PROJECT TITLE :

Image Retrieval Based On Chrominance Feature Of The HMMD Color Space - 2017

ABSTRACT:

This paper proposes a new chrominance feature extraction technique in HMMD color area. Image dependent multi-level thresholding is performed within the HMMD color area to obtain the sixty four-IeveI quantized pictures. The incidence count of every color pixel represents the colour info of these quantized images. This system is tested over Wang's database of ten totally different class images. The distance live of this feature between the question and database image are calculated. Then, the proposed technique performance is evaluated using average precision and recall. Moreover, the proposed technique is a benchmark against the state-of-the-art color feature extraction ways and gives approximately half-dozen.3p.c to 18.05% and seven.fifty fourp.c to fourteen.52 % high precision and recall than the traditional techniques.