中文摘要：

简单线性迭代聚类（SLIC）方法需要人为设定图像分割时所需超像素的数目,这样设定值容易过大或过小,不能保证待分割图像处理的要求。针对该问题,对SLIC超像素分割方法进行了改进,提出利用图像的颜色信息,对所要处理图像的横坐标和纵坐标进行随机扫描,然后分析扫描线像素的颜色信息,从而得到图像所需要的超像素数目。仿真实验表明,该方法可以获得图像所需要的相对合适的超像素数目,用这些超像素对图像进行分割,能达到较理想的分割效果。

英文摘要：

In the Simple Linear Iterative Clustering （SLIC） method, the number of superpixel needs to be set manually. So this number can not be very accurate. We can not ensure this value is the most appropriate one comparatively. This article brings out a modified way to solve the above problem. To get the number of superpixel, the color information of picture can be used to scan some lines randomly and analyze the segmentation situation of similar color in pixel line. The simulation experiment shows that this method can help us get the appropriate comparatively number of superpixel.