Robust Object Segmentation Using Genetic Optimization of Morphological Processing Chains

A semi-automated object segmentation approach has been introduced in this paper. Object segmentation is a crucial task in image processing. The proposed approach learns segmentation from a small number of gold samples. The segmentation is performed in two main sequential steps, namely, target object localization, by applying optimal mathematical morphology procedure, and seg-mentation, by conducting some basic image processing operations. The outstanding feature of this approach is, unlike other existent approaches, that it does not need a prior knowledge or a large number of samples to learn from. The performance of the approach has been examined by a comprehensive well-designed validation set. For all test images, the target object was segmented accurately and the conducted experiments clearly showed that the proposed segmentation approach is highly invariant to noise, rotation, translation, overlapping, and scaling. The architecture of the approach and employed methodologies are explained in detail. Results are provided