A Novel Automatic Seed Placement Approach for Region Growing segmentation in Mammograms
Ahlem Melouah, Soumaya Layachi · 2015
Numerous researchers introduced several mammograms segmentation methods; but, region growing algorithm is the most used one. This algorithm starts by selecting a seed point then grows the seed area relying on the fact that pixels which are close to each other have similar features. To improve the accuracy and effectiveness of region growing segmentation, some works focus on automating seed selection step. This paper proposes a new algorithm to select automatically the seed position for further region growing expansion. It also presents the role of automatic seed placement when segmenting breast lesions in mammograms. The experimental results show the performance and robustness of the proposed method.