Research On Content-based Image Retrieval Technology

With rapid development of internet and technologies of computer and multimedia, information of all kinds of images becomes more and more great。 It is very important to organize, manage and utilizes these multimedia information。 Until now there are still many problems of this research area to be solved。 This thesis focuses on the technology of contented-based image retrieval (CBIR) and get some achievements in this area。One of short points of traditional color histogram is lack of spatial information。 To overcome this shortcoming, color coherence vector was introduced in this thesis, and at the same time, the color number of image was decided by image content。 Image content was divided into two parts: aggregation and non-aggregation, and based on these, we utilize the method of best-weight distance to get the value of similarity about two images。 This kind of measure about similarity of two images not only think of the similarity of the same color but also think of different color。 Experimental results prove that it is better than method of color histogram by reason of the shortcoming about not considering color space and not comparing similar color。Texture character is another important thesis of image research。 An energy-distribution signature is calculated using Gabor filters。 Conternted-based textured image retrieval system establishes color-based and shape-based retrieval。Users usually have different requirements when they search images in database。 So we can retrieval the results according to similarity measurement based on different signature。In similarity measurements mentioned above, the paper analyses the drawback of bin-by-bin-form distance and cross-in-form distance。 The best-weight distance is presented which matches human vision as well as possible。 Also, the principle of the distance is based on transport problem。 The experimental tests the distance is efficient in image retrieval。...