在计算机视觉领域，图像分割（Segmentation）指的是将数字图像细分为多个图像子区域（像素的集合）（也被称作超像素）的过程。超像素由一系列位置相邻且颜色、亮度、纹理等特征相似的像素点组成的小区域。这些小区域大多保留了进一步进行图像分割的有效信息，且一般不会破坏图像中物体的边界信息。(n the field of computer vision, image segmentation (Segmentation) refers to the process of segmenting a digital image into multiple image sub regions (a collection of pixels) (also referred to as hyper pixels). A small region consisting of a series of pixels similar to each other in positions, colors, brightness, texture, etc.. Most of these regions retain the efficient information for further image segmentation, and generally do not destroy the boundary information of objects in the image.)