A Comprehensive Survey of Techniques/Methods For Content Based Image Retrieval System

Author(s):

Vidya Kalpavriksha , VTU Belagavi; Laxmi Kullur, VTU Belagavi; R H Goudar, VTU Belagavi

Keywords:

CBIR, Color, Shape, Texture

Abstract:

In this paper a comprehensive survey on various methods being used for Content Based Image Retrieval (CBIR) using color, shape and texture is been presented. Image retrieval is a process of extracting the contents of a image. This extraction of the contents of the image from large amount of database is not easy. The complete process of this retrieval is called “content based image retrieval”. The paper explains various methods of CBIR based on their related features. By using these methods and different techniques analyzing, searching, storing, browsing, retrieving and similarity of images from the image database can be done automatically. Based on color, the methods include RGB images, RGB color space, RGB color model, HSV color model, color moments, color correlogram etc. Based on shape the methods include Scale Invariant Feature Transform (SIFT), Histogram of oriented gradients (HOG),Nearest neighbor search method, Euclidean distance algorithm method, Prewitt operator method etc. Based on texture the methods included are Co-occurrence matrix, Color Correlogram, Color texture moments, Local Binary Pattern(LBP), Brodatz Texture Album and CUReT Dataset methods.

Other Details:

Manuscript Id	:	IJSTEV3I11177
Published in	:	Volume : 3, Issue : 11
Publication Date	:	01/06/2017
Page(s)	:	303-308

Download Article