Feature extraction methods using biomedical images for content based image retrieval

In informatics world, storing the visual information and visual information retrieval when it is required have been gaining an increasing importance day by day. when the increasing need of visual database both personally and institutionally is taken into consideration, it has become a significant problem to retrieve images from a big image database. A content based image retrieval system is based on the basis of querying by computation upon the content of the image. Here, the aim is to enable the retrieval of the image by using its basic features such as color, shape and texture.In this thesis, Content Based Image Retrieval systems have been throughly studied and statistical texture analysis has been made on the biomedical image database built in order to use in medical applications. These applications are in the form of background of a medical content based retrieval system. In literature, there are some content based retrieval applications on various databases. Moreover there has been some information given about Content Based Image Retrieval System such as QBIC, in literature. However, the aim of this thesis is to compute the textural features on biomedical image by statistical methods and to evaluate the performances of these methods.