Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/354098
Title: | Multimodal Biometric Cryptosystem for Human Authentication using Fingerprint and Ear |
Researcher: | Chanukya Padira SVVN |
Guide(s): | Thivakaran TK |
Keywords: | Computer Science Computer Science Information Systems Engineering and Technology |
University: | Meenakshi Academy of Higher Education and Research |
Completed Date: | 2019 |
Abstract: | ABSTRACT newline newlineThe main objective of the project is to develop a biometric verification system using ear and fingerprint to identify a genuine user .In this contest I proposed two novel methods to improve accuracy. newlineIn initial method First, a pre-processing phase by image enhancement and thinning method, all the images have the samesize. Then, a feature extraction technique includes Minutiae and Singular point technique for fingerprint images. Ear features are extracted by using Speed Up Robust Features (SURF) and Binary Robust Invariant Scalable Key points (BRISK) techniques are used to determine the ear and fingerprint features. Fusion at the feature level is carried out through concatenation for features. At last, matching is carried out by registration and similarity score process, then by using the threshold values, the user is identified as genuine or an imposter. newlineIn second method, there are three important levels involved in the biometric detection which includes Preprocessing, Feature extraction and Segmentation. The novel technique begins with the preprocessing phase including the median filter which gives a helping hand to the job of cropping the image for selecting the position. The preprocessed images are extracted with the assistance of the feature extraction phase in which the shapes and texture features of the face and ear images are efficiently extracted. The consistent images are extracted with the aid of the LGXP approach. Thus, in the feature extraction phase, various features like the shape, texture and ear as well as finger print images are extracted. Later, the extracted features are integrated. The integrated features, in turn, are properly classified by the firefly algorithm in accordance with the distance. newlineThe experimental results showed that the proposed multi-modal biometric methods achieved 95.66% and 96.28% accuracies. The novel techniques are effectively performed in the MATLABplatform. newline |
Pagination: | xvi 115 |
URI: | http://hdl.handle.net/10603/354098 |
Appears in Departments: | Department of Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 267.85 kB | Adobe PDF | View/Open |
02_certificate.pdf | 37.44 kB | Adobe PDF | View/Open | |
03_declaration.pdf | 52.81 kB | Adobe PDF | View/Open | |
04_chapter 1.pdf | 1.7 MB | Adobe PDF | View/Open | |
05_chapter 2.pdf | 680.85 kB | Adobe PDF | View/Open | |
06_chapter 3.pdf | 3.4 MB | Adobe PDF | View/Open | |
07_chapter 4.pdf | 1.23 MB | Adobe PDF | View/Open | |
08_chapter 5.pdf | 598.13 kB | Adobe PDF | View/Open | |
09_chapter 6.pdf | 433.11 kB | Adobe PDF | View/Open | |
10_bibiliograpy.pdf | 558.75 kB | Adobe PDF | View/Open | |
11_annexure.pdf | 2.09 MB | Adobe PDF | View/Open | |
12_content.pdf | 271.47 kB | Adobe PDF | View/Open | |
13_list of tables and figures.pdf | 186.88 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 623.69 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: