Research And Implementation Of SAR Image Retrieval System Based On The Contourlet Transform

A large number of SAR images come forth because of the development of Synthetic Aperture Radar(SAR) technology. This results in a new demand of SAR image processing and its application. Traditional text keyword-based retrieval approach can't satisfy the demand of image retrieval, so content-based image retrieval (CBIR) technology becomes the current research focus. In this thesis, according to the natural science fund of Jiangsu Province, on the basis of analyzing the key techniques of CBIR, we mainly research on the application of the multiscale geometric analysis technology-the Contourlet transform in image retrieval algorithms based on texture feature and shape feature. In addition, these algorithms are applied to achieve a SAR image retrieval system. The main content of this thesis are summarized as follows:1. Some key techniques and algorithms of CBIR, such as the low-level feature descriptions including texture, shape, the similarity measure between images, the indexing methods and so on, are deeply analyzed and studied .2. We research on the texture-based image retrieval algorithm and propose an algorithm of texture feature extraction based on the Contourlet transform in this thesis. The image is decomposed by the Contourlet transform. The mean and standard deviation of the magnitude of the Contourlet coefficients at different scales and directions are computed to extract the texture feature vector.3. We research on the shape-based image retrieval algorithm and propose an algorithm of shape feature extraction based on the Contourlet transform in this thesis. We combine the Canny operator and the Contourlet transform. On the basis of extracting the edge of the image with the Canny operator, Contourlet transform is used to decompose the shape information to different scales and different directions. As a result, shape information under each frequency is remained and loss is reduced.4. We design a content-based SAR image retrieval system using Visual C++ 6.0 and Access2003. A SAR image retrieval system is carried out by using the texture and shape extraction algorithms based on the Contourlet transform. The system can not only implement the texture and shape retrieval of SAR images, but also be used for algorithm evaluation and performance comparison. Besides, different algorithms can be combined to receive better results.