Research And Implementation Of Texture Image Retrieval

With the prevalence of the computer, digital cameras, multimedia technology and Internet, people have paid more attention to digital image resources, the number of digital images are rapidly increasing. It is difficult for common users to find how they really desired images from the numerous images. How to search the desired images from the large-scale image databases efficiently and effectively has become an important and challenging research topic. Content-Based Image retrieval (CBIR) was proposed based on automatically derived image features to solve the above problems.In recent years, CBIR has become the focus of attention and has been applied in many fields.To make up for the lack of wavelet transform description texture features, and to further improve the overall performance of CBIR systems, a novel approach based on Pulse-Coupled Neural Network (PCNN) for texture image retrieval is proposed in this research. The PCNN has many outstanding advantages to progress image. It can be used for many tasks in the field of image processing, such as image enhancement, image segmentation, edge detection, image fusion, feature extraction and target recognition. Outputs by using PCNN and ICM, which are particularly suitable for content-based image retrieval system, represent unique features of original stimulus and are invariant to translation, rotation, scaling and distortion. By adopting PCNN and simplified PCNN model ICM, the dual value image sequence corresponding to different gray levels is obtained. The variance of each image in entropy sequence is then calculated to convert into one dimensional eigenvector, and used to represent the image features. The Euclidean distance and Correlation Coefficient were used to compute the similarity between images. A texture retrieval system based on query image was developed.In this dissertation, the exploratory research work has been done around the feature extraction and similarity matching strategy. Firstly, the basic theory and key technologies of content-based image retrieval system are introduced. Second the low level visual features extraction of image has been studied systematically; these features include color, texture, shape features and so on. And then this paper exhaustively discusses the texture feature extraction techniques. Systematic analysis of the characteristics of pulse coupled neural network and Depth study of the advantages of image processing applied to image retrieval system. Finally, this research proposed a novel approach that based on pulse-coupled neural network for texture image retrieval, evaluated the performance of system, and then pointed certain directions of the future research.