The Research On Texture-Based Image Retieval

As technologies of computer and web are widely applied, how to analyze, store and retrieve these images is an imperative problem. The content-based image retrieval technology can effectively resolve the problem and it also becomes a hotspot in study. The purpose of image retrieval is realization automation, the intelligence of image search and management way, searcher can carry out convenience, fast, accurately to seek. The texture is a primary feature of image and an important method in the content-based image retrieval system.The paper mainly focuses on the texture-based image retrieval techniques.The paper study image retrieval based on texture feature.All algorithms are proved validity through experiments.Feature extraction of image is one of the key technique of image retrieval. Wavelet transform has multi-resolution and multi-direction. It forms feature vectors can describes image in some extents,detail information of texture,and don't need to decompose full image,so arithmetic calculation obviously reduce,and the retrieval results are improved.It can be used for extensive image database retrieval.so wavelet transform is applied to extract image's texture features. This method is compared with the method of based on gray co-occurrence matrix of image,and based on grads- gray co-occurrence matrix of image. This paper designs and realizes a image retrieval system based on texture feature.This text presents program flow diagram,and introduces seeking module,feature extraction module,matching module, displaying of image and the function of all modules.The system adopts Euclidean distance as similarity measurement of image, respectively uses three methods of extracting texture feature. Retrieval is experimented to given image eventually. The experiment results find out ten images and array according to similarity form big to small.The results show texture feature based on wavelet transform accuracy efficiently and obtains the satisfied retrieval results.