一阶边缘检测算法的研究

Study of First-order Edge Detection Algorithm

在机器人视觉伺服控制系统中,用一阶边缘检测算法把目标物与图像背景分离出来,是实现目标物识别和定位的重要手段之一。在此,重点探讨和分析一些重要的一阶边缘检测算法的特点及其原理,用这些算法对车牌号码进行检测,通过分析它们最终的检测效果发现,坎尼边缘检测算子能够检测到真正的弱边缘,可以使边缘的线型的连接程度较好,能把带宽细化成只有一个像素点宽,并且具有较好的定位性能。把坎尼算法应用在机器视觉伺服中是一种较好的边缘检测方法。

In robot visual servo control system,it is important means for using the first -order edge detection algorithm for edge detection method applied to image objects with other factors separated the target to achieve the recognition and positioning. Focusing on a number of important first - order characteristics of the edge detection algorithm, the prmciples are analysed. Through these algorithms on the test plate numbers to an analysis of their effect ,Canny edge detection operator to detect the weak real edge is found,the edge of the line can make the connection degree well,the bandwidth can refine into only one pixet wide,and it has a good positioning performance. Canny edge detector applied to visual servo machine is a better method of edge detection.