{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/latent-fingerprint-recognition-role-of","title":"Latent Fingerprint Recognition: Role of Texture Template","arxiv_id":"1804.10337","date":"2018-04-27","proceeding":null,"authors":["Kai Cao","Anil K. Jain"],"abstract":"We propose a texture template approach, consisting of a set of virtual\nminutiae, to improve the overall latent fingerprint recognition accuracy. To\ncompensate for the lack of sufficient number of minutiae in poor quality latent\nprints, we generate a set of virtual minutiae. However, due to a large number\nof these regularly placed virtual minutiae, texture based template matching has\na large computational requirement compared to matching true minutiae templates.\nTo improve both the accuracy and efficiency of the texture template matching,\nwe investigate: i) both original and enhanced fingerprint patches for training\nconvolutional neural networks (ConvNets) to improve the distinctiveness of\ndescriptors associated with each virtual minutiae, ii) smaller patches around\nvirtual minutiae and a fast ConvNet architecture to speed up descriptor\nextraction, iii) reduce the descriptor length, iv) a modified hierarchical\ngraph matching strategy to improve the matching speed, and v) extraction of\nmultiple texture templates to boost the performance. Experiments on NIST SD27\nlatent database show that the above strategies can improve the matching speed\nfrom 11 ms (24 threads) per comparison (between a latent and a reference print)\nto only 7.7 ms (single thread) per comparison while improving the rank-1\naccuracy by 8.9% against 10K gallery.","url_abs":"http://arxiv.org/abs/1804.10337v1","url_pdf":"http://arxiv.org/pdf/1804.10337v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"latent-fingerprint-recognition-role-of","repo_url":"https://github.com/luannd/MSU-LatentAFIS","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"latent-fingerprint-recognition-role-of","repo_url":"https://github.com/prip-lab/MSU-LatentAFIS","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"graph-matching","task_name":"Graph Matching"},{"task_slug":"template-matching","task_name":"Template Matching"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}