Automatic segmentation of clustered quantum dots based on improved watershed transformation

Automated analysis of the quantum dots (QDs) images is very important in the field of material science. In this frame, efficient QDs segmentation is prerequisite. In this paper, we propose an algorithm of automatic detection and segmentation of the QDs, especially the clustered ones. We depend on fuzzy c-means (FCM) method for initial segmentation of the QDs from the substrate background. Then we present a modified watershed algorithm with markers and a novel marking function. The markers are extracted by adaptive H-minima transformation. Then a marking function based on Quasi-Euclidean distance transform is introduced to accurately and rapidly separate the clustered QDs. We demonstrate the comparisons of our method with the existing approaches. The experimental results show that the proposed method is efficient and accurate with very little running time and has a high quality on QDs segmentation.