-optimal algorithms are developed. This work combines theoretical developments with experimental research and practical algorithms. The description of the mathematical properties of the SSIM index represents the principal theoretical achievement in this thesis. Indeed, it is demonstrated how the SSIM index can be transformed into a distance metric. Local convexity, quasi-convexity, symmetries and invariance properties are also proved. The study of the SSIM index is also generalized to a family of metrics called normalized (or M-relative) metrics. Various analytical techniques for different kinds of SSIM-based optimization are then devised. For example, the best approximation according to the SSIM is described for orthogonal and redundant basis sets. SSIM-geodesic paths with arclength parameterization are also traced between images. Finally, formulas for SSIM-optimal point estimators are obtained. On the experimental side of the research, the structural self-similarity of images is studied. This leads to the confirmation of the hypothesis that the main source of self-similarity of images lies in their regions of low variance. On the practical side, an implementation of local statistical tests on the image residual is proposed for the assessment of denoised images. Also, heuristic estimations of the SSIM index and the MSE are developed. The research performed in this thesis should lead to the development of state-of-the-art image denoising algorithms. A better comprehension of the mathematical properties of the SSIM index represents another step toward the replacement of the MSE with SSIM in image processing applications.

The research domain on the Quality of Experience (QoE) of 2D video streaming has been well established. However, a new video format is emerging and gaining popularity and availability: VR 360-degree video. The processing and transmission of 360-degree videos brings along new challenges such as large bandwidth requirements and the occurrence of different distortions. The viewing experience is also substantially different from 2D video, it offers more interactive freedom on the viewing angle but can also be more demanding and cause cybersickness. Further research on the QoE of 360-videos specifically is thus required.The first goal of this thesis is to complement earlier research by (Tran, Ngoc, Pham, Jung, and Thank, 2017) testing the effects of quality degradation, freezing, and content on the QoE of 360-videos. The second goal is to test the contribution of visual attention as influence factor in the QoE assessment. Data will be gathered through subjective tests where participants watch degraded versions of 360-videos through an HMD with integrated eye-tracking sensors. After each video they will answer questions regarding their quality perception, experience, perceptual load, and cybersickness.Results of the first part show overall rather low QoE ratings and it decreases even more as quality is degraded and freezing events are added. Cyber sickness was found not to be an issue. The effects of the manipulation on visual attention were minimal. Attention was mainly directed by content, but also by surprising elements. The addition of eye-tracking metrics did not further explain individual differences in subjective ratings. Nevertheless it was found that looking at moving objects increased the negative effect of freezing events and made participants less sensitive for quality distortions. The results of this thesis alone are not enough to successfully regard visual attention as an influence factor in 360-video.

Looking at a link between blur and visual discomfort, in the present thesis, blur is viewed as a cause of a cognitive loss, and the discomfort as the immediate consequence of this loss. Among the basic cognitive functions of the Human Visual System (HVS), detection, recognition, and coarse localization functions are strongly conditioned by the individual experience. Conversely, it seems plausible that the fine localization function is committed to stabler and inter-subjective functions of the HVS. After a preliminary discussion of the operators and the ML model used (Part II), the approach presented in Part III of this thesis starts from postulating that, in the absence of vision problems, the HVS performs the fine localization of the observed objects with the best accuracy allowed by its physical macro-structure. This is a fundamental assumption, because it is known from the estimation theory that the maximum accuracy attainable when measuring the fine position of patterns in background noise is obtained by the Fisher Information about positional parameters. In fact, the Fisher Information inverse yields the minimum estimation variance. The proposed approach is based on an abstract, functional model of the Receptive Fields (RF) of the HVS, referred to as Virtual Receptive Field (VRF) and it is tuned to statistical features of natural scenes. It is a complex-valued operator, orientation-selective both in the space domain and in the spatial frequency domain. The role of the VRF model is to extract the Positional Fisher Information (PFI) as a measure of the pattern localizability loss. In the Image Quality Assessment (IQA) Full Reference (FR) environment, subjective assessments refer to the retinal image and lead to the MOS/DMOS values (Difference of Mean Opinion Score). The quality calculated by the IQA metrics is objective and refers to the image reproduced on the display. A parametric scoring function maps these metrics onto the MOS/DMOS values and depends critically on the Viewing Distance (VD) of the subject from the monitor in which the image is reproduced. When objective quality estimates for different VDs are required, as in the case of auditoria, cinemas, classrooms, a re-training procedure must be repeated for each different VDs. In the final part of this thesis (Part IV), the problem of VD is dealt with from a theoretical point of view and a model of the scoring function is defined for the case of blurred images where image degradation substantially depends on the VD. Starting from a Fisher Information loss model applied to the Gaussian distortion case in natural images, we see that the VD is estimated from the data themselves. Several maps are given with the aim of obtaining a DMOS prediction at different distances starting from the data available for a specific distance, without performing new experiments. Moreover, the theoretical results are verified on some most popular IQA FR methods and the problem of VD correction is generalized to the other distortions. Finally, the impact of isolated, long, strong, unidirectional edges on early vision is shown. As for the VD correction, an a-priori linear estimator is presented. It does not require rectification through a re-training procedure. Useful maps for detecting the position and the intensity of the PFI losses in an image are given, and the isoluminance colors allow to highlight strong and isolated edges, maintaining a constant intensity at the same edge level. We have an easy visual feedback on the images themselves to see where the greatest loss of information and the greatest discomfort due to blur are.