Fig 1 - uploaded by Lin Hong
Content may be subject to copyright.
Ridges and automatically detected minutiae points in a fingerprint image. The core is marked with a 2 .
Source publication
With identity fraud in our society reaching unprecedented proportions and with an increasing emphasis on the emerging automatic personal identification applications, biometrics-based verification, especially fingerprint-based identification, is receiving a lot of attention. There are two major shortcomings of the traditional approaches to fingerpri...
Contexts in source publication
Context 1
... fingerprint is the pattern of ridges and valleys on the sur- face of the finger [3]. The uniqueness of a fingerprint can be de- termined by the overall pattern of ridges and valleys as well as the local ridge anomalies [a ridge bifurcation or a ridge ending, called minutiae points (see Fig. 1)]. Although the fingerprints possess the discriminatory information, designing a reliable au- tomatic fingerprint matching algorithm is very challenging (see Fig. 2). As fingerprint sensors are becoming smaller and cheaper [4], automatic identification based on fingerprints is becoming an attractive alternative/complement to the ...
Context 2
... pixel-wise operation which does not change the clarity of the ridge and valley structures. If nor- malization is performed on the entire image, then it cannot com- pensate for the intensity variations in different parts of the image due to the elastic nature of the finger. Separate normalization of each individual sector alleviates this problem. Fig. 10 shows an example of this normalization scheme. For our experiments, we set the values of both and to 100. An even symmetric Gabor filter has the following general form in the spatial ...
Context 3
... capture most of the global ridge directionality information as well as the local ridge characteristics present in a fingerprint. We illus- trate this through reconstructing a fingerprint image by adding together all the eight filtered images. The reconstructed image is very similar to the original image and only been slightly blurred (degraded) [ Fig. 10(a)] due to lack of orthogonality among the filters. At least four directional filters are required to capture the entire global ridge information in a fingerprint [ Fig. 10(k)], but eight directional filters are required to capture the local charac- teristics. So, while four directions are sufficient for classifica- tion [6], eight ...
Context 4
... a fingerprint image by adding together all the eight filtered images. The reconstructed image is very similar to the original image and only been slightly blurred (degraded) [ Fig. 10(a)] due to lack of orthogonality among the filters. At least four directional filters are required to capture the entire global ridge information in a fingerprint [ Fig. 10(k)], but eight directional filters are required to capture the local charac- teristics. So, while four directions are sufficient for classifica- tion [6], eight directions are needed for matching. Our empir- ical results support our claim, we could get better accuracy by using eight directions for matching as compared to only four di- ...
Context 5
... Our empirical results show that using AAD features give slightly better performance than variance features as used in [6]. The 640-dimensional fea- ture vectors (FingerCodes) for fingerprint images of two dif- ferent fingers from the MSU_DBI database are shown as gray level images with eight disks, each disk corresponding to one filtered image in Fig. 11. The gray level in a sector in a disk represents the feature value for that sector in the corresponding filtered image. Note that Fig. 11(c) and (d) appear to be visually similar as are Fig. 11(g) and (h), but the corresponding disks for two different fingers look very ...
Context 6
... fea- ture vectors (FingerCodes) for fingerprint images of two dif- ferent fingers from the MSU_DBI database are shown as gray level images with eight disks, each disk corresponding to one filtered image in Fig. 11. The gray level in a sector in a disk represents the feature value for that sector in the corresponding filtered image. Note that Fig. 11(c) and (d) appear to be visually similar as are Fig. 11(g) and (h), but the corresponding disks for two different fingers look very ...
Context 7
... of two dif- ferent fingers from the MSU_DBI database are shown as gray level images with eight disks, each disk corresponding to one filtered image in Fig. 11. The gray level in a sector in a disk represents the feature value for that sector in the corresponding filtered image. Note that Fig. 11(c) and (d) appear to be visually similar as are Fig. 11(g) and (h), but the corresponding disks for two different fingers look very ...
Context 8
... total of 100 images (approximately 4% of the database) were rejected from the MSU_DBI database because of the fol- lowing reasons: 1) the reference point was located at a corner of the image and therefore an appropriate region of interest could not be established and 2) the quality of the image was poor. See Fig. 12 for examples of images which were rejected. A total of 100 images (approximately 5.6% of the database) were re- jected from the NIST 9 database based on the same criteria. Our quality checker algorithm estimates the dryness of the finger (or smudginess of the fingerprint image) and the extent to which the surface of the finger tip is ...
Context 9
... on the same criteria. Our quality checker algorithm estimates the dryness of the finger (or smudginess of the fingerprint image) and the extent to which the surface of the finger tip is imaged. The estimate of the dry- ness/smudginess is based on the variance of the grayscale in the captured image and a partial fingerprint is detected by tracking Fig. 12. Examples of rejected images: (a) a poor quality image and (b) the reference point is (correctly) detected at a corner of the image and so an appropriate region of interest could not be established. the ridge structure in the image. For algorithm development and parameter selection, an independent database of 250 impres- sions from ten ...
Context 10
... finger did not yield an identical FingerCode because of the rotation and inconsistency in reference point location. For the MSU_DBI database, a total of 6 586 922 matchings were performed. The probability distribution for genuine (correct) matches was estimated with 7472 matches and the imposter distribution was estimated with 6 579 450 matches. Fig. 13(a) shows the two distributions. For the NIST 9 database, a total of 722 419 matchings were performed and the genuine and imposter distributions were estimated with 1640 and 720 779 matching scores, respectively. Fig. 13(b) shows the imposter and genuine distributions for the NIST 9 database. In a biometric system operating in a ...
Context 11
... genuine (correct) matches was estimated with 7472 matches and the imposter distribution was estimated with 6 579 450 matches. Fig. 13(a) shows the two distributions. For the NIST 9 database, a total of 722 419 matchings were performed and the genuine and imposter distributions were estimated with 1640 and 720 779 matching scores, respectively. Fig. 13(b) shows the imposter and genuine distributions for the NIST 9 database. In a biometric system operating in a verification mode, there are four possible outcomes: 1) genuine acceptance; 2) imposter ...
Context 12
... In such a case, the "ideal" ROC curve is a step function at the zero False Acceptance Rate. On the other extreme, if the genuine and imposter distributions are exactly the same, then the ROC is a line segment with a slope of with an end point at zero False Acceptance Rate. In prac- tice, the ROC curve behaves in between these two extremes. Fig. 14(a) and (b) compare the ROC's of a state-of-the-art minutiae-based matcher [1] with our filter-based matcher on the MSU_DBI and the NIST 9 databases, respectively. Since the ROC curve of the minutiae-based matcher is above the filter-based matcher, we conclude that our matcher does not perform as well as the state-of-the-art ...
Context 13
... have combined mul- tiple biometrics (i.e., fingerprint and face) to improve the per- formance of a verification system [29], [30], but this involves the cost of additional sensors and inconvenience to the user in providing multiple cues. Jain et al. [31] have shown that matching accuracy can be improved by combining "indepen- dent" matchers. Fig. 14(a) and (b) show such an improvement in matching accuracy results by using the Neyman-Pearson [32] rule to combine scores obtained from the proposed filter-based and minutiae-based [1] matchers. The Neyman-Pearson rule used for this combination can be summarized as ...
Context 14
... matching scores from the rest of the 84 users. For the NIST 9 database, the training was performed using the scores from the first 450 users and the testing on the scores obtained by gener- ating all the matching scores from the rest of the 450 users. A two-dimensional plot of the scores from the two matchers for the MSU_DBI database is shown in Fig. 15. The performance im- provement resulting from a combination of matchers is shown in Fig. 14(a) and ...
Context 15
... using the scores from the first 450 users and the testing on the scores obtained by gener- ating all the matching scores from the rest of the 450 users. A two-dimensional plot of the scores from the two matchers for the MSU_DBI database is shown in Fig. 15. The performance im- provement resulting from a combination of matchers is shown in Fig. 14(a) and ...
Context 16
... refined. We expect to improve the performance significantly by developing algorithms to over- come the following main shortcomings of our technique: 1) the reference point cannot be located accurately in noisy images and 2) the matching scheme is not able to tolerate large deforma- tion in the ridge pattern due to finger pressure differences (see Fig. 16). About 99% of the total compute time for verification ( 3 s on a SUN ULTRA 10) is taken by the convolution of the input image with eight Gabor ...
Context 17
... fingerprints are stored as templates, the identification ef- fectively involves a "bit" comparison. As a result, the identifi- cation time would be relatively insensitive to the database size. Further, our approach to feature extraction and matching is more amenable to hardware implementation than, say, a string-based fingerprint matcher [1]. Fig. 15. Two-dimensional distribution of genuine and imposter scores from the proposed filter-based and the minutiae-based matchers for the MSU_DBI database. The matching distance obtained from the filter-based method was inverted (100-distance) to obtain a matching score for making the outputs of the two matchers consistent. Fig. 16. Errors ...
Context 18
... matcher [1]. Fig. 15. Two-dimensional distribution of genuine and imposter scores from the proposed filter-based and the minutiae-based matchers for the MSU_DBI database. The matching distance obtained from the filter-based method was inverted (100-distance) to obtain a matching score for making the outputs of the two matchers consistent. Fig. 16. Errors in filter-based matching; fingerprint images from the same finger which do not match: (a) and (b) do not match because of failure of reference location, and (c) and (d) do not match because of change in inter-ridge distance due to finger pressure ...
Context 19
... I NTRODUCTION W ITH and smartcards THE advent and of an electronic increased banking, emphasis e-commerce, on the pri- vacy and security of information stored in various databases, automatic personal identification has become a very important topic. Accurate automatic personal identification is now needed in a wide range of civilian applications involving the use of passports, cellular telephones, automatic teller machines, and driver licenses. Traditional knowledge-based [password or personal identification number (PIN)] and token-based (passport, driver license, and ID card) identifications are prone to fraud because PIN’s may be forgotten or guessed by an imposter and the tokens may be lost or stolen. As an example, Mastercard credit card fraud alone now amounts to more than 450 million U.S. dollars annually [2]. Biometrics , which refers to identifying an individual based on his or her physiological or behavioral characteristics has the capability to reliably distinguish between an authorized person and an imposter. A biometric system can be operated in two modes: 1) verification mode and 2) identification mode. A biometric system operating in the verification mode either accepts or rejects a user’s claimed identity while a biometric system operating in the identification mode establishes the identity of the user without a claimed identity information. In this work, we have focused only on a biometric system operating in the verification mode. Among all the biometrics (e.g., face, fingerprints, hand geometry, iris, retina, signature, voice print, facial thermogram, hand vein, gait, ear, odor, keystroke dynamics, etc. [2]), fingerprint-based identification is one of the most mature and proven technique. A fingerprint is the pattern of ridges and valleys on the surface of the finger [3]. The uniqueness of a fingerprint can be determined by the overall pattern of ridges and valleys as well as the local ridge anomalies [a ridge bifurcation or a ridge ending, called minutiae points (see Fig. 1)]. Although the fingerprints possess the discriminatory information, designing a reliable automatic fingerprint matching algorithm is very challenging (see Fig. 2). As fingerprint sensors are becoming smaller and cheaper [4], automatic identification based on fingerprints is becoming an attractive alternative/complement to the traditional methods of identification. The critical factor in the widespread use of fingerprints is in satisfying the performance (e.g., matching speed and accuracy) requirements of the emerging civilian identification applications. Some of these applications (e.g., fingerprint-based smartcards) will also benefit from a compact representation of a fingerprint. The popular fingerprint representation schemes have evolved from an intuitive system design tailored for fingerprint experts who visually match the fingerprints. These schemes are either based on predominantly local landmarks (e.g., minutiae-based fingerprint matching systems [1], [5]) or exclusively global information (fingerprint classification based on the Henry system [6]–[8]). The minutiae-based automatic identification techniques first locate the minutiae points and then match their relative placement in a given finger and the stored template [1]. A good quality fingerprint contains between 60 and 80 minutiae, but different fingerprints have different number of minutiae. The variable sized minutiae-based representation does not easily lend itself to indexing mechanisms. Further, typical graph-based [9]–[11], and point pattern-based [1], [12], [13] approaches to match minutiae from two fingerprints need to align the unregistered minutiae patterns of different sizes which makes them computationally expensive. Corre- lation-based techniques [14], [15] match the global patterns of ridges and valleys to determine if the ridges align. The global approach to fingerprint representation is typically used for indexing [6]–[8], and does not offer very good individual discrimination. Further, the indexing efficacy of existing global representations is poor due to a small number of categories that can be effectively identified and a highly skewed distribution of the population in each category. The natural proportion of fingerprints belonging to categories whorl (whorl and double loop put together), loop (right and left loop put together), and arch (arch and tented arch put together), is 0.279, 0.655, and 0.066, respectively. Both these approaches utilize representations which cannot be easily extracted from poor quality fingerprints. The smooth flow pattern of ridges and valleys in a fingerprint can be viewed as an oriented texture field [16]. The image intensity surface in an ideal fingerprint image is comprised of ridges whose direction and height vary continuously, which consti- tutes an oriented texture. Most textured images contain a limited range of spatial frequencies, and mutually distinct textures differ significantly in their dominant frequencies [17]–[19]. Textured regions possessing different spatial frequency, orientation, or phase can be easily discriminated by decomposing the texture in several spatial frequency and orientation channels. For typical fingerprint images scanned at 500 dpi, there is a little variation in the spatial frequencies (inter-ridge distances) among different fingerprints. This implies that there is an optimal scale (spatial frequency) for analyzing the fingerprint texture. Every point in a fingerprint image is associated with a dominant local orientation and a local measure of coherence of the flow pattern. A symbolic description of a fingerprint image can be derived by computing the angle and coherence at each point in the image. Fingerprints can be identified by using quantitative measures associated with the flow pattern (oriented texture) as features. It is desirable to explore representation schemes which combine global and local information in a fingerprint. We present a new representation for the fingerprints which yields a relatively short, fixed length code, called FingerCode [6] suitable for matching as well as storage on a smartcard. The matching reduces to finding the Euclidean distance between these FingerCodes and hence the matching is very fast and the representation is amenable to indexing. We utilize both the global flow of ridge and valley structures and the local ridge characteristics to generate a short fixed length code for the fingerprints while maintaining a high recognition accuracy. The proposed scheme of feature extraction tessellates the region of interest of the given fingerprint image with respect to a reference point (Fig. 3). A feature vector is composed of an ordered enumeration of the features extracted from the (local) information contained in each subimage (sector) specified by the tessellation. Thus, the feature elements capture the local information and the ordered enumeration of the tessellation captures the invariant global relationships among ...
Similar publications
We present a fingerprint matching scheme that utilizes a ridge feature map to match fingerprint images. The technique described here obviates the need for extracting minutiae points to match fingerprint images. The proposed scheme uses a set of 16 Gabor filters, whose spatial frequencies correspond to the average inter-ridge spacing in fingerprints...
Citations
... Indeed, due to lodging, wheat stems tend to bend in certain directions, making these directions ideal candidates for extracting textural features using a suitable tool, such as the Gabor filters. Gabor filter has the property that textural features can be effectively extracted from images (Jain et al., 2000). The filter parameters need to be set properly for this purpose. ...
Crop lodging in agricultural fields is one of the major factors that limit cereal crop yields. Wheat, the most popular cereal crop in most countries, is also affected by this phenomenon, which may result in a significant decrease in both yield and quality. Therefore, addressing wheat lodging is crucial for producers. This study aims to detect and identify wheat lodging through aerial images and classify its severity based on ratio, angle and location of the lodging. To achieve this goal, a multi-task approach was proposed involving three phases. First, automatic dataset generation methodology was conducted on orthomosaic imagery of three dates. Next, a comprehensive assessment of wheat lodging (ratio, angle and location) was performed, which has received little research attention. Third, applying and improving selected classification models for classifying image datasets was conducted. Combining convolutional neural networks and temporal sequences in a single model provided an opportunity to use spatiotemporal information extracted from the wheat image datasets. Time dependency and individual dates were both considered in the classification task. The limited number of data and imbalanced classes challenges, resulting from real field conditions data collection, were overcome by applying a new loss function to the classifier models. The overall accuracy of wheat lodging classification reached over 91% in these two states using the proposed approach. Based on this research, wheat lodging was detected more accurately by the proposed models despite the small and imbalanced dataset. The developed methodology paves the way for comprehensive and automatic wheat lodging detection, and the methodology can be adapted for similar crops that suffer lodging issues with suitable modifications.
... Biometric templates are generated from biometric measurements (e.g., a face or fingerprint image). These measurements undergo a sequence of transformations, an extraction of the features (e.g., using Gabor filtering [13,18]) followed eventually by a Scale-then-Round process [2] to accommodate representations better handled by cryptographic schemes, i.e., binary or integer vectors. These templates are then protected either through their mere encryption, or using a Biometric Template Protection (BTP), e.g., a fuzzy commitment scheme [16], a cancelable biometric transformation such as Biohashing [15,17] or any other salting method. ...
... The templates in B result from a chain of treatments, an extraction of the features (e.g., using Gabor filtering [13,18]) followed eventually by a Scale-then-Round process [2] to accommodate better handled representations, usually binary vectors. These feature vectors can also be transformed into a binary form using a tokenized projection-based protection scheme, where tokens may be secret or public. ...
A biometric recognition system can operate in two distinct modes, identification or verification. In the first mode, the system recognizes an individual by searching the enrolled templates of all the users for a match. In the second mode, the system validates a claimed identity by comparing the fresh template with the enrolled template for this identity. Both the experimentally determined false match rate and false non-match rate through recognition threshold adjustment define the recognition accuracy, and hence the security of the system. The biometric transformation schemes usually produce binary templates that are better handled by cryptographic schemes. One of the requirements for these transformation schemes is their irreversibility. In this work, we rely on probabilistic modelling to quantify the security strength of binary templates. We investigate the influence of template size, database size and threshold on the probability of having a near-collision, and we highlight two attacks on biometric systems. We discuss the choice of parameters through the generic presented attacks.
... This portion is the high frequency core area of the fingerprint. To separate out the high (discriminative) frequency portion, the enhancement [28] and ROI extraction [29] steps are elaborated below. ...
... With the addition of the ridge dictionary, fingerprint orientation gets corrected to a great extent. Further, the fingerprint image with corrected ridge orientation is smoothed using a low-pass filter (Jain et al., 2000) in which the image is initially converted to a continuous vector field as depicted by Equations (2)-(3). ...
Fingerprint based human identification is one of the authentic biometric recognition systems due to the permanence and uniqueness of the finger impressions. There is the extensive usage of fingerprint recognition in personalized electronic devices, security systems, banking, forensic labs, and especially in law enforcement agencies. Although the existing systems can recognize fingerprints, they lack in case of poor quality and latent fingerprints. The latent fingerprints are captured by law enforcement agencies during the crime scene to find the criminal. Consequently, it is essential to develop a novel system that can efficiently recognize both complete and latent fingerprints. The current work proposes an efficient Gravitational Search Decision Forest (GSDF) method, which is a combination of the gravitational search algorithm (GSA) and the random forest (RF) method. In the proposed GSDF approach, the mass agent of GSA determines the solution by constructing decision trees in accordance with the random forest hypothesis. The recognition of the fingerprints is accomplished by mass agents in the form of a final generated decision forest from the set of hypothesis space as the mass agents can create multiple hypotheses using random proportional rules. The experiments for fingerprint recognition are conducted for both the latent fingerprints (NIST SD27 dataset) and the complete fingerprints (FVC2004 dataset). The effectiveness of the proposed GSDF approach is analyzed by evaluating the results with machine learning classifiers (random forest, decision tree, back propagation neural networks, and k-nearest neighbor) as well. The comparative analysis of the proposed approach and incorporated machine learning classifiers indicates the outperformed performance of the proposed approach.
... Fingerprint identification and verification are one of the most significant and reliable identification methods. It is virtually impossible that two people have the same fingerprint, having a probability 1 / 1.9 X 1015 (Hong et. al., 2000). In fingerprint identification and verification applications world-wide, a large volume of fingerprints was collected and stored for a wide range of applications, including forensics, civilian, commercial and lawenforcement applications. Automatic identification of humans based on fingerprints requires the input fingerprint to be matche ...
Humans have distinctive and unique traits which can be used to distinguish them thus, acting as a form of identification. Biometrics identifies people by measuring some aspect of individual’s anatomy or physiology such as hand geometry or fingerprint which consists of a pattern of interleaved ridges and valleys. The aim of this research is to analyse humans fingerprint texture in order to determine their gender, and correlation of RTVTR and Ridge Count on gender detection. The study is to analyze the effectiveness of physical biometrics (thumbprint) in order to determine gender in humans. Humans have distinctive and unique traits which can be used to distinguish them thus, acting as a form of identification. Biometrics identify people by measuring some aspect of individual’s anatomy or physiology such as hand geometry or fingerprint which consists of a pattern of interleaved ridges and valleys. This work developed a system that determines human gender using fingerprint analysis trained with SVM+CNN (for gender classification). To build an accurate fingerprint based model for gender detection system using fingerprint pattern analysis, there are certain steps that must be taken, which include; Data collection (in conducting research, the first step is collecting data in the form of a set of fingerprint image), Pre-processing Data (before entering the training data, pre-processing data is performed, which is resize the fingerprint image 96x96 pixels). Training Data (in this processing the dataset will be trained using the Convolutional neural network and Support vector machine methodology. This training data processing is a stage where CNN + SVM are trained to obtained high accuracy from the classification conducted). Result Verification (after doing all the above process, at this stage, we will display the results of gender prediction based on fingerprint images in the application that has been making). SOCOFing database is made up of 6,000 fingerprint images from 600 African subjects. It contains unique attributes such as labels for gender, hand and finger name as well as synthetically altered versions with three different levels of alteration for obliteration, central rotation, and z-cut. The values for accuracy, sensitivity and precision using the CNN classifier at threshold 0.25 were 96%, 97.8% and 96.92% respectively. At threshold 0.45 the values were 96.3%, 97.6% and 97.6% respectively. At threshold 0.75 the values were 96.5%, 97.3% and 97.9% respectively. In case of the SVM classifier, at threshold 0.25 were 94.3%, 96.6% and 95.8% respectively. At threshold 0.45 the values were 94.5%, 96.4% and 96.2% respectively. At threshold 0.75 the values were 94.8%, 97.3% and 96.8% respectively. From the 600 fingerprints classified, it was observed that a total of 450 fingerprints were detected for male and 150 for female. Results were obtained for gender accuracy, sensitivity and precision through several thresholds to compare the two classifiers. However, it should be verified that the results obtained showed that the CNN classification yielded better accuracy, sensitivity and precision than SVM.
... Bio-Hashing technique based on biometric data salting is used for different biometric paradigms, a few examples of which are presented in [7,39]. Bio-Hashing has been performed on the FingerCode [20] with respect to each minutia to secure the original minutiae information in [7]. In [39], Teoh et al. have devised a two-factor feature transformation technique. ...
Authentication systems have evolved immensely ever since the use of biometrics in the field of security has started. Nonetheless, extensive usage of biometric systems has also resulted in growing fear of identity theft as storing biometric user templates in databases opens up severe security concerns. Consequently, the scope of biometrics invariably depends on the ability of the system to manifest security and robustness against biometric identity theft along with achieving acceptable recognition performance. This paper proposes a user template protection technique to secure a fingerprint based user template used in a biometric authentication system. It is based on the fingerprint shell, which computes alignment-free, singular point independent, and non-invertible user templates. The secure user template generated by the proposed technique comprises fingerprint shell curves computed using the transformed pair-polar structure of the minutia points. The proposed technique has been evaluated on five fingerprint databases of Fingerprint Verification Competitions, namely, FVC2002, FVC2004, and FVC2006. The effectiveness of the technique is analyzed in terms of revocability, diversity, security, and recognition performance. The comparison of results with that of the existing techniques demonstrates the robustness and efficacy of the proposed technique.
... The impression of the pattern of the ridges and valleys on the surface of the fingertip represents the fingerprint, and the combination of the minutiae points, which determines its uniqueness [10]. Due to their uniqueness, fingerprint images are usually utilized for the purpose of the user's authentication, thus making their protection a significant issue. ...
This article presents a recent blind and robust fingerprint image watermarking scheme based on a two-dimensional discrete cosine transform (2D-DCT). The main focus is to compress the fingerprint image watermarked data for the purpose of reducing the volume of storage or sending over the network. The fingerprint features might be affected by the embedded watermark, compression of fingerprint images and the sending across network, thereby leading to various sets of features or watermark data. In order to address this goal in a differential way, the watermark sequence bit two sub-vectors were utilized. The two sub-vectors were achieved by the two-dimensional discrete cosine transform of the host image. Throughout the extraction stage, the essential distinction between the corresponding sub-vectors of the watermarked fingerprint image resulted explicitly in an embedded watermark sequence. The advantage of the proposed method is that it can develop a new simple blind and robust watermarking scheme by 2D-DCT frequency domain on the whole image. Accomplished results relative to other reliable compression schemes showed that the proposed scheme has greater or equivalent robustness to common image processing and geometric attacks, such as cropping, resizing, and rotation. To extract watermark data, the initial fingerprint image was not necessary. The proposed study was tested using 80 fingerprint images from 10 persons, for each from CASIA-FingerprintV5 and FVC2002 fingerprint databases. Eight fingerprint images for each individual were set as the format at which the watermark was embedded in each one.
... Biometric templates are generated from biometric measurements (e.g., a fingerprint image). They result from a chain of treatments, an extraction of the features (e.g., using Gabor filtering (Manjunath and Ma, 1996;Jain et al., 2000)) followed eventually by a Scale-then-Round process (Ali et al., 2020) to accommodate better handled representations, i.e., binary or integervalued vectors. These templates are then protected either through their mere encryption, or using a Biometric Template Protection (BTP), e.g., a cancelable biometric transformation such as Biohashing (Jin et al., 2004;Lumini and Nanni, 2007) or any other salting method. ...
Biometric recognition encompasses two operating modes. The first one is biometric identification which consists in determining the identity of an individual based on her biometrics and requires browsing the entire database (i.e., a 1:N search). The other one is biometric authentication which corresponds to verifying claimed biometrics of an individual (i.e., a 1:1 search) to authenticate her, or grant her access to some services. The matching process is based on the similarities between a fresh and an enrolled biometric template. Considering the case of binary templates, we investigate how a highly populated database yields near-collisions, impacting the security of both the operating modes. Insight into the security of binary templates is given by establishing a lower bound on the size of templates and an upper bound on the size of a template database depending on security parameters. We provide efficient algorithms for partitioning a leaked template database in order to improve the generation of a master-template-set that can impersonates any enrolled user and possibly some future users. Practical impacts of proposed algorithms are finally emphasized with experimental studies.
... There are systems in which multiple feature extraction and matching techniques are used depending on the quality of the acquired fingerprint [3]. Techniques are available which are good at extracting features and matching the poor quality fingerprints [12][13][14]. Enhancement algorithms are also employed if there is a scope of improving the quality of the fingerprint. The enhancement techniques generally try to stitch the broken ridges, balances contrast level, etc. ...
Fingerprint image quality ensures that only the high-quality fingerprints containing a good amount of features are used for verification. Fingerprint matching accuracy depends heavily on the quality of the fingerprints. In this paper, we have proposed a new fingerprint image quality method based on the Directional Filter Banks (DFB). The fingerprint image is decomposed into subbands using the DFB. The similarity between the different subbands is used to calculate the fingerprint image quality. We have compared the performance of the proposed method with the ten existing fingerprint quality estimation methods with special emphasis on widely used fingerprint quality metrics NFIQ and NFIQ 2.0. The experimental results on the top three high-quality fingerprints in the FVC 2004 DB2A dataset show an equal error rate (EER) of 5.59% for the proposed method as compared with 5.74% and 7.95% for NFIQ 2.0 and NFIQ, respectively. Ranked EER, ROC, and error-reject curves show that DFB-based method is a good predictor of matching performance, and it outperforms the existing fingerprint quality methods. We have also analyzed the effect of the partial fingerprints on the fingerprint recognition system using the FVC 2004 DB1A dataset. The presence of partial fingerprints has an adverse effect on the recognition system. The equal error rate increases from 2.34 to 13.93% in partial fingerprint recognition. The proposed DFB-based method rightly assigns low-quality values to partial fingerprints.
... This classification will reduce search space in large database [2]. It can be used as a reference point to register fingerprint images and participate in minutiae matching [3,4]. ...
Detecting the singular point accurately and efficiently is one of the most important tasks for fingerprint recognition. In recent years, deep learning has been gradually used in the fingerprint singular point detection. However, the existing deep learning-based singular point detection methods are either two-stage or multi-stage, which makes them time-consuming. More importantly, their detection accuracy is yet unsatisfactory, especially for the low-quality fingerprint. In this paper, we make a Real One-Stage Effort to detect fingerprint singular points more accurately and efficiently, and therefore, we name the proposed algorithm ROSE for short, in which the multi-scale spatial attention, the Gaussian heatmap and the variant of focal loss are integrated together to achieve a higher detection rate. Experimental results on the datasets FVC2002 DB1 and NIST SD4 show that our ROSE outperforms the state-of-the-art algorithms in terms of detection rate, false alarm rate and detection speed.
