Efficient Hand-dorsa Vein Pattern Recognition using KNN Classification with Completed Histogram CB in TP Feature Descriptor
C. Premavathi1, P. Thangaraj2

1C. Premavathi, Department of Computer Science, Navarasam Arts and Science College for Women, Arachalur (Tamil Nadu), India.
2P. Thangaraj, Department Computer Science and Engineering, Bannari Amman Institute of Technology, Sathyamangalam (Tamil Nadu), India.
Manuscript received on 10 December 2018 | Revised Manuscript received on 29 December 2018 | Manuscript Published on 09 January 2019 | PP: 50-55 | Volume-7 Issue-4S November 2018 | Retrieval Number: E1868017519/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Hand-dorsa Vein Recognition System identifies an individual using the human hand vein features. Image capturing, extracting the features, keeping the features in a descriptor and making classification are important methods in hand-dorsa vein Recognition. In this paper, the feature descriptor and classification method is proposed for an efficient recognition system. A completed CB in TP has been proposed to represent selected features from Hand vein image system. K-nearest classification method with various proximity measure calculations is analysed to make an efficient classification system. A new minimum distance classification is proposed with dataset and the results are checked for accuracy and reliability. The proposed technique is calculated on a NCUT Dataset contains 2040 images from Prof. Yiding Wang, North China University of technology (NCUT) (Wang et al, 2010). Proximity process as Chi-square, City block, Euclidean, Chebychev along with Murkowski are calculated and compared used for the better performance. The new results prove to facilitate the future feature descriptor achieved excellent performance for classification system.
Keywords: Feature Descriptor, K-Nearest Image Classification.
Scope of the Article: Classification