The study by [11] implemented a fingerprint classification system using a fuzzy neural network classifier and its output in the recognition method. 11501157, 1999. 37, no. Figure 15 shows a sample of fingerprints of the dataset. 6, 2017. The main goal of the GoogleNet architecture was to get high accuracy so that the motivation for the GoogleNet is creating Inception CNN module to make a deeper CNN by adding 22 layers and almost 12 times fewer parameters than AlexNet. In the images below youll get a feel of what many of the points look like. System 4 clarifies the mathematical model for the composite whorl with the S core pattern, which is shown in Figure 8; see [4]. [18], they proposed a new partial fingerprint matching for all sensors in mobile devices using minutiae and ridge-form features (RSFs). 2011. The first experiments are performed by applying ResNet-18 to the proposed model. W. Wang, Y. Yang, X. Wang, W. Wang, and J. Li, Development of convolutional neural network and its application in image classification: a survey, Optical Engineering, vol. To get an acceptable and accurate image, the problems such as images with scars, too dry or too moist fingers, or incorrect pressure must be overcome. A fingerprint is a distinct pattern of ridges and valleys on the finger surface of an individual. [18]. The preprocessing stage has two functions: ridge enhancement restoration and segmentation of fingerprint images. Influence of Skin Diseases on Fingerprint Recognition. 22332243, 2004. An ulnar loop enters and exits near the pinky (or the ulna) where the radial loop enters and exits toward the thumb (or the radius). end. Finally, Section 6 concludes the paper. In order to address the limitations of existing contact-based fingerprint identification systems, improve recognition accuracy, and reduce time analysis function, a fingerprint database with categories and prepossessing using neural network classification technique has attracted growing attention in order to improve accuracy and reduce the time for fingerprint classification and matching. Ultimately, if the image does not belong to the previous systems, then the image belongs to the form of loops pattern. The ridge ending is, as indicated by the name, the spot where a ridge ends. Bush Les. endstream endobj startxref Creative Commons Attribution-NonCommercial 4.0 International License. System 1 shows the mathematical model for the concentric whorl pattern, which is illustrated in Figure 5; see [4]. 2005 Jul 7;235(1):71-83. Im capitalizing for clarity purposes only. Human Fingerprints a combination of nerves and skin. A pattern type in which the friction ridges would appear to enter on one side of theimpressionand flow or tend to flow out the other side with a rise or wave in the centre. When a ridge forks or diverges into branch ridges, it is called ridge bifurcation. It presented the latest definition of the initiation block in CNN, where it integrates multiscale convolutionary transformations through the notion of separating, transforming, and merging. hb```T:``f`s s@\vgnIT\:9gss)seGfW\5\KrfYU=[n/\r{j v}c_\a{Z=uwwV7 qV/VTb@U"/ 17, pp. G. T. Candela, P. Grother, C. Watson, R. A. Wilkinson, and C. Wilson, Pcasys- a Pattern-Level Classification Automation System for Fingerprints Nist, NIST, Maryland, USA, 1995. Hj1E{Rg.C If we draw a line between the two Deltas, at least one ridge that stems from the Pattern Area should be cut by the line. A picture paints a thousand words. Arch patterns can be further categorized into two subgroups: . The testing accuracy rises to 81.25% in iteration 96. Preprocessing stage aims to improve and enhance the quality of the image. To find a match-out fingerprint of such a large dataset, a correct classification of the images is required. 8, pp. System 6 shows the mathematical model for the tented arch pattern, which is shown in Figure 10; see [35]. Kasey Wertheim. And youre so right about Delta!!! Feedspot and Expertido.org named her Murder Blog as Best 100 Crime Blogs on the Net. She also blogs at the Kill Zone (Writer's Digest "101 Best Websites for Writers") and Writers Helping Writers. J.-H. Hong, J.-K. Min, U.-K. Cho, and S.-B. These approaches are categorized based on rule-based, structural-based, frequency-based, and syntactic [6]. 2D grayscale to 3D color: the main idea to convert from 2D grayscale to 3D color is that the CNNs just accept the image with 3D color so that the following MATLAB function is used for converting: AugmentedTrainingSet=augmentedImageDatastore(imageSize, trainingSet, ColorPreprocessing, gray2rgb); augmentedTestSet=augmentedImage-Datastore (imageSize, testSet, ColorPreprocessing, gray2rgb); The separation of the object and background is known as binarization. Moreover, Hong and Wang et al. Aw, thank you for making my day, Nicholas. Fingerprint Examiners look at the differences between ridges, number of minutiae, and location on the impression. In this method, to achieve better outcomes, this strategy must reduce the dependence of ridge points on minutiae. ":I^NgO5pCZN4}YeH uyXM] $WN=8(P#kLq-=oZeN8K|%$^HI_O_Tszh)M]KLBadH. Algorithm 1 for the proposed model is given in the following box. The time required to build a model for training and testing for all iterations is 28 seconds. This minutia illustrates an abrupt end to a continuous ridge path. Cho, Fingerprint classification using one-vs-all support vector machines dynamically ordered with nave Bayes classifiers, Pattern Recognition, vol. You can unsubscribe anytime. Hence, the input fingerprint images need to be preprocessed. In today's business world, managed services are more critical than ever. 1 Department of Computer Science, . Y. Lecun, L. D. Jackel, B. Leon, C. Cartes, J. S. Denker, and H. Drucker, Learning algorithms for classification: a comparison on handwritten digit recognition, Neural Networks: the Statistical Mechanics Perspective, World Scientific, Singapore, pp. 21, pp. Latent fingerprints can be made visible by dusting with powder or via chemical reaction. Binarization and thinning model has been used in order to improve image quality. Results in primary and secondary ridge formation with anastomoses. That said, I do NOT collect emails from the comment section. CNN is considered one of the best strategies for learning image content and shows state-of-the-art results related to image recognition, segmentation, detection, and retrieval-related tasks [4244]. FVC2004 was also revealed via mailing lists and online magazines affiliated with biometrics. 397406, 2002. The Galton classification scheme is shown in Figure 1. A pattern type in which the friction ridges enter on one side of the impression, and flow or tend to flow out the other side with a rise or wave in the centre. 18, no. j. Q=fprintf ('it is a bifurcation'); bifurcation=bifurcation+1. The data used to support the findings of this paper are available online and FVC2004 free Database fingerprint dataset. A. K. Jain, H. Lin Hong, S. Pankanti, and R. Bolle, An identity-authentication system using fingerprints, Proceedings of the IEEE, vol. Your blog is properly stuffed with information about fingerprints. Fingerprint Ridge Patterns and Characteristics. A fingerprint image has a lot of redundant information when it is captured. While Wang et al. The proposed model was implemented and tested using MATLAB based on the FVC2004 dataset. What is a bifurcation fingerprint? The preprocess for fingerprint images can be done using some processes such as image enhancement, normalization, filtering, noise reduction, binarization, and thinning [36]. Friction ridges are also found on our palms, feet, and toes. Advertise with TechnologyAdvice on Webopedia and our other IT-focused platforms. Where the ridge of the fingerprint stops. In this paper, we have implemented and tested the proposed model with three CNN architectures, namely, GoogleNet, AlexNet, and ResNet. AlexNet [49] is considered as the first deep CNN architecture, which showed groundbreaking results for image classification and recognition tasks. Its a fascinating field. Fingerprint formation. No. The type of fingerprint on the left is a loop. In this paper, we propose a classification and matching fingerprint model, and the classification classifies fingerprints into three main categories (arch, loop, and whorl) based on a pattern mathematical model using GoogleNet, AlexNet, and ResNet Convolutional Neural Network (CNN) architecture and matching techniques based on bifurcation . Minutia and patterns are very important in the analysis of fingerprints. She's appeared on the Emmy award-winning true crime series, Storm of Suspicion, and three episodes of A Time to Kill on DiscoveryID (airs in 2023). Distinctive Ridge Point (DRP), a recent fingerprint feature, has been developed [30], along with an enhancement triangle-based representation that includes minutiae. Nature vs. Nurture: What Drives People to Kill? Therefore, this research categorized the dataset into three main fingerprint patterns (arches, loops, and whorls) that make CNN works efficiently. 1196, no. One little tidbit is that usually every print from the scene is referred to in the biz as a latent. The data have trained with various convolutional layers to find the best architecture of CNN. Most of the fingerprints datasets have a large size. Several techniques for fingerprint matching have been proposed. VF e@QLDz@1!P& BwO>3=b\@:eW fN@:04@ You wont want to miss this one! By default, imbinarize uses Otsus method, which chooses the threshold value to minimize the intraclass variance of the threshold black and white pixel [37]. Fingerprint matching techniques are divided into three main types: Input: Introduce fingerprint Image. The performance evaluation of the proposed model is done under three different CNN architectures, ResNet, AlexNet, and GoogleNet. It is observable that when the accuracy is high, the loss is low. There are two main theories that aim to explain the processes underlying friction ridge formation. Short ridges (or dot) are ridges which are significantly shorter than the average ridge length on the fingerprint. To complete the categorization method on the dataset for the three main patterns in fingerprint (whorls, arches, and loops), we have to state the three mathematical models for arch patterns which are plain, tented, and strong arch; see [55]. Youre so welcome, Paula. There are several types of minutiae found in each print. A. Khan, A. Sohail, U. Zahoora, and A. S. Saeed, A survey of the recent architectures of deep convolutional neural networks, Artificial Intelligence Review, vol. But then, their fingerprint would be even more distinguishable because of the scar. 13, no. These factors determine the points of identification. A free database biometric authentication system is presented, with a tamper-resistant smartcard serving as the storage device. 10971105, 2017. Imbinarize uses a 256-bin image histogram to compute Otsus threshold. The core is the approximate centre of the fingerprint impression, where the friction ridges converge. They are at http://bias.csr.unibo.it/fvc2004/ website [6]. Any cookies on my website are used to ensure normal website functions (for instance, Youtube videos wont work without their own identifiers). In the proposed architecture, preprocessing consists of three phases: 2D grayscale to 3D color, binarization, and thinning. N. K. Ratha, K. Karu, S. Chen, and A. K. Jain, A real-time matching system for large fingerprint databases, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. The Pattern Area contains the Core, Delta(s), if any, and Ridges. Excellent and bang-on material here. The applied imbinarize function (I) using MATLAB generates a binary image from 2D grayscale or 3D color image by replacing all values above a globally determined threshold with 1s and setting all other values to 0s. This looks like a pitch-fork and branches off into 3 ridge lines. Primary, major, secondary, sub secondary, final, and key classification are the six main classifications. Although this is a process with defined steps, when making a . Garry Rodgers recently postedCOURTROOM COMEDY GREAT LINES FROM GOOFY LAWYERS. Figure 18 shows accuracy and loss for AlexNet-8, showing training and testing accuracy and training and testing loss with respect to iteration. B. Herbert, A. Ess, T. Tuytelaars, and L. Van Gool, Speeded- up robust features (surf), Computer Vision and Image Understanding, vol. Im so glad you found it useful! A ridge bifurcation is defined as the point where a ridge forks or diverges into branch ridges. Mapping the volar pad precedes ridge formation. Most of the fingerprint extraction and matching. 66% of the population have Loops. This can affect analysis by modern-day biometric pattern recognition systems. Collectively, these features are called minutiae. GoogleNet-22 architecture classifier results. In simpler terms, a Delta is where the ridges form a triangular-shape. Recently, a receiver operating characteristic (ROC) [31] curve model was suggested that used a weighted empirical approach to account for both the order constraint and the within-cluster correlation structure. And as for If a fingerprint has no Delta, well, at least it will get there on time. One friction ridge branches off and divides into two friction ridges. endstream endobj 809 0 obj <>stream This is what makes them unique. Figure 19 obtained results from the experiment. end. Therefore, fingerprints have been confirmed to be good and secure biometrics. Cant wait to read about your internet woes in your hilarious style, Garry! Tented arch They consist of friction ridge units that can be characterised by patterns (arrangement of ridges) and minutiae (small specific details such as bifurcations and ridge endings). A determination made by the examiner that there is sufficient agreement between the friction ridge patterns of a known and unknown fingerprint to conclude that both fingerprints may have originated from the same source. D. Peralta, S. Garca, J. M. Benitez, and F. Herrera, Minutiae-based fingerprint matching decomposition: methodology for big data frameworks, Information Sciences, vol. 21, pp. With AlexNet, feature extraction stages have been expanded from 5 (LeNet) to 8 to render CNN accessible for different image categories [47]. J Theor Biol. These details are used to characterize the fingerprint in question and forinclusionandexclusionpurposes. On the other hand, the pattern of the spiral has three equilibrium points, the origin is spiral out, and the points (1, 0) and (1, 0) are cusps. hbbd``b`> The pattern is the unique characteristics of the ridges and valleys that make up the print. Id like to see something like two examiners must confirm match rather than one. Examples of fingerprint patterns Loop Tented arch Whorl Double Loop Examples of minutiae Ridge ending Dot An uninterrupted friction ridge that continues until there is a ridge ending. This technique is able to train a NN with 152 layers while still having lower complexity than VGGNet [54]. Individualistic details of various levels that are observed within a fingerprint and are used to individually characterize the fingerprint in question and be used for inclusion and exclusion purposes. 4, 2011. Second, the ResNet in this research had 18 layers, so it is between AlexNet and GoogleNet in the accuracy and time computation. Biometric based access control is becoming increasingly popular in the current era because of its simplicity and user-friendliness. Fingerprints are the pattern of the epidermal ridges on fingers. You beat me to the topic and did a great job. 55, no. Patent is rarely used as a scene term even though its technically different. Obviously I cant list all 150 points here, but here are the most common. The proposed model implementation runs using a MATLAB environment with a PC containing 4GB of RAM and 4 Intel cores i5 (2.0GHz each). I hope this helps you to write a more convincing scene. In the verification stage, the decision is determined based on the percentages or matching results of similarity. The point at, or in front of, and nearest the center of, the divergence of the Type Line. From the obtained results, GoogleNet provides better results in terms of accuracy. More recently neural network fingerprint classification method is [16] where the proposed method is retrained over AlexNet, GoogleNet, and ResNet with an average precision of 95.55%, 92.51, and 94, 88 respectively. Its a lot harder than it looks. Here a loop like peak is seen, but the recurve does not occur resulting in a steep arch. Please read full Privacy Policy. A good quality fingerprint . I. Jawarneh and N. Alsharman, The mathematical model and deep learning features selection for whorl fingerprint classifications, InterNational Journal of Computational Intelligence Systems, vol. end. Journal of Biomedicine and Biotechnology Volume 2012, Article ID 626148, 14 pages doi:10.1155/2012/626148. Therefore, leaving a comment is considered a clear affirmative, specific, and unambiguous action as defined by the GDPR. In addition, the image has connection orbits between the origin and the other two points on the left and right sides. In biometrics and fingerprint scanning, bifurcation refers to the point in a fingerprint where a ridge divides to form two ridges. %PDF-1.5 % This pattern is characterized by ridges that enter on one side of the print, go up, and exit on the opposite side. The bottom right fingerprint is a arch. These skip connections are also known as gated units or gated recurrent units and have a clear resemblance to recent effective elements introduced in RNNs [53]. ML algorithms are a field of Artificial Intelligence (AI) that provides computers with intelligence by studying the underlying relationships between the data and making decisions without explicit programming. 198212, 2017. [13] proposed a combination of the SVM and the naive Bayes to classify the fingerprints based on the number of fingerprint core and delta points. Friction ridges are also found on our palms, feet, and toes. XHaRgm]bG)5j9sziY40CLCauEeLD}LW|C"y(") #yW% gYdnV:b$RYE^47rd62wE KqJr9b#b=['L3BGWTw5!s>RuA? Parameters setting for CNN architecture classifier. 110, no. Minutiae is extracted from the thinning image, and matching is conducted in a specific part in the database that is determined during classify input image stage to reduce time complexity function. 209225, 2018. 85, no. The Fingerprint Examiner will first look at the pattern type (loops, arches, or whorls). Its marked Delta and underlined in green. This minutia is an example of a second-level fingerprint detail. The ulna bone is on the pinky side of the lower arm, and the radius bone is on the thumb side. To gain from CNNs representational ability, AlexNet was simultaneously trained on two NVIDIA GTX 580 GPUs to address hardware deficiencies. One of the . 402421, 1999. S. Christian, W. Liu, Y. Jia et al., Going deeper with convolutions, in Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, June 2015, https://doi.org/10.1109/CVPR.2015.7298594. Arches are the least common type of fingerprint, occurring only about 5% of the time. Learn more about ridge characteristics in the following slides. Since the late 1990s, several and numerous ML algorithms have been implemented to mimic sensory human responses such as speech and vision but have generally failed to achieve satisfaction at the human level [39, 40]. What catches their eye first? Many departments require a 12 point match, but theres no country-wide rule. Embarrassing doesnt come close to the right word. Id heard of loops, arches and whorls before, and latent prints, but I had no idea all of that was just the tip of the iceberg. Arches can be subdivided into two categories: tented arch and plain arch For a fingerprint to be considered an arch, the pattern must have ridges entering one side and exiting the other side. Clinical relevance of loss of fingerprints. Language of Forensics: Fingerprints by Vivienne Luk is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License, except where otherwise noted. SVM is an ML algorithm that adopts a robust approach to fingerprint classification. The overall validation accuracy is 75%. K. Castillo-Rosado and J. Hernndez-Palancar, Latent fingerprint matching using distinctive ridge points, Informatica, vol. The classification scheme is based on the extraction of the fingerprint feature, which involves encoding the singular points along with their relative positions and directions from a fingerprint image of the binaries. Simulation of the concentric whorl fingerprint: (a) phase portrait of the model and (b) image of the concentric whorl fingerprint. Ridges divdes into two ridges then divides one again. To the best of our knowledge, this is the first such attempt to tackle complex fingerprint classification issues using CNN. "9Ph"uipX"tcF?M&E9/qmV},I.J`r8(z,P*78@+6. Y. Yao, G. Marcialis, M. Pontil, P. Frasconi, and F. Roli, Combining flat and structured representations for fingerprint classification with recursive neural networks and support vector machines, Pattern Recognition, vol. E. Erwin, N. N. B. Karo, A. Y. Sari, and N. Aziza, The enhancement of fingerprint images using gabor filter, Journal of Physics: Conference Series, vol. Furthermore, the fingerprint could be utilized to solve one of the most difficult problems in the system and network security: user authentication. So when I got homeIdelved deeper intofingerprints in preparation of this post. Also, no deltas are present. Learn more about minutiaepoints, fingerprint, biometrics Image Processing Toolbox 805 0 obj <> endobj The embedded access points for trusted data and resources access in HPC systems [58] are one of interesting related work that discussed one possible solution for user authentication in network security and it is a hardware implementation in hight-performance computing field. Its access point prototype, which was created with FPGA technology, a smartcard read/write device, and the AES algorithm to encrypt the biometric template, yielded intriguing results in terms of recognition rates. Gastrulation early single ball of cells reorganises into three, Endoderm forms other major organ systems including gastrointestinal tract, respiratory tract, urinary tract. The fourth observation is to ensure the characteristics are in the same relative position. As a result, various research studies employing machine learning to classify fingerprints have been done. Accidental Whorl (found in only 0.01 % of the population) is comprised of two Deltas. Criminal courts require 8 to 12 minutiae to match. In the stage of feature extraction, the block of the relevant information is extracted that will be applied for identification with the template fingerprint. From the results in Table 2, we can observe that training accuracy rises to 100% in iterations 87, 90, and 102. 408, pp. Webopedia resources cover technology definitions, educational guides, and software reviews that are accessible to all researchers regardless of technical background. System 7 represents the mathematical model for the strong arch pattern, which is explained in Figure 11; see [35]. Where two ridges of the fingerprint cross, forming an X patern. Edward Henry classification of fingerprints classes. <3. Double Loop. Another brilliant post! I had no idea, either. Collectively they are called minutiae. Sue lives with her husband in the Lakes Region of New Hampshire and writes two psychological thriller series, Mayhem Series and Grafton County Series (Tirgearr Publishing) and true crime/narrative nonfiction (Rowman & Littlefield Group). Section 5 presents results and discussion. end. If you use this information in your book, these words are not normally capitalized. 289307, 2022. Volar pads less prominent then disappear; friction ridge units develop, Volar skin is layered epidermis on top of shapeless fibrous dermis, By end of second trimester, sweat ducts and pores appear along epidermal ridges. There have been numerous algorithms of fingerprint classification developed [12, 1926]. 3, pp. Whorl These datasets are commonly used as benchmarks for evaluating fingerprint matchers in the context of fingerprint verification. Sir Francis Galtonwas the first person to classify fingerprints into different types based on the three basic features: Loops, Arches, and Whorls. Looking forward for more post from you. 201218, 2020. The creation of four new databases was performed using three commercially available scanners and a synthetic fingerprint generator [57]. R. Cappelli, A. Lumini, D. Maio, and D. Maltoni, Fingerprint classification by directional image partitioning, Pattern Analysis and Machine Intelligence, IEEE Transactions on, vol. I missed this post, Sue. The testing accuracy rises to 100% in iterations 60, 63, 66, 75, 78, 90, 93, 96, 99, 102, 105, and 108. . FUZZ-IEEE02. 6, pp. Determination made by the examiner that there is sufficient reliable detail of characteristics in disagreement that one can conclude that two fingerprints (friction ridge impressions) did not originate from the same source. A. Nagar, K. Nandakumar, and A. K. Jain, Securing fingerprint template: fuzzy vault with minutiae descriptors, in Proceedings of the 2008 International Conference for Pattern Recognition, Tampa, USA, December 2008. As a result, the additional time complexity is required for statistical assessment of performance fingerprint matching data. A. Krizhevsky, I. Sutskever, and E. Hinton Geoffrey, Imagenet classification with deep convolutional neural networks, Advances in Neural Information Processing Systems, vol. F. Galton, Finger Prints, McMillan & Co., London, UK, 1892. And in Germany, the standard is 12 points. Theyre looking for the most obvious points. Pseudosingularity points will be discovered and extracted in the circumstances mentioned above, allowing for fingerprint categorization and matching [27]. 3, no. Simulation of the upper right-lower left (UR-LL) spiral whorl fingerprint: (a) phase portrait of the model and (b) image of the upper right-lower left (UR-LL) spiral whorl fingerprint. There are three different types of whorls: plain whorl central pocket whorl double loop whorl and accidental whorl Each of these is different in its own way, but each type has two deltas. Copyright 2022 Nesreen Alsharman et al. A ridge is defined to be a single curved segment whereas a valley is the area between two adjacent ridges. This is important because the Delta(s) determine how a fingerprint is classified. AlexNet was proposed by Lee et al. Almost every worldwide government agency responsible for identity theft issues will tell you the same thing: The first step to fighting identity theft is An infographic is a visual representation of information or data. S. Minaee, E. Azimi, and A. Minutiae Extraction from Fingerprint Images - a Review Roli Bansal1, Priti Sehgal2 and Punam Bedi3. The usage of AlexNet [47, 48] since the beginning of deep CNNs was limited to hand digit recognition tasks and did not scale well to all classes of images. In another study conducted by Lee et al. 206214, 2013. 2022 TechnologyAdvice. Combining these characteristics is a common occurrence. A line that connects two other lines together. However permanent scarring can occur with deep lesions. They consist of friction ridge units that can be characterised by patterns (arrangement of ridges) and minutiae (small specific details such as bifurcations and ridge endings). S. Shah and P. Sastry, Fingerprint classification using a feedback-based line detector, IEEE Transactions on Systems, Man, and Cybernetics, Part B, vol. I have this code but in this it only shows limited no. Can Graphologists ID Serial Killers' Handwriting? Fingerprints. Within each type of fingerprint there are different types of arches, loops, whorls. Look at the picture below and identify the different minutiae. The number of minutiae that are required to match between two prints varies. This minutia is an example of a second-level fingerprint detail. It is observable that when the accuracy is high, the loss is low. hand (Olsen, Robert D. Fingerprint Mechanics 1977) Delta The delta is that point on a ridge at or in front of and nearest the center of the divergence of the type lines. Novel hardware implementation in the field of high-performance computing is presented in [58] that could be integrated with the proposed method to produce a more effective and efficient algorithm. 431454, 2019. The ridges are established upon a grid of sensory nervous tissue. K. Karu and A. K. Jain, Fingerprint classification, Pattern Recognition, vol. Small space in the middle of two lines. While the second-sub secondary, WCDX and special loop division are three extensions.. More precisely, these six divisions with extensions are referred to as FBI modified Henry fingerprint . Note: Flat or plain impressions give a truer reading of how the ridges appear. The two most prominent local ridge characteristics are ridge ending and ridge bifurcation. If you continue to use this site, you consent to our use of cookies. H. O. Nyongesa, S. Al-Khayatt, S. M. Mohamed, and M. Mahmoud, Fast robust fingerprint feature extraction and classification, Journal of Intelligent and Robotic Systems, vol. The pattern is the unique characteristics of the ridges and valleys that make up the print. 'L''iBn$]FxGiCd\xo`8; ^OofO4(xC;9t H1fesM ()}kGglN`k\Y8UO}=cP{8& 1 The geometric properties of major ridge curves in a fingerprint image called orientation field flow curves were used to achieve a manual fingerprint categorization (OFFCs) [28].