Supervised deep learning methods for semantic image segmentation 

Image segmentation plays a fundamental role in a wide range of omputer vision applications, such as medical image analysis, robotic perception, video surveillance, etc. It provides key information for the overall understanding of an image. Many traditional image segmentation methods have been proposed in the literature, including thresholding, region-based segmentation, region growing, k-means ckustering, watershed methos and edge detection segmentation , etc. The methods use the knowledge of image processing and mathematics to segment the image in a simple and fast way. However , their accuracy cannot be guaranteed interms of details.