With the development of dermoscopic technology,the high-resolution dermoscopic images obtained are relatively large.Using SLIC algorithm to construct superpixels can reduce memory requirements and reduce computational time when using Ncut clustering to fuse superpixels in the future.Therefore,a dermoscopic image segmentation algorithm is proposed based on the combination of improved SLIC and Ncut clustering algorithm.The experimental results show that compared with the comparison algorithm,the proposed improved algorithm performs better in segmentation of low contrast and uneven brightness skin mirror images in the ISIC2018 dataset.
NETL
NSTL
万方数据
