Shape Modeling with Front Propagation: A Level Set Approach
IEEE Transactions on Pattern Analysis and Machine Intelligence
Segmentation of tissue boundary evolution from brain MR image sequences using multi-phase level sets
Computer Vision and Image Understanding
Pixon-based image segmentation with Markov random fields
IEEE Transactions on Image Processing
| Hi-index | 0.00 |
Image segmentation is a key process of any image recognition system. Loose repeat algorithm and K-means algorithm are used to resolve the stomach epidermis tumor segmentation, but many conglutinated cells cannot be separated by others with the help of traditional segmentation algorithms. So the Vincent watershed algorithm as well as the Inver watershed algorithm is designed to do segmentation experiments about stomach epidermis tumors. A lot of experiments about several kinds' images were done to build the right segmentation theory. The conclusion of these experiments show that these two algorithms could get good segmentation results while the image has few conglutinated areas. Moreover, the Inver algorithm may get better result than the Vincent algorithm while the image has many conglutinated areas. As a defect of the Inver algorithm, it may take many very small areas after segmentation. Finally, boundary tracing algorithm is designed to obtain the boundaries of tumor cells.