Content Based Image Retrieval Using Texture Structure Histogram and Texture Features

摘要

In Content Based Image Retrieval one of the most important features is texture. In this paper we present a image retrieval based on Texture Structure Histogram (TSH) and Gabor texture feature extraction. In TSH technique to describe the texture feature, we use the edge orientation and color information method. The HSV color space will be used in this technique for humans visual perception mechanism. To make the image content more reasonable non-equal interval quantization scheme will be used. One of the important texture feature analysis method is Gabor texture feature extraction. Using the mean and variation of Gabor filtered image the image texture will be retrieved. To get the same dominant direction rotation normalization will be used. In this paper the comparison of both the texture techniques will be discussed. This mechanism gives a good retrieval performance and more efficient techniques.