A Novel Technique to Detect Local Features in Digital Images

Author(s):

Hitul Angrish , CGC Technical Campus, Jhanjeri; Navjot Kaur; Raman Chadha, CGC Technical Campus, Jhanjeri, CGC Technical Campus, Jhanjeri

Keywords:

Interest Point, Features, Features Detection, Feature Description. Feature Extraction

Abstract:

In the computer vision field the subject of keypoints detection or feature detection has been a subject of popularity in the field of research from the past 20 years. However in the last few years the domain has shifted dramatically due to entrance of the hand held devices like mobile, tablets and phablets most likely the smart phones and the tablet computers has demanded the creation of new platforms. These platforms have the limited memory and lesser computational power. Thus a demand for the faster and efficient in memory utilization keypoint descriptor has been created for applications like object reorganization and panorama camera picture stitching. In this paper a new proposed technique named ZEON has been implemented and results of that new technique has been compared with SURF, BRISK and FREAK techniques of feature detection in digital images. The comparison is done between techniques is in terms of the Number of Features detected and Time Taken by each of the descriptors to do so. Thus by dividing the time by the number of feature we can calculate the average time taken for a particular feature lesser the number greater is Descriptor in terms of speed and lesser computation power. In the end conclusion of the results and future work to be continued has been discussed.

Other Details:

Manuscript Id	:	IJSTEV2I3017
Published in	:	Volume : 2, Issue : 3
Publication Date	:	01/10/2015
Page(s)	:	59-64

Download Article