Dissertations / Theses on the topic 'Fingerprints Classification'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 33 dissertations / theses for your research on the topic 'Fingerprints Classification.'
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 dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
Massimiliani, Lorenzo. "Classification and clustering of video fingerprints: preliminary results." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2021. http://amslaurea.unibo.it/22973/.
Full textSutanto, Kevin. "RNA Sequence Classification Using Secondary Structure Fingerprints, Sequence-Based Features, and Deep Learning." Thesis, Université d'Ottawa / University of Ottawa, 2021. http://hdl.handle.net/10393/41876.
Full textSpäth, Bastian, Matthias Philipp, and Thomas Bartnitzki. "Machine performance and acoustic fingerprints of cutting and drilling." TU Bergakademie Freiberg, 2017. https://tubaf.qucosa.de/id/qucosa%3A23182.
Full textTian, Ye. "Développement d'une méthode de géolocalisation à l'intérieur de bâtiments par classification des fingerprints GSM et fusion de données de capteurs embarqués." Thesis, Paris 6, 2015. http://www.theses.fr/2015PA066027/document.
Full textGPS has long been used for accurate and reliable outdoor localization, but it cannot operate in indoor environments, which suggests developing indoor localization methods that can provide seamless and ubiquitous services for mobile users.In this thesis, indoor localization is realized making use of received signal strength fingerprinting technique based on the existing GSM networks. A room is defined as the minimum location unit, and support vector machine are used as a mean to discriminate the rooms by classifying received signal strengths from very large number of GSM carriers. At the same time, multiple sensors, such as accelerometer and gyroscope, are widely available for modern mobile devices, which provide additional information that helps location determination. The hybrid approach that combines the GSM fingerprinting results with mobile sensor and building layout information using a particle filter provides a more accurate and fine-grained localization result.The results of experiments under realistic conditions demonstrate that correct room number can be obtained 94% of the time provided the derived model is used before significant received signal strength drift sets in. Furthermore, if the training data is sampled over a few days, the performance can remain stable exceeding 80% over a period of months, and can be further improved with various post-processing techniques. Moreover, including the mobile sensors allows the system to localize the mobile trajectory coordinates with high accuracy and reliability
Zhang, Fan. "Fingerprint classification with combined neural networks." Thesis, University of Surrey, 2009. http://epubs.surrey.ac.uk/2216/.
Full textMohamed, Suliman M. "Fingerprint-based biometric recognition allied to fuzzy-neural feature classification." Thesis, Sheffield Hallam University, 2002. http://shura.shu.ac.uk/20071/.
Full textKim, Dae Wook. "Data-Driven Network-Centric Threat Assessment." Wright State University / OhioLINK, 2017. http://rave.ohiolink.edu/etdc/view?acc_num=wright1495191891086814.
Full textWang, Yi, and alice yi wang@gmail com. "Ridge Orientation Modeling and Feature Analysis for Fingerprint Identification." RMIT University. Computer Science and Information Technology, 2009. http://adt.lib.rmit.edu.au/adt/public/adt-VIT20091009.152317.
Full textWescher, Agnes. "Molekularbiologische Typisierung von Streptococcus canis isoliert aus subklinisch mastitiskranken Kühen in hessischen Milchviehbetrieben." Doctoral thesis, Universitätsbibliothek Leipzig, 2009. http://nbn-resolving.de/urn:nbn:de:bsz:15-20090609-095913-0.
Full textEschenhagen, Martin. "Molekulare Untersuchung zweier Belebtschlammanlagen unter besonderer Berücksichtigung der biologischen Phosphorelimination." Doctoral thesis, Technische Universität Dresden, 2003. https://tud.qucosa.de/id/qucosa%3A24360.
Full textBleul, Catrin. "Molekularbiologische Analyse mikrobieller Gemeinschaften in Talsperrensedimenten." Doctoral thesis, Saechsische Landesbibliothek- Staats- und Universitaetsbibliothek Dresden, 2004. http://nbn-resolving.de/urn:nbn:de:swb:14-1097570982718-83940.
Full textVarga, Adam. "Identifikace a charakterizace škodlivého chování v grafech chování." Master's thesis, Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií, 2021. http://www.nusl.cz/ntk/nusl-442388.
Full text吳宜龍. "A Wavelet-Based Fingerprints Classification System." Thesis, 2004. http://ndltd.ncl.edu.tw/handle/80955253641628571144.
Full text中華大學
電機工程學系碩士班
92
One of the objectives of this thesis is to find a set of more suitable features via wavelet transform to identify fingerprints images on-line. A wavelet-based fingerprints classification system had been designed, in which, the fingerprints features are extracted from the wavelet coefficients of the gray-scale fingerprints images. The system parameters are adjusted on the basis of a training set of fingerprints images. Since the pre-processing tasks such as the image enhancement, directional filtering, and ridge thinning that are usually performed on the classical minutiae-based fingerprints classification methods can be eliminated, the wavelet approach provides an efficient way to the fingerprints classification system.
Chen, Chun-Liang, and 陳俊良. "A study on efficient preprocessing and classification for fingerprints." Thesis, 2001. http://ndltd.ncl.edu.tw/handle/80947327177553012199.
Full text淡江大學
資訊工程學系
89
For most of the automatic fingerprint identification systems(AFIS), the performance of identification is significantly affected by that of preprocessing step which is a necessary step for almost all the AFIS’s. Such a preprocessing step includes segmentation, enhancement, noise reduction, binarization and thinning ,etc. An improved preprocessing technique is proposed in this thesis.Firstly, we partition a fingerprint image into small blocks with size 8 x 8 pixels. The variance of gray values in each block is evaluated to distinguish the fingerprint area from the background area, and then the fingerprint area is smoothed to reduce noise. A ridge direction computation algorithm is proposed, which calculates a few pixels locating at ridges or a valleys, instead of all pixels in the block, so that the unnecessary computation is reduced.Fingerprint classification is to assign a given fingerprint to a specific category, which can be done according to the number and locations of the singular points in the fingerprint. This fingerprint classification is used to facilitate the management of large fingerprint databases and to speed up the process of fingerprint matching. Singular points can be found using the ridge direction map and some specific masks, and can be classified into five types : left loop, right loop, whorl , arch and tented arch.The experimental results indicate that the proposed preprocessing method has demonstrated good performance in both the processing speed and the degree of correctness for feature extraction.
Jen, Cheng-Lin, and 任正麟. "Fingerprint Classification Using Singularities." Thesis, 2001. http://ndltd.ncl.edu.tw/handle/67813355446298106764.
Full text國立清華大學
資訊工程學系
89
Fingerprint is an important biometric feature because it’s believed that fingerprint is unique and easiness and the research is studied for a long time. Fingerprint classification provides information for identification. According to the definition of the FBI, fingerprints are classified to eight classes. In the thesis, we only classify fingerprints to four classes: Arch, Left Loop, Right Loop, and Whorl. The thesis describes a set of algorithms using directional image and singularities for fingerprint classification. The approach consists of four major steps. (i)Enhancement, (ii)Directional image computing, (iii)Singular points detection, and (iv)Classification We test the algorithm for the first 800 thumb fingerprint images from NIST Special Database 14. The average recognition rate is 87%.
Yeh, Chun-Nan, and 葉俊男. "A study on Fingerprint Classification." Thesis, 2002. http://ndltd.ncl.edu.tw/handle/27620304293167355153.
Full text國立交通大學
電資學院學程碩士班
90
Fingerprint classification provides an important indexing mechanism in a fingerprint database. An accurate and consistent classification can greatly reduce the fingerprint matching time for a large database. In this thesis, we present a new classification method for fingerprint images. In the proposed method, we classify fingerprints into five classes: arch, left loop, right loop, whorl, and tented arch. The major steps of this method include image enhancement, direction matrix extraction, singular points extraction and classification. Finally, we use the 1900 thumb fingerprints of NIST-4 database to evaluate the performance of the proposed method. The experimental result shows that we are able to achieve a classification accuracy of 88 percent (with 10% rejection).
Lu, Nan-Zone, and 盧南彰. "Minutiae Based Fingerprint Classification System." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/97401644914423900514.
Full text國立清華大學
電機工程學系
97
Fingerprint Classification is a key technique in automatic fingerprint identification systems (AFIS). How to reduce the time of computing in an AFIS with a huge database is an important and necessary issue. Since on July 2006, the international standards organization (ISO) established the standard data format (template) of fingerprint based on minutiae (ISO 19794-2). The minutiae based fingerprint template becomes the international standard of fingerprint authentication/identification systems. However, handling image-based classification system into minutiae-based classification system is still a problem. This thesis present a fingerprint classification algorithm based on minutiae . The fingerprint category is clasified into one of the three classes:right loop, left loop and arch/whorl. Experimental results on live-scan database FVC2002-db1a demonstrate the validity of the proposed algorithm.
Yunn-Ruey, Perng, and 彭韻瑞. "A Study on Fingerprint Classification." Thesis, 1998. http://ndltd.ncl.edu.tw/handle/75883844453924266573.
Full text中原大學
電子工程研究所
86
In this paper, we propose a fingerprints classification method, using the ridge directional map on fingerprints, which is the necessary information in minutiae matching, without adding the other data for classification, and classify the input fingerprint patterns to seven genus: plain arch, tended arch, radial loop, ulnar loop, double loop and accidental. The technique is using characteristic mask to search the three kinds of characteristics: delta, core and whorl in the input fingerprint then classify the input fingerprint patterns for increase the accuracy at minutiae matching and decrease the matching time. A part of the fingerprints image we treat for input patterns are from the database of NIST(National Institute of Standard Technical) and the others are scanned on paper, totally 31 patterns. The input patterns can be successfully classified while their characteristics structure unbroken and the quality of image is not too poor. On the other hand, we also proposed a new technique on refining the fingerprint pattern after thinning it and the generation of ridge directional map, which can work effectively.
Wang, Yao-I., and 王耀億. "An AFIS Using Fingerprint Classification." Thesis, 2001. http://ndltd.ncl.edu.tw/handle/63412069894813032772.
Full text國立清華大學
資訊工程學系
89
In this thesis, we are devoted to the implementation of an automatic fingerprint identification system (AFIS). Fingerprint is considered to be a popular characteristic for personal identification which is necessary for access control, criminal identification and credit card usage. An AFIS can reduce time and laborious effort taken by manual identification effectively. Our AFIS is a minutiae-based matching system. Each fingerprint is recorded with its minutiae and is also classified into one of loop, whorl and arch types. The data set is acquired by FPS 110 Silicon Fingerprint Sensor manufactured by Veridicom. We will describe the whole procedures and show the experimental results.
Cavallaro, Anneliese. "Perceptual Expertise in Fingerprint Classification." Thesis, 2019. http://hdl.handle.net/2440/129100.
Full textFingerprint examiners classify crime scene prints as belonging a left or right hand and to a finger-type – thumb index, middle, ring or little – to help narrow their search for known candidate prints. While fingerprint examiners have been found to have impressive perceptual expertise little in known about their perceptual abilities in this aspect of the fingerprint examination process. The present study served as a first test of fingerprint classification expertise, probing experienced (n = 30) and novice (n = 30) examiners in their ability to classify a controlled, fully rolled, set of prints by hand-type and finger-type in a 10-alternative forced-choice task. Using a yoked novice-expert design performance was measured at two levels of specificity: a coarse-grained level accounting for hand-type classifications (i.e. “left” versus “right”), and a fine-grained level accounting for finger-type classifications (i.e., “thumb”, “index’, “middle”, “ring”, “little”). The results revealed experienced fingerprint examiners were indeed sensitive to the type of hand a fingerprint originated from and were significantly better than novices at these classifications. The experts were also able to classify fingerprints by finger-type, performing significantly above chance. Novices, on the other hand, did not differ from chance at classifying fingerprints by finger-type. These expert-novice differences remained large, even when accounting for response times when classifying prints by hand and finger-type. These data suggest that fingerprint experts are able to generalise their highly specific perceptual expertise with fingerprint to coarser grained levels of analysis: moving from identity to hand and finger classification.
Thesis (B.PsychSc(Hons)) -- University of Adelaide, School of Psychology, 2019
Chiou, Tzone-Kaie, and 邱宗楷. "Using Fuzzy Feature Extraction Fingerprint Classification." Thesis, 2001. http://ndltd.ncl.edu.tw/handle/92760623300716087147.
Full text元智大學
資訊工程學系
89
Fingerprint classification is a useful task for a large database of fingerprint recognition system. Accurate classification can speed up the process of fingerprint recognition. The fingerprint classification method proposed in this paper is based on human thinking and uses fuzzy theory. The key point of human thinking to classify fingerprint is attempting to find out fingerprint ridge, singular points (cores or deltas), direction of ridge, wrinkles or scars as global features. Firstly, in order to determine the fingerprint ridge direction, we need to transform the fingerprint image into 50x50 direction pattern. Then we use a set of pre-defined fuzzy mask to find out the singular points. Finally we use relationship between the singular points to classify the fingerprint. The experimental results of our method exhibit the best performance, a very low sensitivity and good classification accuracy.
Chou, Wen-Chi, and 周文祺. "A Study on Fingerprint Classification Systems." Thesis, 2002. http://ndltd.ncl.edu.tw/handle/89324443782441152146.
Full text國立中央大學
資訊工程研究所
90
Fingerprints are one of the most popular biometrics techniques in both of verification and identification mode because the fingerprints of an individual are unique. To facilitate the management of large fingerprint database and to speedup the process of fingerprint identification, we will first classify fingerprints into several categories such as arch, tented arch, left loop, right loop, and whorl. Several different approaches have been proposed for fingerprint classification. Each has its own advantages and limitations. In this thesis, a new fingerprint classification system is introduced. The proposed system tries to use feature as fewer as possible, while to achieve correct recognition as high as possible. In this system, we first propose an efficient method to transform fingerprint image into block directional image. Then a registration point detection method is applied to locate the center of each block directional image. In the following, several feature are extracted from a window whose center is located at the detected registration point. Finally, a class of Hyper Rectangular Composite Neural Networks (HRCNNs) is trained for fingerprint classification. The system was tested on 4000 images in the NIST-4 database.
Lu, Zhuang-Yuan, and 呂壯元. "Design of the Fingerprint Classification System." Thesis, 2000. http://ndltd.ncl.edu.tw/handle/87539053350097829226.
Full text淡江大學
電機工程學系
88
Preprocessing is a important step of the image processing. For each pixel’s neighborhoods be a unit of fingerprint image , we use sort method and K-mean algorithm to find out the cluster center and threshold value of two objects---ridge and valley respectively. Then for each unit is smoothed and shaped by filter with smooth parameter S and sharp parameter K. After had preprocessing , the original irregular histogram became bimodal histogram , and helpful to binarzation. It might cause classification fail by extracting singular points and correlate locations if fingerprint image is broken and obscure. Therefore we issue a new feature─directional and minute feature, due to it can show the crude and minute textures of the fingerprint image, on the other way it can solve that if image is broken or obscure and get well classification result. For extracting directional and minute features, we must build feature -directional- matrix first. The procedures of building feature-directional —matrix are binarzation image, thinning , calculating ridges slope, smoothing and quantification. Feature-directional-matrix discuss in chapter 2 of this paper respectively. Chapter 4 shows the framework of fingerprint classification system. Chapter 5 presents the experiment result, we can class more kinds of fingerprints and get well classification rate by we issuing the features(direction and minute feature) and classification framework(Neural-Backproprgation-Networks). Final is conclusion and future work.
Wu, Ming Yo, and 吳明祐. "A New Approach for Fingerprint Classification." Thesis, 1996. http://ndltd.ncl.edu.tw/handle/89542190642664821939.
Full text國立交通大學
資訊科學學系
84
A new fingerprint classification method is proposed in this thesis. We classify fingerprints based on the global ridge shapes. The fingerprint image is first locally thresholded and the background is removed. After thresholding, we use the line following technique to obtain a thinned image. Then we establish a 6*5 directional matrix, which represents the global ridge shapes of the fingerprint. In order to establish the directional matrix, we insert five vertical lines into the thinned image and for each intersection point between one of the vertical lines and the thinned fingerprint, we calculate its ridge slope. Then we quantize these slopes into four directions using nonuniform quantization. After all directions of the intersection points are found, the intersection points in each line are divided into six parts. In each part, we calculate the number of each appearing direction and find the direction with maximum number. Then a directional matrix of 6*5 is generated, and some features are extracted from the directional matrix to represent the global ridge shapes. We finally classify fingerprints based on the combinations of these features. Experimental results are given to show that the system has a high classification rate.
Rong, Shys Shyu, and 徐世榮. "Design of a Fingerprint Classification System." Thesis, 2004. http://ndltd.ncl.edu.tw/handle/11305942329801154902.
Full text淡江大學
電機工程學系
92
In the biometrics, using fingerprints is one of the most popular biometrics techniques in both verification and identification systems because the fingerprints of an individual are unique. To facilitate the management of large fingerprint database and to speed up the process of fingerprint identification, fingerprints will be first classified into several categories. Several different effective approaches have been proposed for fingerprint classification. Each has its own advantages and limitations. In this thesis, a new fingerprint classification system is introduced. The proposed system directly extracts the directional information from the thinned image of the fingerprint. The proposed octagon mask is used to find the center point of the interesting region. Then, the direction information of the interesting region is used to be the feature vectors for classifying. In the system, not only is the amount of computation reduced but also can the extracted information be used for identification on AFIS. The system is tested on 2000 images in the NIST-4 database.
Chen, Yu-yi, and 陳育誼. "Fingerprint Image Classification Based on Singular Points." Thesis, 2002. http://ndltd.ncl.edu.tw/handle/74077019496841793955.
Full text國立清華大學
資訊工程學系
90
An automatic fingerprint identification system (AFIS) is one of the most important biometric technologies. How to reduce the time of computing in an AFIS with a huge database is an important and necessary issue. Fingerprint classification provides a practical method. In this thesis, we present a fingerprint classification algorithm based on singular points with novel criteria of a classification scheme. A fingerprint is classified into one of the four classes: arch, right loop, left loop, and whorl. The fingerprint classification was tested on 27,000 images in the Nist14 database as well as on 28 images in a live-scan database. The recognition rate of 83.13% for the Nist14 database and 96.4% for the live-scan database have been achieved.
Hsaio, Fu-Chung, and 蕭輔中. "Fast Fingerprint Classification Based on Normalized Histogram Statistic." Thesis, 2007. http://ndltd.ncl.edu.tw/handle/ghucju.
Full text國立東華大學
資訊工程學系
95
The fingerprint classification is essential for the fingerprint matching. It provides an important classified index for each fingerprint to enhance the efficiency of fingerprint matching in a large fingerprint database. Recently, some fingerprint classification methods were proposed. The Qi et al’s method provides a high classification speed but the lower accuracy. Nevertheless, Shah et al’s method provides the high accuracy but low efficiency. Therefore, our proposed classification method gains the high accuracy and needs simple computation by using the normalized histogram statistic. In this thesis, we proposed three fingerprint classification methods based on statistical analysis. Before classification, we have to preprocess the fingerprint images for enhancing the resolution and locating the core of fingerprint. The first is a pixel-histogram based method but it will be interfered by noise. The second is a line-segment-histogram based classification which reduces the interference but has the problem for classifying the whorl fingerprint. Therefore, a hybrid method combining the “pixel” and the “line-segment” is proposed. Our method can classify the fingerprints into four types: arch, whorl, left-loop and right-loop, and meantime need only simple operations and statistics. We test our methods with the NIST-4 fingerprint database. Experimental results show that our hybrid method is fast and effective for the fingerprint classification.
Hsu, Ching-Fu, and 許景復. "Fingerprint recognition based on classification and orientation field coding." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/64272392529698100972.
Full text國立高雄應用科技大學
光電與通訊研究所
96
In this paper we present an innovative fingerprint classification scheme based on hierarchical singular point detection and traced orientation flow. Contrary to conventional methods, fingerprint is classified into seven categories: right loop, left loop, plain arch, tented arch, plain whorl, S-type (twin loop), and eddy. A novel technique for histogram specification is devised to enhance the separation between ridge and valley for the captured fingerprint. Then we transform the enhanced gray level image into an energy image and segment the impression region for orientation field estimation by projection on eight specified angles to locate the boundary. Hierarchical singular points detection through Poincare index method is used for the preliminary filtering of fingerprint class. Finally, the class label is assigned per the traced orientation flow and related threshold setting. The performance of the proposed method has been validated through experiments on the NIST4 database. For the 2133 images in the tests set, the classification accuracy reaches 87.36% with rejection rate 1.5% in average. In terms of speed, our system is faster, operating at an average processing time 0.9 sec per fingerprint on an AMD64 Athlon CPU 3.0 GHz PC.
Chang, Jeng-Horng, and 張正弘. "Fingerprint CLassification by Ridge DIstribution Sequences and Ridge Distribution Model." Thesis, 2001. http://ndltd.ncl.edu.tw/handle/78873947177873333329.
Full text國立中央大學
資訊工程研究所
89
Ridges and ravines are the main components constituting a fingerprint. Traditional Automatic Fingerprint Identification Systems (AFIS) are mainly based on minutiae matching techniques. The minutiae for fingerprint identification are defined by ridge terminations and ridge bifurcations. Most AFIS perform ridge line following process to automatically detect minutiae based on binary or skeleton fingerprint images. For low-quality fingerprint images, the preprocessing stage of an AFIS produces redundant minutiae or even destroys real minutiae. The minutiae detection algorithms in direct gray-scale domain have been developed to overcome these problems. The first step of gray-scale minutiae detection algorithm is to determine ridge locations and then perform gray-scale ridge line following algorithm to extract minutiae. However, the existing gray-scale minutiae detection techniques can only work on partial fingerprint images due to the ignorance of image background. Moreover, the gray value variation inside a ridge also generates redundant ridge points. In this dissertation, we propose a novel method, based on gray-level histogram decomposition, to locate the ridge points in complete fingerprint images. By decomposing the gray-level histogram, redundant ridge points can be eliminated according to some statistical parameters. Experimental results demonstrate that the correct rate can be over 96% even applied to poor-quality fingerprint images. For automatic fingerprint classification problem, a novel method is introduced which is a combination of structural and syntactic approaches. The goal of the proposed Ridge Distribution (R-D) Model is to present the idea of the possibility for classifying a fingerprint into the complete seven classes in the Henry''s classification. From our observation, there exist only ten basic ridge patterns which construct fingerprints. Fingerprint classes can be interpreted as a combination of these ten ridge patterns with different ridge distribution sequences. In this thesis, the classification task is performed depending on the global distribution of the ten basic ridge patterns by analyzing the ridge shapes and the sequence of ridges distribution. The regular expression for each class is formulated and a NFA model is constructed accordingly. An explicit rejection criterion is also defined in this thesis. For the seven-class fingerprint classification problem, our method can achieve the classification accuracy of 93.4% with 5.1% rejection rate. For the five-class problem, the accuracy rate of 94.8% is achieved. Experimental results reveal the feasibility and validity of the proposed approach in fingerprint classification.
Yi, Hung Ming, and 洪銘曎. "Blur detection for fingerprint classification based on gradient and SVD." Thesis, 2011. http://ndltd.ncl.edu.tw/handle/39075506475585587790.
Full text國立高雄應用科技大學
光電與通訊研究所
99
Automatic fingerprint classification is nowadays one of the most important and reliable biometric technologies. This is because of the fingerprint distinctiveness, persistence, ease of acquisition, and high matching accuracy rates. However, the performance of classification relies heavily on the quality of the input fingerprint images. Due to various factors such as skin conditions, e.g. dry, wet, cuts, scars, and bruises, non-uniform finger pressure, noise introduced by sensor and inherently poor-quality fingers, e.g. manual workers and elders, a significant percentage of fingerprint images is of poor quality. In fact, a single fingerprint may contain regions with quality of good, medium, and blur. Thus an enhanced method which can mend the ridge structure of a blur region is necessary. In this thesis, we propose an effective algorithm for fingerprint image patch, which can much improve the clarity and continuity of ridge structures based on the novel mutual-lighting SVD compensation with blur region detection. The algorithm consists of two stages. The first stage extracts the blur region using wavelet entropy filtering with region growing. The second stage yields the patched image by doing lighting compensation mutually for the dark and light regions based on the image mean. Experimental results for NIST-4 and FVC databases show that the patched image quality is much better than the other existing methods for improving singular point detection and fingerprint classification.
Wu, Cheng-Jung, and 吳振榮. "Dry and Wet Fingerprint Classification Using Ridge Features for Multiple Resolutions." Thesis, 2017. http://ndltd.ncl.edu.tw/handle/csv265.
Full textLe, Ngoc Tuyen, and 黎玉線. "Image Enhancement Using Adaptive Singular Value Decomposition for Face Recognition and Fingerprint Classification." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/88627310951461650212.
Full text國立高雄應用科技大學
電子工程系碩士班
103
The development of face recognition and fingerprint classification systems in real world still remains a major challenge for the scientific researchers. One of the most crucial reasons affecting the efficiency of these systems is the quality of the input image. To improve the quality of images in pre-processing step, this dissertation applies the useful properties of the singular value decomposition in image processing to improve quality of the color face and fingerprint images. For face recognition, this study proposes three methods to enhance color face images. First, we propose the innovative illumination compensation algorithm, two separated singular value decomposition, in the spatial domain. Second, we introduce an efficient brightness detector for lighting detection and an illumination compensation method, adaptive singular value decomposition in the two-dimensional discrete Fourier domain. Third, we propose a novel illumination compensation method called adaptive singular value decomposition in the 2D discrete wavelet domain. These methods can resolve the illumination variation problem on color face images when there is insufficient light and, at the same time, improve the effective of recognition system. Fingerprint image enhancement is one of the most major steps in an automated fingerprint identification system. In this study, an effective algorithm for fingerprint image quality improvement is proposed. The algorithm consists of two stages. The first stage is decomposing the input fingerprint image into four subbands by applying two-dimensional discrete wavelet transform. At the second stage, the compensated image is produced by adaptively obtaining the compensation coefficient for each subband based on the image content and the referred Gaussian template. The experimental results indicated the efficiency of the proposed method.
Mieloch, Krzysztof. "Hierarchically linked extended features for fingerprint processing." Doctoral thesis, 2008. http://hdl.handle.net/11858/00-1735-0000-0006-B3BC-A.
Full text