A blockwise relaxation labeling scheme and its application to edge detection in cardiac MR image sequences

We present a segmentation scheme for magnetic resonance (MR) image sequences based on vector quantization of a block partitioned image followed by a relaxation labeling procedure. By rst searching a coarse segmentation, the algorithm yields very fast and eeective performance on images that are inherently noisy, and can eeectively utilize the correlation in a sequence of images for robust performance and eecient implementation. The algorithm deenes feature vectors by the local histogram on a block partioned image and approximates the local histograms by normal distributions. The relative entropy is chosen as the meaningful distance measure between the feature vectors and the templates. After initial computation of the normal distribution parameters , a Block-wise Classiication Maximization algorithm classiies blocks in the block partitioned image by minimizing their relative entropy distance for a coarse resolution segmentation; and nally ner resolution is obtained by Contextual Bayesian Relaxation Labeling in which label update is performed pixel-wise by incorporating neighborhood information. Sequence processing is then performed to segment all images in the sequence. The scheme is applied to left ventric-ular boundary detection in short axis MR image sequences and results are presented to show that the algorithm successfully extracts the endocardial contours and that sequence processing signiicantly improves edge detection performance and can avoid local minima problem.