Research On Image Retrieval Based On Color And Texture Feature

With the development of the multimedia network technology and the application of the image is extensive, Content-based Image Retrieval (CBIR) becomes a hotspot in multimedia retrieval field. CBIR extracts visual features as the retrieval features, such as color, texture, shape etc. Generally, features of the example image needed are extracted firstly, and then compared with the features of other images in the database after the result is showed to the user. Extracting features from image is the key issues in CBIR.This paper firstly investigates the correlated technology of CBIR, and then lucubrates the image retrieval based on the color feature,the texture feature and both of them.In spatial domain, we firstly study the extract methods of color and texture features. In the color features aspect, we introduce an image retrieval algorithm based on HVS Weighted color features. Firstly, we split the image into sub-blocks and draw the color feature consists of dominant colors in each sub-block. Then weighting the gained color features by making use of Human Visual System. So we can obtain the weighted dominant color feature. In the texture features aspect, we introduce a retrieval algorithm based on the gray-difference primitive co-occurrence matrix. The methods imbibe the co-occurrence matrix'merit, and combine the texture primitive of structure analysis, so that the texture feature is much fine. On this foundation, we introduce an image retrieval algorithm using both color and texture features. The experiment result testifies the validity of the proposed algorithm.In wavelet domain, we introduce a color features'extract method based on HSI low-frequency subband color features after partition in the color features aspect. Firstly, according to the image attention from human eyes, we split the image into sub-blocks. Then extract HSI low-frequency subband color features of each sub-block after wavelet transform, and we can obtain the image color features by weighting. In the texture features aspect, we put forward a texture features'extract method based on the gray-difference primitive co-occurrence matrix of high-frequency subband and the orientation features of each subband, and add right to each subband by analyzing its importance. The result is better. On this foundation, the image retrieval algorithm using both color and texture features improve the image retrieval precision.