Wavelet use for reduction of watershed transform over-segmentation in biomedical images processing

Problems of multi-dimensional signal enhancement, segmentation, feature extraction and components classification is essential in many engineering and biomedical applications. The paper is devoted to the use of watershed transform and wavelet transform for MR image components detection and discussion of over-segmentation problems. The goal of the paper is in (i) analysis of image de-noising, (ii) discussion of image enhancement, and (iii) multi-resolutional approach application for reduction of over-segmentation problems. Proposed algorithms include the use of wavelet transform and gradient methods in the preprocessing stage and application of the watershed transform for enhanced images. Resulting algorithms are verified for simulated images and applied for a selected MR biomedical images containing different structures.