ISSN : 2319-7323

INTERNATIONAL JOURNAL OF COMPUTER SCIENCE ENGINEERING

ABSTRACT

Title	:	IMAGE RETRIEVAL SYSTEM BASED ON COLOR (QUANTIZED HSV) AND TEXTURE FEATURE (GLCM)
Authors	:	Nishad Ahmed, Parismita Sarma
Keywords	:	Image Retrieval (IR), Content based image retrieval (CBIR), Color, Texture, HSV Color space, Gray Level Co-Occurrence Matrix (GLCM), Euclidian distance
Issue Date	:	May 2017
Abstract	:	An Image Retrieval (IR) is a system which allows user to browse, search and retrieve digital images based on visual Features such as color, texture and shape. Image retrieval based on single feature cannot provide a good result for accuracy and efficiency. High level feature like object, Semantic etc describes the concept of human brain effort which will reduce the query efficiency and low level features such as color, texture and shape will reduce the query accuracy. So, it is good to use multi feature in image retrieval system. The most important visual features are Color and texture which are used in this review paper. A review on Retrieval of Digital images based on color and texture feature have been done in this paper and also a new system has been proposed for efficient Image Retrieval System using multi-feature. In this paper, three methods for image retrieval system based on color, texture and combination of both color and texture feature have been proposed. Here, color feature has been extracted by using quantized HSV Color space. In HSV (Hue, Saturation, Value) color space, first quantifying the color space in non-equal intervals, then constructing one dimension feature vector and after that representing the color feature by finding the summation of one dimension feature vector. Similarly, the work of texture feature extraction has been done by using gray-level co-occurrence matrix (GLCM) and the work of both Color and texture feature have been extracted by using the quantification of HSV color space and Gray-level co-occurrence matrix (GLCM).The technique shows the advantage of using multi-feature than the single feature in image retrieval system. In order to express the similarity between different images or feature vectors of images, Euclidian Distance is used
Page(s)	:	124-132
ISSN	:	2319-7323
Source	:	Vol. 6, No.5