Analysis and Design of Image Segmentation Algorithm Based on Super-pixel and Graph Cut
暂无分享,去创建一个
Image segmentation has always been a research hotspot in computer vision. Traditional image segmentation is based on pixels, and it is difficult to achieve the desired target effect. The manual segmentation method is too cumbersome and has limited precision. There is a certain degree of instability in the selection of interactively segmented pixels, which will affect the final segmentation results to a large extent. In this paper, the superpixel block generated by the SLIC algorithm is used to preprocess the image, and then based on the Graph cut algorithm, the selection of seed points is improved, and the pixels of the background area are attempted to be optimized to obtain a more excellent target point sample. On top of the image segmentation, the foreground object is further divided with a frame, the pixels in the background are optimized, and the advantages of other segmentation algorithms are extracted to achieve the goal of global optimization, and interactive image segmentation is achieved. Keywords-Interactive Segmentation; Super-Pixel;Graph Cut; Maximum Flow/ Minimum Cut; K-Means;SLIC
[1] Saurabh R Bhonde,et al. Improvement of K-Means clustering Algorithm , 2012 .
[2] Marie-Pierre Jolly,et al. Interactive graph cuts for optimal boundary & region segmentation of objects in N-D images , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.
[3] Hao Wu,et al. Image Segmentation Algorithm Based On Clustering , 2018, 2018 International Conference on Machine Learning and Cybernetics (ICMLC).
[4] D. Greig,et al. Exact Maximum A Posteriori Estimation for Binary Images , 1989 .