Adaptive Algae Image Segmentation Based on Morphological Watersheds

According to the differences in algae cells´ shapes, various watershed algorithms are used to adaptively segment the algae cell images. Considering the noises´ characteristics in these images, a new method is introduced in this paper. The basic idea is to segment the algae images, collected at scene, before denoising. The result shows that this algorithm can extract edges and inside details effectively and filter noises as well. And it works well in many images.