To see the other types of publications on this topic, follow the link: Image collection.
Journal articles on the topic 'Image collection'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 journal articles for your research on the topic 'Image collection.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse journal articles on a wide variety of disciplines and organise your bibliography correctly.
1
Riahi Samani, Zahra, and Mohsen Ebrahimi Moghaddam. "Image Collection Summarization Method Based on Semantic Hierarchies." AI 1, no. 2 (May 18, 2020): 209–28. http://dx.doi.org/10.3390/ai1020014.
Full textAbstract:
The size of internet image collections is increasing drastically. As a result, new techniques are required to facilitate users in browsing, navigation, and summarization of these large volume collections. Image collection summarization methods present users with a set of exemplar images as the most representative ones from the initial image collection. In this study, an image collection summarization technique was introduced according to semantic hierarchies among them. In the proposed approach, images were mapped to the nodes of a pre-defined domain ontology. In this way, a semantic hierarchical classifier was used, which finally mapped images to different nodes of the ontology. We made a compromise between the degree of freedom of the classifier and the goodness of the summarization method. The summarization was done using a group of high-level features that provided a semantic measurement of information in images. Experimental outcomes indicated that the introduced image collection summarization method outperformed the recent techniques for the summarization of image collections.
2
Guo, Feng Ying. "The Hardware Design and Implement of an Image Information Collection System." Advanced Materials Research 804 (September 2013): 211–15. http://dx.doi.org/10.4028/www.scientific.net/amr.804.211.
Full textAbstract:
It introduces the hardware design and implement of the image information collection system which is one part of Fluorescent magnetic particle nondestructive detection system of Cannonball. The image information collection system includes Open collecting card model Initialize collection card modelApply for EMS memory modelCollecting image modelWrite bitmap information head model Show pictures modeClose collection card model. Through collecting image information to the EMS memory and then processing the image information, the magnetic images on the cannonball are showed clearly on computer, and cracks on the cannonball can easily to be judged. The software of the image information collection system is developed by Visual C++.
3
Kim, Namhoon, Eunha Chun, and Eunju Ko. "Country of origin effects on brand image, brand evaluation, and purchase intention." International Marketing Review 34, no. 2 (April 10, 2017): 254–71. http://dx.doi.org/10.1108/imr-03-2015-0071.
Full textAbstract:
Purpose The purpose of this paper is to analyze how national stereotype, country of origin (COO), and fashion brand’s images influence consumers’ brand evaluations and purchase intentions regarding fashion collections. Korea (Seoul) and overseas (New York and Paris) collections are compared and analyzed. Design/methodology/approach The authors conduct a structural equation modeling and multi-group analysis using data collected from Seoul, New York, and Paris. Findings Consumers make higher brand evaluations and ultimately have stronger purchase intentions toward fashion collections from countries that have stronger COO and fashion brand images. In the context of fashion collections, COO image is greatly influenced by a nation’s political economic and cultural artistic images. In addition, comparing the domestic Seoul fashion collection with New York and Paris collections reveals that a national stereotype images, COO images of fashion collection, and fashion brand’s images cause different brand evaluation and purchase intention. Originality/value The overarching value of the study is that it expands COO research, which has been limited to actual products. Also, the results provide a basic foundation for establishing marketing strategy based on COO image as a way to enhance the development and image of fashion collection.
4
Blatter, Jeremy. "Wellcome Library Moving Image and Sound Collection, (http://wellcomelibrary.org/about-us/about-the-collections/moving-image-and-sound-collection/)." Medical History 57, no. 3 (May 30, 2013): 459–60. http://dx.doi.org/10.1017/mdh.2013.29.
Full text
5
Caspers, Max. "Image Recognition to Enhance the Value of Collections." Biodiversity Information Science and Standards 2 (June 13, 2018): e26320. http://dx.doi.org/10.3897/biss.2.26320.
Full textAbstract:
Techniques for image recognition through machine learning have advanced rapidly over recent years and applications using this technique are becoming increasingly common.. Applications using image recognition have enormous potential not only for research, education, conservation and capacity-building but certainly also for collections management. Perhaps by now an even bigger challenge than the technological one is supplying content in the form of large amounts of validated images. With an estimated 44 million objects, the collection of Naturalis Biodiversity Center has plenty of physical source material. During a five-year digitization programme (2010–2015) at Naturalis, 4.4 million herbarium sheets were imaged and since the start of the “Butterflies in Bags” project, 50,000 papered butterflies (out of more than 500,000) have been digitized and photographed by volunteers in a standardized manner. Still there are large parts of our collection that are not digitized at specimen level, let alone imaged, but hold great potential for collections work. This poster presents a workflow for efficient scanning of insect drawers and automated segmentation of those images to “feed” deep learning-based image recognition with images of individual insects. It will also demonstrate how this will aid in enhancing the value of our collections. With proper expert validation early on in the process, the software could mature and become more independent in such a way that ultimately, it could be used by non-specialist professionals to identify the majority of common species. The technique would pinpoint anomalies based on self-learned patterns, both in unidentified and in already identified specimens, and link those back to the taxonomic specialist. Not only does image recognition aid taxonomy, it may also hold potential for conservation and management by, for example, detecting damaged specimens or managing space utilization of drawers.
6
Martin, William. "Satellite image collection optimization." Optical Engineering 41, no. 9 (September 1, 2002): 2083. http://dx.doi.org/10.1117/1.1495856.
Full text
7
Saba, Khalaf. "Pseudo-Subdural Collection on Ultrasound: A Mirror Image Artefact." Journal of Clinical Research and Reports 11, no. 4 (October 5, 2022): 01–03. http://dx.doi.org/10.31579/2690-1919/270.
Full text
8
Turner, J. N., D. H. Szarowski, B. Roysam, and T. J. Holmes. "Light Microscopy Image Collection: Confocal, Widefield and Deconvolution." Microscopy and Microanalysis 3, S2 (August 1997): 371–72. http://dx.doi.org/10.1017/s1431927600008746.
Full textAbstract:
Light microscopy instrumentation and biological applications continue to expand rapidly, and an important aspect of this expansion is three-dimensional (3-D) imaging and image analysis. For many biological specimens, optical sectioning, i.e. collecting images at a sequence of depths under controlled conditions, provides true 3-D data through the entire specimen. This is especially important for specimens whose thickness exceeds the depth-of-field of the microscope objective lens. This sequence of optical sections is the basis for 3-D image reconstruction providing information from all three specimen dimensions instead of just the traditional two in the image plane of the microscope. The analysis of images in 3-D provides insights into the structure and function of biological specimens that are not available through other means. However, 3-D microscopy also presents additional choices for image collection. The first of which is whether to use widefield or confocal microscopy and the second is whether to utilize digital deblurring or deconvolution methods.
9
Legland, David, and Marie-Françoise Devaux. "ImageM: a user-friendly interface for the processing of multi-dimensional images with Matlab." F1000Research 10 (April 30, 2021): 333. http://dx.doi.org/10.12688/f1000research.51732.1.
Full textAbstract:
Modern imaging devices provide a wealth of data often organized as images with many dimensions, such as 2D/3D, time and channel. Matlab is an efficient software solution for image processing, but it lacks many features facilitating the interactive interpretation of image data, such as a user-friendly image visualization, or the management of image meta-data (e.g. spatial calibration), thus limiting its application to bio-image analysis. The ImageM application proposes an integrated user interface that facilitates the processing and the analysis of multi-dimensional images within the Matlab environment. It provides a user-friendly visualization of multi-dimensional images, a collection of image processing algorithms and methods for analysis of images, the management of spatial calibration, and facilities for the analysis of multi-variate images. ImageM can also be run on the open source alternative software to Matlab, Octave. ImageM is freely distributed on GitHub: https://github.com/mattools/ImageM.
10
Pangallo, Matteo. "Shakespeare Quartos Archive. Image Collection." Renaissance and Reformation 42, no. 3 (December 11, 2019): 170–73. http://dx.doi.org/10.7202/1066364ar.
Full text
11
Qu, Jingye, and Jiangping Chen. "An investigation of benchmark image collections: how different from digital libraries?" Electronic Library 37, no. 3 (June 3, 2019): 401–18. http://dx.doi.org/10.1108/el-10-2018-0195.
Full textAbstract:
Purpose This paper aims to introduce the construction methods, image organization, collection use and access of benchmark image collections to the digital library (DL) community. It aims to connect two distinct communities: the DL community and image processing researchers so that future image collections could be better constructed, organized and managed for both human and computer use. Design/methodology/approach Image collections are first identified through an extensive literature review of published journal articles and a web search. Then, a coding scheme focusing on image collections’ creation, organization, access and use is developed. Next, three major benchmark image collections are analysed based on the proposed coding scheme. Finally, the characteristics of benchmark image collections are summarized and compared to DLs. Findings Although most of the image collections in DLs are carefully curated and organized using various metadata schema based on an image’s external features to facilitate human use, the benchmark image collections created for promoting image processing algorithms are annotated on an image’s content to the pixel level, which makes each image collection a more fine-grained, organized database appropriate for developing automatic techniques on classification summarization, visualization and content-based retrieval. Research limitations/implications This paper overviews image collections by their application fields. The three most representative natural image collections in general areas are analysed in detail based on a homemade coding scheme, which could be further extended. Also, domain-specific image collections, such as medical image collections or collections for scientific purposes, are not covered. Practical implications This paper helps DLs with image collections to understand how benchmark image collections used by current image processing research are created, organized and managed. It informs multiple parties pertinent to image collections to collaborate on building, sustaining, enriching and providing access to image collections. Originality/value This paper is the first attempt to review and summarize benchmark image collections for DL managers and developers. The collection creation process and image organization used in these benchmark image collections open a new perspective to digital librarians for their future DL collection development.
12
Gu, Yi, Chaoli Wang, Jun Ma, Robert J. Nemiroff, David L. Kao, and Denis Parra. "Visualization and recommendation of large image collections toward effective sensemaking." Information Visualization 16, no. 1 (July 25, 2016): 21–47. http://dx.doi.org/10.1177/1473871616630778.
Full textAbstract:
In our daily lives, images are among the most commonly found data which we need to handle. We present iGraph, a graph-based approach for visual analytics of large image collections and their associated text information. Given such a collection, we compute the similarity between images, the distance between texts, and the connection between image and text to construct iGraph, a compound graph representation which encodes the underlying relationships among these images and texts. To enable effective visual navigation and comprehension of iGraph with tens of thousands of nodes and hundreds of millions of edges, we present a progressive solution that offers collection overview, node comparison, and visual recommendation. Our solution not only allows users to explore the entire collection with representative images and keywords but also supports detailed comparison for understanding and intuitive guidance for navigation. The visual exploration of iGraph is further enhanced with the implementation of bubble sets to highlight group memberships of nodes, suggestion of abnormal keywords or time periods based on text outlier detection, and comparison of four different recommendation solutions. For performance speedup, multiple graphics processing units and central processing units are utilized for processing and visualization in parallel. We experiment with two image collections and leverage a cluster driving a display wall of nearly 50 million pixels. We show the effectiveness of our approach by demonstrating experimental results and conducting a user study.
13
B.D.C.N, Prasad, M. Sailaja, and V. Suryanarayana. "Analysis on Content Based Image Retrieval Using Image Enhancement and Deep Learning Convolutional Neural Networks." ECS Transactions 107, no. 1 (April 24, 2022): 19777–89. http://dx.doi.org/10.1149/10701.19777ecst.
Full textAbstract:
"Content-Based" means that an image contents search analyzes instead of meta data, including keywords, tags or image descriptions. The word contents could apply in this sense to colours, structures, textures or any details extracted from the picture itself. CBIR is desirable as searches relying purely on metadata depend on the quality and the completeness of annotations. CBIR method for the recovery of images from huge, unshaped image databases is commonly used. The CBIR method is used. Therefore, users are not satisfied with standard knowledge collection methods. In addition, there are more images available to users, as well as the advent of web creation and transmission networks. Consequently, there is a permanent and important output of digital images in many regions. Hence the rapid access to these enormous picture collections and the identical image retrieval from this broad image collection presents major challenges and demands efficient techniques. The efficiency of a content-based image retrieval system depends on the characteristic representation and similarity calculation. We therefore have a simple but powerful, profound, CNN-based, and feature-extraction and classification-based imaging system. Some promising results have been obtained from a range of empirical studies on a variety of CB IR tasks through the image database. Content-based image recovery systems (CBIR) allow you to find images identical to a query image among a picture dataset. The best-known CBIR system is Google's search by image feature.
14
Saleh, Emad Isa. "Image embedded metadata in cultural heritage digital collections on the web." Library Hi Tech 36, no. 2 (June 18, 2018): 339–57. http://dx.doi.org/10.1108/lht-03-2017-0053.
Full textAbstract:
Purpose The purpose of this paper is to investigate the availability of embedded metadata within images of digital cultural collections. It is designed to examine a proposed hypothesis that most digitally derived images of cultural resources are stripped of their metadata once they are placed on the web. Design/methodology/approach A sample of 603 images were selected randomly from four cultural portals which aggregate digitized cultural collections, then four steps in the data collection process took place to examine image metadata via the web-based tool and windows application. Findings The study revealed that 28.5 percent of the analyzed images contained metadata, no links exist between image embedded metadata and its metadata record or the pages of the websites analyzed, and there is a significant usage of Extensible Metadata Platform to encode embedded metadata within the images. Practical implications The findings of the study may encourage heritage digital collection providers to reconsider their metadata preservation practices and policies to enrich the content of embedded metadata. In addition, it will raise awareness about the potential and value of embedded metadata in enhancing the findability and exchange of digital collections. Originality/value This study is ground breaking in that it is one of the early studies, especially in the Arab world, which aim to recognize the use of image embedded metadata within cultural heritage digital collections on the web.
15
Therrell, Grace. "More product, more process: metadata in digital image collections." Digital Library Perspectives 35, no. 1 (February 11, 2019): 2–14. http://dx.doi.org/10.1108/dlp-06-2018-0018.
Full textAbstract:
Purpose The purpose of this paper is to discuss the implications of current theories that advocate for minimal levels of description in digital collections. Specifically, this paper looks at the archival theory of “More Product, Less Process” and its encouragement of collection-level description. The purpose of the study was to analyze how levels of description impact resource retrieval. Design/methodology/approach This study analyzed 35 images from a New York Public Library (NYPL) digital collection present on the NYPL website and on Flickr. The methodology was designed to reflect users’ information seeking behavior for image collections. There were two research questions guiding this study: what are the descriptive terms used to describe items in digital collections? and what is the success rate of retrieving resources using assigned descriptive terms? Findings The results of this study revealed that the images from the NYPL collection were more difficult to find on the institution’s website as compared with Flickr. These findings suggest that lesser levels of description in digital collections hinder resource retrieval. Research limitations/implications These findings suggest that lesser description levels hurt the findability of resources. In the wake of theories such as “More Product, Less Process”, information professionals must find ways to assign metadata to individual materials in digital image collections. Originality/value Recent research concerning description levels of digital collections is several years old and focuses mostly on the usefulness of collection-level metadata as a supplement to or substitute for item-level metadata. Few, if any, studies exist that explore the implications of description levels on resource retrievability and findability. This study is also unique in that it discusses these implications in the context of less-is-more theories of archival processing.
16
Schwendy, Mischa, Ronald E. Unger, and Sapun H. Parekh. "EVICAN—a balanced dataset for algorithm development in cell and nucleus segmentation." Bioinformatics 36, no. 12 (April 1, 2020): 3863–70. http://dx.doi.org/10.1093/bioinformatics/btaa225.
Full textAbstract:
Abstract Motivation Deep learning use for quantitative image analysis is exponentially increasing. However, training accurate, widely deployable deep learning algorithms requires a plethora of annotated (ground truth) data. Image collections must contain not only thousands of images to provide sufficient example objects (i.e. cells), but also contain an adequate degree of image heterogeneity. Results We present a new dataset, EVICAN—Expert visual cell annotation, comprising partially annotated grayscale images of 30 different cell lines from multiple microscopes, contrast mechanisms and magnifications that is readily usable as training data for computer vision applications. With 4600 images and ∼26 000 segmented cells, our collection offers an unparalleled heterogeneous training dataset for cell biology deep learning application development. Availability and implementation The dataset is freely available (https://edmond.mpdl.mpg.de/imeji/collection/l45s16atmi6Aa4sI?q=). Using a Mask R-CNN implementation, we demonstrate automated segmentation of cells and nuclei from brightfield images with a mean average precision of 61.6 % at a Jaccard Index above 0.5.
17
Kalukin, Andrew, Satoshi Endo, Russell Crook, Manoj Mahajan, Robert Fennimore, Alice Cialella, Laurie Gregory, Shinjae Yoo, Wei Xu, and Daniel Cisek. "Image Collection Simulation Using High-Resolution Atmospheric Modeling." Remote Sensing 12, no. 19 (October 1, 2020): 3214. http://dx.doi.org/10.3390/rs12193214.
Full textAbstract:
A new method is described for simulating the passive remote sensing image collection of ground targets that includes effects from atmospheric physics and dynamics at fine spatial and temporal scales. The innovation in this research is the process of combining a high-resolution weather model with image collection simulation to attempt to account for heterogeneous and high-resolution atmospheric effects on image products. The atmosphere was modeled on a 3D voxel grid by a Large-Eddy Simulation (LES) driven by forcing data constrained by local ground-based and air-based observations. The spatial scale of the atmospheric model (10–100 m) came closer than conventional weather forecast scales (10–100 km) to approaching the scale of typical commercial multispectral imagery (2 m). This approach was demonstrated through a ground truth experiment conducted at the Department of Energy Atmospheric Radiation Measurement Southern Great Plains site. In this experiment, calibrated targets (colored spectral tarps) were placed on the ground, and the scene was imaged with WorldView-3 multispectral imagery at a resolution enabling the tarps to be visible in at least 9–12 image pixels. The image collection was simulated with Digital Imaging and Remote Sensing Image Generation (DIRSIG) software, using the 3D atmosphere from the LES model to generate a high-resolution cloud mask. The high-resolution atmospheric model-predicted cloud coverage was usually within 23% of the measured cloud cover. The simulated image products were comparable to the WorldView-3 satellite imagery in terms of the variations of cloud distributions and spectral properties of the ground targets in clear-sky regions, suggesting the potential utility of the proposed modeling framework in improving simulation capabilities, as well as testing and improving the operation of image collection processes.
18
Kesiman, Made Windu Antara, and Gede Aditra Pradnyana. "A Scheme Towards Automatic Word Indexation System for Balinese Palm Leaf Manuscripts." Journal of ICT Research and Applications 15, no. 2 (October 7, 2021): 105–19. http://dx.doi.org/10.5614/itbj.ict.res.appl.2021.15.2.1.
Full textAbstract:
This paper proposes an initial scheme towards the development of an automatic word indexation system for Balinese lontar (palm leaf manuscript) collections. The word indexation system scheme consists of a sub module for patch image extraction of text areas in lontars and a sub module for word image transliteration. This is the first word indexation system for lontar collections to be proposed. To detect parts of a lontar image that contain text, a Gabor filter is used to provide initial information about the presence of text texture in the image. An adaptive sliding patch algorithm for the extraction of patch images in lontars is also proposed. The word image transliteration sub module was built using the long short-term memory (LSTM) model. The results showed that the image patch extraction of text areas process succeeded in optimally detecting text areas in lontars and extracting the patch image in a suitable position. The proposed scheme successfully extracted between 20% to 40% of the keywords in lontars and thus can at least provide an initial description for prospective lontar readers of the content contained in a lontar collection or to find in which lontar collection certain keywords can be found.
19
Martell, Charles. "Editorial: Collection Development: The JANUS Image." College & Research Libraries 46, no. 2 (March 1, 1985): 109–10. http://dx.doi.org/10.5860/crl_46_02_109.
Full text
20
Dauter, Z. "Image-plate data collection in Hamburg." Acta Crystallographica Section A Foundations of Crystallography 49, s1 (August 21, 1993): c12. http://dx.doi.org/10.1107/s010876737809964x.
Full text
21
Toet, Alexander. "The TNO Multiband Image Data Collection." Data in Brief 15 (December 2017): 249–51. http://dx.doi.org/10.1016/j.dib.2017.09.038.
Full text
22
Kramer, Elsa F. "IUPUI image collection: a usability survey." OCLC Systems & Services: International digital library perspectives 21, no. 4 (December 2005): 346–59. http://dx.doi.org/10.1108/10650750510631712.
Full text
23
Rorvig, M. E., C. H. Turner, and J. Moncada. "The NASA Image Collection Visual Thesaurus." Journal of the American Society for Information Science 50, no. 9 (1999): 794–98. http://dx.doi.org/10.1002/(sici)1097-4571(1999)50:9<794::aid-asi8>3.0.co;2-y.
Full text
24
Chu, Boce, Feng Gao, Yingte Chai, Yu Liu, Chen Yao, Jinyong Chen, Shicheng Wang, Feng Li, and Chao Zhang. "Large-Area Full-Coverage Remote Sensing Image Collection Filtering Algorithm for Individual Demands." Sustainability 13, no. 23 (December 6, 2021): 13475. http://dx.doi.org/10.3390/su132313475.
Full textAbstract:
Remote sensing is the main technical means for urban researchers and planners to effectively observe targeted urban areas. Generally, it is difficult for only one image to cover a whole urban area and one image cannot support the demands of urban planning tasks for spatial statistical analysis of a whole city. Therefore, people often artificially find multiple images with complementary regions in an urban area on the premise of meeting the basic requirements for resolution, cloudiness, and timeliness. However, with the rapid increase of remote sensing satellites and data in recent years, time-consuming and low performance manual filter results have become more and more unacceptable. Therefore, the issue of efficiently and automatically selecting an optimal image collection from massive image data to meet individual demands of whole urban observation has become an urgent problem. To solve this problem, this paper proposes a large-area full-coverage remote sensing image collection filtering algorithm for individual demands (LFCF-ID). This algorithm achieves a new image filtering mode and solves the difficult problem of selecting a full-coverage remote sensing image collection from a vast amount of data. Additionally, this is the first study to achieve full-coverage image filtering that considers user preferences concerning spatial resolution, timeliness, and cloud percentage. The algorithm first quantitatively models demand indicators, such as cloudiness, timeliness, resolution, and coverage, and then coarsely filters the image collection according to the ranking of model scores to meet the different needs of different users for images. Then, relying on map gridding, the image collection is genetically optimized for individuals using a genetic algorithm (GA), which can quickly remove redundant images from the image collection to produce the final filtering result according to the fitness score. The proposed method is compared with manual filtering and greedy retrieval to verify its computing speed and filtering effect. The experiments show that the proposed method has great speed advantages over traditional methods and exceeds the results of manual filtering in terms of filtering effect.
25
Yangwen, Zheng. "Chinese Collection 457: the Call for Global History." Bulletin of the John Rylands Library 91, no. 1 (March 2015): 35–44. http://dx.doi.org/10.7227/bjrl.91.1.3.
Full textAbstract:
With the help of the Jesuits, the Qianlong emperor (often said to be Chinas Sun King in the long eighteenth century) built European palaces in the Garden of Perfect Brightness and commissioned a set of twenty images engraved on copper in Paris. The Second Anglo-Chinese Opium War in 1860 not only saw the destruction of the Garden, but also of the images, of which there are only a few left in the world. The John Rylands set contains a coloured image which raises even more questions about the construction of the palaces and the after-life of the images. How did it travel from Paris to Bejing, and from Belgium to the John Rylands Library? This article probes the fascinating history of this image. It highlights the importance of Europeans in the making of Chinese history and calls for studies of China in Europe.
26
Decker, Peter, Axel Christian, and Willi E. R. Xylander. "VIRMISCO – The Virtual Microscope Slide Collection." ZooKeys 741 (March 7, 2018): 271–82. http://dx.doi.org/10.3897/zookeys.741.22284.
Full textAbstract:
Digitisation allows scientists rapid access to research objects. For transparent to semi-transparent three-dimensional microscopic objects, such as microinvertebrates or small body parts of organisms, available databases are scarce. Most mounting media used for permanent microscope slides deteriorate after some years or decades, eventually leading to total damage and loss of the object. However, restoration is labour-intensive, and often the composition of the mounting media is not known. A digital preservation of important material, especially types, is important and an urgent need. The Virtual Microscope Slide Collection – VIRMISCO project has developed recommendations for taking microscopic image stacks of three-dimensional objects, depositing and presenting such series of digital image files or z-stacks as an online platform. The core of VIRMISCO is an online viewer, which enables the user to virtually focus through an object online as if using a real microscope. Additionally, VIRMISCO offers features such as search, rotating, zooming, measuring, changing brightness or contrast, taking snapshots, leaving feedback as well as downloading complete z-stacks as jpeg files or video file. The open source system can be installed by any institution and can be linked to common database or images can be sent to the Senckenberg Museum of Natural History Görlitz. The benefits of VIRMISCO are the preservation of important or fragile material, to avoid loan, to act as a digital archive for image files and to allow determination by experts from the distance, as well as providing reference libraries for taxonomic research or education and providing image series as online supplementary material for publications or digital vouchers of specimens of molecular investigations are relevant applications for VIRMISCO.
27
Oktaviantina, Adek Dwi. "CITRAAN DALAM KUMPULAN PUISI ABDUL SALAM HS “MALAIKAT WARINGIN”." BEBASAN Jurnal Ilmiah Kebahasaan dan Kesastraan 6, no. 2 (March 9, 2020): 137. http://dx.doi.org/10.26499/bebasan.v6i2.118.
Full textAbstract:
Poetry is arranged beautiful words and understood by the readers. Therefore, poetry is inseparable from the accuracy of the language which is arranged aesthetically and creatively so that the meaning of poetry can be conveyed properly. Interpretive descriptive of images is used as the technique of this research. Image, as a way of looking at the meaning, is identified through the human senses. The method used is interpretative qualitative research method. The formulation of the problem is how the image in the poetru collection of Abdul Salam HS entitled "Malaikat Waringin". The purpose of this research is to describe the image in Abdul Salam HS‟ Poetry Collection. The results from analyzing fifty poems contained in the poetry collection entitled "Malaikat Waringin", it showed that there are 17 visual image data, 9 auditory image data, 4 smell and taste image images, and 6 visualization image data.
28
Moskolaï, Waytehad Rose, Wahabou Abdou, Albert Dipanda, and Kolyang Kolyang. "A Workflow for Collecting and Preprocessing Sentinel-1 Images for Time Series Prediction Suitable for Deep Learning Algorithms." Geomatics 2, no. 4 (October 1, 2022): 435–56. http://dx.doi.org/10.3390/geomatics2040024.
Full textAbstract:
The satellite image time series are used for several applications such as predictive analysis. New techniques such as deep learning (DL) algorithms generally require long sequences of data to perform well; however, the complexity of satellite image preprocessing tasks leads to a lack of preprocessed datasets. Moreover, using conventional collection and preprocessing methods is time- and storage-consuming. In this paper, a workflow for collecting, preprocessing, and preparing Sentinel-1 images to use with DL algorithms is proposed. The process mainly consists of using scripts for collecting and preprocessing operations. The goal of this work is not only to provide the community with easily modifiable programs for image collection and batch preprocessing but also to publish a database with prepared images. The experimental results allowed the researchers to build three time series of Sentinel-1 images corresponding to three study areas, namely the Bouba Ndjida National Park, the Dja Biosphere Reserve, and the Wildlife Reserve of Togodo. A total of 628 images were processed using scripts based on the SNAP graph processing tool (GPT). In order to test the effectiveness of the proposed methodology, three DL models were trained with the Bouba Ndjida and Togodo images for the prediction of the next occurrence in a sequence.
29
MacKenzie, Adrian, and Anna Munster. "Platform Seeing: Image Ensembles and Their Invisualities." Theory, Culture & Society 36, no. 5 (June 3, 2019): 3–22. http://dx.doi.org/10.1177/0263276419847508.
Full textAbstract:
How can one ‘see’ the operationalization of contemporary visual culture, given the imperceptibility and apparent automation of so many processes and dimensions of visuality? Seeing – as a position from a singular mode of observation – has become problematic since many visual elements, techniques, and forms of observing are highly distributed through data practices of collection, analysis and prediction. Such practices are subtended by visual cultural techniques that are grounded in the development of image collections, image formatting and hardware design. In this article, we analyze recent transformations in forms of prediction and data analytics associated with spectacular performances of computation. We analyze how transformations in the collection and accumulation of images as ensembles by platforms have a qualitative and material effect on the emergent sociotechnicality of platform ‘life’ and ‘perception’. Reconstructing the visual transformations that allow artificial intelligence assemblages to operate allows some sense of their heteronomous materiality and contingency.
30
Sun, Ke Mei. "Wireless Image Transmission System." Applied Mechanics and Materials 336-338 (July 2013): 1661–64. http://dx.doi.org/10.4028/www.scientific.net/amm.336-338.1661.
Full textAbstract:
A wireless image transmission system is proposed, which in order to offset the disadvantages of wire video monitor laying points hard, and disadvantages of wire video monitor implementing hard in bad circumstance. The core of the system is TI series of C5000 high speed DSP processor and radio frequency wireless communication chip. Through DSP transplantation of JPEG compression algorithm, complete image collection and image compression through DSP, use wireless receiving and sending module to complete wireless images data transmission. Use wireless receiving and sending module instead of emitter and network as image data transmission device, low cost, not required network fee, system can real-time complete terminal images collection and show. It can apply community guard and video monitor of deploying points urgently in the electric power system, factory, bank and other important department, which dont require higher performance.
31
Hoogendoorn, S. P., H. J. Van Zuylen, M. Schreuder, B. Gorte, and G. Vosselman. "Microscopic Traffic Data Collection by Remote Sensing." Transportation Research Record: Journal of the Transportation Research Board 1855, no. 1 (January 2003): 121–28. http://dx.doi.org/10.3141/1855-15.
Full textAbstract:
To gain insight into the behavior of drivers during congestion, and to develop and test theories and models that describe congested driving behavior, very detailed data are needed. A new data-collection system prototype is described for determining individual vehicle trajectories from sequences of digital aerial images. Software was developed to detect and track vehicles from image sequences. In addition to longitudinal and lateral position as a function of time, the system can determine vehicle length and width. Before vehicle detection and tracking can be achieved, the software handles correction for lens distortion, radiometric correction, and orthorectification of the image. The software was tested on data collected from a helicopter by a digital camera that gathered high-resolution monochrome images, covering 280 m of a Dutch motorway. From the test, it was concluded that the techniques for analyzing the digital images can be applied automatically without much problem. However, given the limited stability of the helicopter, only 210 m of the motorway could be used for vehicle detection and tracking. The resolution of the data collection was 22 cm. Weather conditions appear to have a significant influence on the reliability of the data: 98% of the vehicles could be detected and tracked automatically when conditions were good; this number dropped to 90% when the weather conditions worsened. Equipment for stabilizing the camera—gyroscopic mounting—and the use of color images can be applied to further improve the system.
32
Liu, Weiping, Jennifer Fung, W. J. de Ruijter, Hans Chen, John W. Sedat, and David A. Agard. "Automated electron tomography: from data collection to image processing." Proceedings, annual meeting, Electron Microscopy Society of America 53 (August 13, 1995): 26–27. http://dx.doi.org/10.1017/s0424820100136507.
Full textAbstract:
Electron tomography is a technique where many projections of an object are collected from the transmission electron microscope (TEM), and are then used to reconstruct the object in its entirety, allowing internal structure to be viewed. As vital as is the 3-D structural information and with no other 3-D imaging technique to compete in its resolution range, electron tomography of amorphous structures has been exercised only sporadically over the last ten years. Its general lack of popularity can be attributed to the tediousness of the entire process starting from the data collection, image processing for reconstruction, and extending to the 3-D image analysis. We have been investing effort to automate all aspects of electron tomography. Our systems of data collection and tomographic image processing will be briefly described.To date, we have developed a second generation automated data collection system based on an SGI workstation (Fig. 1) (The previous version used a micro VAX). The computer takes full control of the microscope operations with its graphical menu driven environment. This is made possible by the direct digital recording of images using the CCD camera.
33
Jin, Shaojie, Ying Gao, Shoucai Jing, Fei Hui, Xiangmo Zhao, and Jianzhen Liu. "Traffic Flow Parameters Collection under Variable Illumination Based on Data Fusion." Journal of Advanced Transportation 2021 (August 15, 2021): 1–14. http://dx.doi.org/10.1155/2021/4592124.
Full textAbstract:
Accurate traffic flow parameters are the supporting data for analyzing traffic flow characteristics. Vehicle detection using traffic surveillance pictures is a typical method for gathering traffic flow characteristics in urban traffic scenes. In complicated lighting conditions at night, however, neither classical nor deep-learning-based image processing algorithms can provide adequate detection results. This study proposes a fusion technique combining millimeter-wave radar data with image data to compensate for the lack of image-based vehicle detection under complicated lighting to complete all-day parameters collection. The proposed method is based on an object detector named CenterNet. Taking this network as the cornerstone, we fused millimeter-wave radar data into it to improve the robustness of vehicle detection and reduce the time-consuming postcalculation of traffic flow parameters collection. We collected a new dataset to train the proposed method, which consists of 1000 natural daytime images and 1000 simulated nighttime images with a total of 23094 vehicles counted, where the simulated nighttime images are generated by a style translator named CycleGAN to reduce labeling workload. Another four datasets of 2400 images containing 20161 vehicles were collected to test the proposed method. The experimental results show that the method proposed has good adaptability and robustness at natural daytime and nighttime scenes.
34
Hodosh, M., P. Young, and J. Hockenmaier. "Framing Image Description as a Ranking Task: Data, Models and Evaluation Metrics." Journal of Artificial Intelligence Research 47 (August 30, 2013): 853–99. http://dx.doi.org/10.1613/jair.3994.
Full textAbstract:
The ability to associate images with natural language sentences that describe what is depicted in them is a hallmark of image understanding, and a prerequisite for applications such as sentence-based image search. In analogy to image search, we propose to frame sentence-based image annotation as the task of ranking a given pool of captions. We introduce a new benchmark collection for sentence-based image description and search, consisting of 8,000 images that are each paired with five different captions which provide clear descriptions of the salient entities and events. We introduce a number of systems that perform quite well on this task, even though they are only based on features that can be obtained with minimal supervision. Our results clearly indicate the importance of training on multiple captions per image, and of capturing syntactic (word order-based) and semantic features of these captions. We also perform an in-depth comparison of human and automatic evaluation metrics for this task, and propose strategies for collecting human judgments cheaply and on a very large scale, allowing us to augment our collection with additional relevance judgments of which captions describe which image. Our analysis shows that metrics that consider the ranked list of results for each query image or sentence are significantly more robust than metrics that are based on a single response per query. Moreover, our study suggests that the evaluation of ranking-based image description systems may be fully automated.
35
Jalali-Yazdi, Farzad, and Eric Gouaux. "Image collection strategies for single particle cryoEM." Microscopy and Microanalysis 27, S1 (July 30, 2021): 1904–6. http://dx.doi.org/10.1017/s1431927621006930.
Full text
36
SHAFI, MAHMOOD I., and DAVID M. LUESLEY. "Modern Image Capture and Data Collection Technology." Clinical Obstetrics and Gynecology 38, no. 3 (September 1995): 640–43. http://dx.doi.org/10.1097/00003081-199509000-00023.
Full text
37
Hsieh, Che-Jen, Jung-Shian Li, and Wei-Cheng Liu. "Digital evidence collection for web image access." Security and Communication Networks 6, no. 6 (September 28, 2012): 765–76. http://dx.doi.org/10.1002/sec.610.
Full text
38
Akhmetvaleev, R. R., I. A. Lackman, D. V. Popov, and M. V. Krasnoperov. "Image segmentation technique to support automatic marking of objects in endoscopic images." Informatization and communication, no. 2 (February 16, 2021): 146–54. http://dx.doi.org/10.34219/2078-8320-2021-12-2-146-154.
Full textAbstract:
The aim of this study is to develop a method for visual segmentation of various objects of endoscopic images based on a collection of endoscopic images. The method was developed on the basis of a collection of images obtained by ENVD LLC on a contractual basis with medical organizations of the Republic of Bashkortostan, Russia. The collection consists of 70 endoscopic images recording clinical cases diagnosed in accordance with the Paris Tumor Classification of Gastrointestinal Diseases. A number of machine vision operations were carried out, including image preprocessing, image sampling, and subsequent clustering for the purpose of image segmentation. Results: A technique for the analysis of endoscopic images was developed, which makes it possible to obtain the contours of objects of interest to a specialist performing endoscopy. Conclusion. The developed solution allows to speed up and improve the procedure for marking endoscopic images, which in turn prepares a platform for further processing of endoscopic images, for example, nosological classification of neoplasms.
39
Li, Q., Z. Zhang, W. Lu, J. Yang, Y. Ma, and W. Yao. "From pixels to patches: a cloud classification method based on bag of micro-structures." Atmospheric Measurement Techniques Discussions 8, no. 10 (October 5, 2015): 10213–47. http://dx.doi.org/10.5194/amtd-8-10213-2015.
Full textAbstract:
Abstract. Automatic cloud classification has attracted more and more attention with the increasing development of whole sky imagers, but it is still in progress for ground-based cloud observation. This paper proposes a new cloud classification method, named bag of micro-structures (BoMS). This method treats an all-sky image as a collection of micro-structures mapped from image patches, rather than a collection of pixels. And then it constructs an image representation with a weighted histogram of micro-structures. Lastly, a support vector machine (SVM) classifier is applied on the image representation because SVM is appealing for sparse and high dimensional feature space. Five different sky conditions are identified: cirriform, cumuliform, stratiform, clear sky and mixed cloudiness that often appears in all-sky images but is seldom addressed in literature. BoMS is evaluated on a large dataset, which contains 5000 all-sky images that are captured by a total-sky cloud imager located in Tibet (29.25° N, 88.88° E). BoMS achieves an accuracy of 90.9 % for 10 fold cross-validation, and it outperforms the state-of-the-art method with an increase of about 19 %. Furthermore, influence of key parameters in BoMS are investigated to verify their robustness.
40
Woodward-Greene, M. Jennifer, Jason M. Kinser, Tad S. Sonstegard, Johann Sölkner, Iosif I. Vaisman, and Curtis P. Van Tassell. "PreciseEdge raster RGB image segmentation algorithm reduces user input for livestock digital body measurements highly correlated to real-world measurements." PLOS ONE 17, no. 10 (October 13, 2022): e0275821. http://dx.doi.org/10.1371/journal.pone.0275821.
Full textAbstract:
Computer vision is a tool that could provide livestock producers with digital body measures and records that are important for animal health and production, namely body height and length, and chest girth. However, to build these tools, the scarcity of labeled training data sets with uniform images (pose, lighting) that also represent real-world livestock can be a challenge. Collecting images in a standard way, with manual image labeling is the gold standard to create such training data, but the time and cost can be prohibitive. We introduce the PreciseEdge image segmentation algorithm to address these issues by employing a standard image collection protocol with a semi-automated image labeling method, and a highly precise image segmentation for automated body measurement extraction directly from each image. These elements, from image collection to extraction are designed to work together to yield values highly correlated to real-world body measurements. PreciseEdge adds a brief preprocessing step inspired by chromakey to a modified GrabCut procedure to generate image masks for data extraction (body measurements) directly from the images. Three hundred RGB (red, green, blue) image samples were collected uniformly per the African Goat Improvement Network Image Collection Protocol (AGIN-ICP), which prescribes camera distance, poses, a blue backdrop, and a custom AGIN-ICP calibration sign. Images were taken in natural settings outdoors and in barns under high and low light, using a Ricoh digital camera producing JPG images (converted to PNG prior to processing). The rear and side AGIN-ICP poses were used for this study. PreciseEdge and GrabCut image segmentation methods were compared for differences in user input required to segment the images. The initial bounding box image output was captured for visual comparison. Automated digital body measurements extracted were compared to manual measures for each method. Both methods allow additional optional refinement (mouse strokes) to aid the segmentation algorithm. These optional mouse strokes were captured automatically and compared. Stroke count distributions for both methods were not normally distributed per Kolmogorov-Smirnov tests. Non-parametric Wilcoxon tests showed the distributions were different (p< 0.001) and the GrabCut stroke count was significantly higher (p = 5.115 e-49), with a mean of 577.08 (std 248.45) versus 221.57 (std 149.45) with PreciseEdge. Digital body measures were highly correlated to manual height, length, and girth measures, (0.931, 0.943, 0.893) for PreciseEdge and (0.936, 0. 944, 0.869) for GrabCut (Pearson correlation coefficient). PreciseEdge image segmentation allowed for masks yielding accurate digital body measurements highly correlated to manual, real-world measurements with over 38% less user input for an efficient, reliable, non-invasive alternative to livestock hand-held direct measuring tools.
41
Agard, D. A., A. J. Koster, M. B. Braunfeld, and J. W. Sedat. "Automated data collection for electron tomography." Proceedings, annual meeting, Electron Microscopy Society of America 50, no. 2 (August 1992): 1044–45. http://dx.doi.org/10.1017/s0424820100129851.
Full textAbstract:
Three-dimensional imaging has become an important addition to the variety of methods available for research on biological structures. Non-crystalline samples can be examined by high resolution electron tomography which requires that projection data be collected over a large range of specimen tilts. Practical limitations of tomography are set by the large number of micrographs to be processed, and by the required (and tedious) recentering and refocusing of the object during data collection; especially for dose sensitive specimens. With automated electron tomography a number of these problems can be overcome. First, the images are recorded directly in digital format, using a cooled slow scan CCD camera, and, with automatic tracking and correction for image shift and focus variation, a pre-aligned dataset is obtained, with every image recorded under well defined imaging conditions.At UCSF, we use intermediate voltage electron tomography to study higher-order chromatin structure. Of central interest is elucidating the higher-order arrangement of the 30nm chromatin fiber within condensed chromosomes through several phases of the cell cycle and, in collaboration with Chris Woodcock, the structure of the 30 nm fiber.
42
Garris, M. D. "Design and Collection of a Handwriting Sample Image Database." Social Science Computer Review 10, no. 2 (July 1992): 196–214. http://dx.doi.org/10.1177/089443939201000205.
Full textAbstract:
NIST, under sponsorship from the Bureau of the Census, has collected a database consisting of 2,100 pages of binary image data of handprinted characters including numerals and text. NIST Special Database 1 contains handwriting samples from 2,100 writers geographically distributed across the United States, NIST is currently using this database to research field-isolation, box detection and removal, character segmentation, and writer-independent neural character recognition. In addition to advancing the design of algorithms, this database and its study can aid the social sciences. Observations can be compiled and used to improve form design and field layout strategies. Region-based studies and comparisons are also possible due to regionally distributed and referenced populations of writers in the database. In creating this large database, obstacles to the effective archiving of images have been dealt with and eliminated. This paper describes the database's content in terms of its collection, organization, and usefulness. The strategies and conventions used to develop NIST Special Database 1 can be applied to any discipline requiring the use of image archives.
43
Mai, Nicole Tham Ley, Syahmi Syahiran Bin Ahmad Ridzuan, and Zaid Bin Omar. "Content-based Image Retrieval System for an Image Gallery Search Application." International Journal of Electrical and Computer Engineering (IJECE) 8, no. 3 (June 1, 2018): 1903. http://dx.doi.org/10.11591/ijece.v8i3.pp1903-1912.
Full textAbstract:
Content-based image retrieval is a process framework that applies computer vision techniques for searching and managing large image collections more efficiently. With the growth of large digital image collections triggered by rapid advances in electronic storage capacity and computing power, there is a growing need for devices and computer systems to support efficient browsing, searching, and retrieval for image collections. Hence, the aim of this project is to develop a content-based image retrieval system that can be implemented in an image gallery desktop application to allow efficient browsing through three different search modes: retrieval by image query, retrieval by facial recognition, and retrieval by text or tags. In this project, the MPEG-7-like Powered Localized Color and Edge Directivity Descriptor is used to extract the feature vectors of the image database and the facial recognition system is built around the Eigenfaces concept. A graphical user interface with the basic functionality of an image gallery application is also developed to implement the three search modes. Results show that the application is able to retrieve and display images in a collection as thumbnail previews with high retrieval accuracy and medium relevance and the computational requirements for subsequent searches were significantly reduced through the incorporation of text-based image retrieval as one of the search modes. All in all, this study introduces a simple and convenient way of offline image searches on desktop computers and provides a stepping stone to future content-based image retrieval systems built for similar purposes.
44
Ørnager, Susanne, and Haakon Lund. "Billedindeksering og sociale medier." Nordisk Tidsskrift for Informationsvidenskab og Kulturformidling 8, no. 1 (August 26, 2019): 2–21. http://dx.doi.org/10.7146/ntik.v8i1.115599.
Full textAbstract:
This article focuses on the methodologies, organization, and communication of digital image collection research that utilizes social media content. “Image” is here understood as a cultural, conventional, and commercial—stock photo—representation. Two methodologies i.e. PRISMA and Grounded Theory are employed to examine, categorize and analyze images and comprehend how humans consider them. The literature review covers research since 2005, when major social media platforms emerged. It demonstrates that the images on social media have not changed the overall direction of research into image indexing and retrieval, though new topics on crowdsourcing and tagging have emerged in it. A citation analysis includes an overview of co-citation maps that demonstrate the nexus of image research literature and the journals in which they appear. The results point to new possibilities influencing the research by providing large image collections as new testbeds for improving or testing research hypothesis on a new scale.
45
Ouhda, Mohamed, Khalid El Asnaoui, Mohammed Ouanan, and Brahim Aksasse. "A Content Based Image Retrieval Method Based on K-Means Clustering Technique." Journal of Electronic Commerce in Organizations 16, no. 1 (January 2018): 82–96. http://dx.doi.org/10.4018/jeco.2018010107.
Full textAbstract:
With the appearance of many devices that are used in image acquisition comes a large number of images every day. The rapid access to these huge collections of images and retrieval of similar images (Query) from this huge collection of images presents major challenges and requires efficient algorithms. The main goal of the proposed system is to provide an accurate result with lower computational time. For this purpose, the authors apply a new method based on k-means clustering technique to match image's descriptors. This article provides a detailed view of the solution the authors have adopted and which perfectly meets their needs. For validation, they apply all of these techniques on two image databases in order to evaluate the performance of their system.
46
Labbé, Mathilde. "‘Nécessité de procéder obliquement’: Shaping Michel Butor's Image." Nottingham French Studies 58, no. 3 (December 2019): 315–31. http://dx.doi.org/10.3366/nfs.2019.0260.
Full textAbstract:
‘Difficile d'écrire sur soi ou sur ses écrits’ are the first words of a 2003 monograph that Michel Butor published about himself in Seghers's collection ‘Poètes d'aujourd'hui’. This series, like other collections of illustrated monographs, builds the writer's image through a combination of different types of discourse and several actors. Butor's volume on his own work tends to blur such distinctions and to challenge the collection's model, not only because of the intervention of the writer, but also because of Butor's interest in the processes through which his image is crafted. This paper examines a series of six monographs published on Michel Butor between 1964 and 2009, firstly, in order to shed light on the way the image of a high profile writer is carved in collections of illustrated monographs, and, secondly, to reach a better understanding of Michel Butor's relation to critical discourse and to his own image.
47
Pflugrath, J. W. "The finer things in X-ray diffraction data collection." Acta Crystallographica Section D Biological Crystallography 55, no. 10 (October 1, 1999): 1718–25. http://dx.doi.org/10.1107/s090744499900935x.
Full textAbstract:
X-ray diffraction images from two-dimensional position-sensitive detectors can be characterized as thick or thin, depending on whether the rotation-angle increment per image is greater than or less than the crystal mosaicity, respectively. The expectations and consequences of the processing of thick and thin images in terms of spatial overlap, saturated pixels, X-ray background andI/σ(I) are discussed. Thed*TREKsoftware suite for processing diffraction images is briefly introduced, and results fromd*TREKare compared with those from another popular package.
48
Baroni, Travis C., Brendan J. Griffin, and Frank J. Lincoln. "3D Reconstructions Using Charge Contrast Imaging." Microscopy and Microanalysis 5, S2 (August 1999): 284–85. http://dx.doi.org/10.1017/s1431927600014744.
Full textAbstract:
Three-dimensional reconstructions are a useful visualisation aid for understanding the volumetric relationships between features in an object. Typically, three dimensional reconstructions have evolved with regards to biological applications, where the collection and manipulation of the data has matured to a point where volumetric rendering has become almost routine. Two common data collection procedures involve either virtual sectioning using CLSM, where layers within the object are imaged successively without mechanical sectioning, or alternatively the projection of the object is imaged, via TEM while being rotated through a range of angles. This paper describes a novel application of three dimensional imaging by collecting 2 dimensional serial section images using a recently described technique for imaging materials using a VPSEM, known as Charge Contrast Imaging.Figure 1 shows a typical image of a polished grain of gibbsite. Previous work has correlated the appearance of the concentric rings to laboratory controlled precipitation events, thus showing the technique provides information on the precipitation history
49
Aliaga, Daniel G., Paul Rosen, Voicu Popescu, and Ingrid Carlbom. "Image warping for compressing and spatially organizing a dense collection of images." Signal Processing: Image Communication 21, no. 9 (October 2006): 755–69. http://dx.doi.org/10.1016/j.image.2006.07.005.
Full text
50
Li, Qingyong, Zhen Zhang, Weitao Lu, Jun Yang, Ying Ma, and Wen Yao. "From pixels to patches: a cloud classification method based on a bag of micro-structures." Atmospheric Measurement Techniques 9, no. 2 (March 1, 2016): 753–64. http://dx.doi.org/10.5194/amt-9-753-2016.
Full textAbstract:
Abstract. Automatic cloud classification has attracted more and more attention with the increasing development of whole sky imagers, but it is still in progress for ground-based cloud observation. This paper proposes a new cloud classification method, named bag of micro-structures (BoMS). This method treats an all-sky image as a collection of micro-structures mapped from image patches, rather than a collection of pixels. It represents the image with a weighted histogram of micro-structures. Based on this representation, BoMS recognizes the cloud class of the image by a support vector machine (SVM) classifier. Five classes of sky condition are identified: cirriform, cumuliform, stratiform, clear sky, and mixed cloudiness. BoMS is evaluated on a large data set, which contains 5000 all-sky images captured by a total-sky cloud imager located in Tibet (29.25° N, 88.88° E). BoMS achieves an accuracy of 90.9 % for 10-fold cross-validation, and it outperforms state-of-the-art methods with an increase of 19 %. Furthermore, influence of key parameters in BoMS is investigated to verify their robustness.