Indexing Based Approach for Implementation of Latent Fingerprint Matching System

Abstract

Fingerprints have been considered an important tool for investigating crimes due to their uniqueness and persistence To overcome these problems an ALFIS is presented in the presence of extensive manual intervention with the extraction of latent features A successful matching depends on a reliable extraction on ridge Machine Learning by far has played an important role in increasing the performance of the fingerprint recognition system Various type of model via ML and neural networking has been used for a better image quality estimation Reducing of search time in finding the matching identity of the query image fingerprint indexing technique has been introduced It is observed from the conventional approaches that extraction of the correct features is not simple In identification system, comparison of the speed of the two fingerprints depends on the similarity measure between the FP database and the query print The selection of optimal features consumed more amount of time as will it increased the The capability of the image pre processing and enhancement techniques to improve the image intensity has helped to obtain better features In latent fingerprint indexing the presence of disturbing textures and background noise are two major factors which affect the recognition quality Classification using Deep Learning with LFI Image enhancement technique Preprocessing and Feature Extraction methods are being used for sparsity test and statistical analysis LFI for faster retrieval from dataset with Image enhancement technique is being used which separated the foreground and the background blocks this helped in achieving the segmentation accuracy This research work is to show the result of each phase using different preprocessing improvement techniques Neural Network algorithms reduces training and testing times Combining local match with global match increases indexing efficiency For this, a new machine learning based segmentation algorithm was used for the preprocessing phase to increase the accuracy of the match