Image segmentation based on multi-scale local feature

Yang Wang, Qianqian Wang · 2010 3rd International Congress on Image and Signal Processing · 2010

Image segmentation is the basic problem of image processing and computer vision, and it's also the key step from image processing to image analysis. The existing image segmentation methods were firstly reviewed and then the concept of multi-scale local features was introduced. A region-driven image segmentation of multi-scale local energy was proposed later. Finally, these algorithms were evaluated by the manual segmentation result and the evaluation results showed that the method based on multi-scale local energy could obtain accurate segmentation results with high availability.

Read the paper · More papers on PaperTik