Research And Implementation Of Color-Based Stone Categories Algorithms And Testing System

With the rapid development of information technology and network technology, the picture information database of huge,different contents is appearing constantly. And the picture resources especially on Internet,are doubling and redoubling constantly.As the same time,the development of stone industry leads to the growth of the stone-image information database.In order to retrieve the needed image rapidly and efficiently,the Content-based Image Retrieval(CBIR) has emerged in 20^th century 90's.CBIR analyses the image on the basis of the color,texture and shape embedded in the image.This paper applies the theory of CBIR to research the retrieval of stone image.According to the features of the same stone-image, the image retrieval system adopts color feature as its retrieval method in this paper.Since color feature is relatively robust to background complication and independent of image size and orientation,it is widely used in image feature representations.In this paper the color space model and the methods of color feature extracting is introduced firstly,and the algorithms for color feature representation of stone images are proposed. They selected the HSL color space consistent with human's visual characteristics,used the perception of human,the weights of HSL color space are non-interval quantified,and then the vectors of color features are constructed.Secondly,the similarity measure algorithms are proposed in detail.Finally,the algorithms proposed in the paper are tested and evaluated.The results demonstrate that the improved approach can obtain better retrieval results than others.This issue is funded by Fujian Province Scientific Project "Research on Key-technologies of Stone Retrieval Based on the similarity of images".The testing system is developed and used to test the retrieval effect of algorithms. The developing system is Windows XP and the development environment is Visual Studio.NET2003 of Microsoft Company.The results of testing are satisfied.Further work will be done to develop this system so as to release an applied government information retrieval system.