EDA: Novel Technique for Image Matching

Author(s):

Midhu Yesodh , THEJUS ENGINEERING COLLEGE; Neenu P. A, THEJUS ENGINEERING COLLEGE; Jisney Thomas, THEJUS ENGINEERING COLLEGE; Princy P, THEJUS ENGINEERING COLLEGE

Keywords:

EDA, Encoded Images, Heterogeneous Images, IR, Modality Gap, Sketch

Abstract:

This paper recognizes the problem of matching images from different modalities. These different modality images are known as heterogeneous images. These type of images have large differences between them. These differences are known as modality gap. To reduce this gap, here proposed a new algorithm called EDA. It has wide applications like IR to optical image matching, Sketch to optical image matching etc. The main advantage of EDA approach is we can do matching of different modalities using the same feature descriptor. The key idea is to learn an encoding mechanism which will turn images in different modalities into their encoded images to reduce the modality gap between them. Here Sketch to optical image matching is done as an example.

Other Details:

Manuscript Id	:	IJSTEV1I10152
Published in	:	Volume : 1, Issue : 10
Publication Date	:	01/05/2015
Page(s)	:	348-352

Download Article