一种图像分割的目标描述方法及实现

An Approach of Object Description for Image Segmentation and Its Implementation

提出一种二值分割图像目标描述的实现方法 ,用以获取目标特征量。在对二值化图像形成目标块的过程中使用扩展像素标记法 ,可以在标记目标所属像素的同时标记目标的边界点 ,从而得到目标的单像素宽边界。在此基础上获得目标的面积、周长和质心坐标等参数 ,为图像目标识别做好准备。最后给出了算法C ++编程的具体思路和实现。仿真实验证实了该算法实用可行。

A practical approach of object description after image segmentation to obtain objects' features is introduced in this paper. In this approach an advanced pixel labeling technique is employed to form object region in the binary image because it can label all pixels belonging to the object and extract their edges with one pixel width in a single procedure. Consequently, the object's area, perimeter and coordinate value of center of mass are obtained, which are necessary for target recognition. Meanwhile, the details about the programming of this approach in C++ are described. Simulation results show that this approach is practical and feasible.