The Key Issues Studied On Image Retriebal Of Integtated Color And Shape

With the popularization and development of Internet,the number of image data grows dramatically fast,and how to retrieve image efficiently and quickly becomes an important issue in the field of image's application. Traditional text-based image retrieval technology has been unable to meet the growing needs of Image Retrieval .In order to manage and retrieve large amounts of images,the CBIR has emerged to be one of the hot research areas in image domain.CBIR analyses the image on the basis of the color,texture and shape embedded in the image.Index of the image is built on the feature vector.The system retrieves the above information from the image database to satisfy the users'requirment. Color is the most widely used visual attributes. Classical color-based histogram is easy to compute, insensitive to translation, rotation and scale, and quite resistant to noises as well. But its drawback is lack of spatial information, prone to false hits when distinguishing images in large database with similar color composition but different spatial distribution. In order to solve probably inaccurate retrieval results caused by only visual feature,we present an image retrieval method based on color feature and shape feature.In this thesis,the research background,the important of the issue,the up-to-date applicationa and the development of the key techniques of content-based image retrieval are reviewed.First,we concentrate on the research of image retrieval using color,texture and shape information.Then,we discuss the method combine color and shape image retrieval methods from different aspects inclding features extracted,features empressed and similarity measurements.This kind of image retrieval is a better method which simulates the human comprehension of image ,and hence the semantic gap is narrowed.The proposed method is proved its validity and efficiency though theoretical derivation and experimental results.Secondly In order to make the result of retrieve as far as possible with the understanding of the same image content, reduce semantic gap, relevance feedback technology is proposed in this thesis.The main research work and innovation of this thesis are as follows:(1)After I read a lot of papers, for color histogram can not accurately express the color image spatial distribution of this shortcoming, I propose a method that block the image.The block can concentrate the information of the image,but at the same time increased the complexity of user access, in order to reduce the burden of its customers and reduce retrieval time of system, K-means clustering method will be used to solve the drawback.2)Recalling back the papers,most of them use the general color histogram, the cumulative histogram or partial cumulative histogram to extract image color feature. In the text ,I do experiment with matlab to improve the image's histogram and find the standardized histogram is better than the general brightness of the image, so I extracted each of the block color histogram, figure balance, standardization use it as the characteristics of image retrieval.3)Image similarity matching is an important part of the retrieval, in this paper I used a method that has been proved and published. This method can reduce the complexity of time.4)In the image's shape retrieval,I use seven shape moment invariants as the feature to retrieval.In order to allow the edge of the image more clearly, I do many experiments at last chose Canny operator of the image edge detection, and introduce the expansion of corrosion morphology extraction algorithm image contours.5)With VC + +, and Access matlab designed a content-based image retrieval system for a variety of image retrieval and backed up by relevant feedback technology to realize image retrieval and achieved better search results.