This dissertation work is concerned with the extraction of uniform texture regions based on conditional one-dimensional histograms and the method of fast and adaptive texture feature estimation for segmentation and the boundary coding (representation) of the extracted regions using the run-length and chain codes.
Firstly, and efficient feature transform method is proposed for extracting an uniform texture region using the conditional one-dimensional histogram. The conditional one- dimensional histogram and a method for extracting a homogeneous texture region are described. The method of calculating covariance matrix between texture features is proposed. By applying this procedure to the artificial texture image, some desirable results are obtained and experiments show that the feature transform method can be used to extract a homogeneous texture region based on the conditional one-dimensional histograms.
Secondly, a fast and adaptive algorithm to estimate texture statistics based on the SGLDM (spatial gray level dependence matrix) is presented for texture image segmentation. Several recursive equations are proposed by modifying the conventional one in order to get the texture statistics at high speed. And an adaptive window selection method is studied to find the more accurate texture statistics. By applying the proposed method to the texture images, it is shown that the proposed method is useful to estimate texture statistics for image segmentation. Finally, a new algorithm is described for converting the extracted regions (a binary image) into chain codes using its run-length codes. The basic idea of the conventional chain coding algorithm is to follow boundary pixels by convolving a 3x3 window the image and to sequentially generate chain codes. The proposed algorithm has two phase, namely run-length coding and chain code generation. We use connectivity information between runs as well as their coordinates in the phase of run-length coding. In the second phase (chain code generation) the connectivity information extracted in the first phase is utilized for sequentially tracking runs containing the boundary pixels to be followed. This algorithm has an advantage that we can detect easily the inclusion relationship between boundaries at the same time as chain code generation.