边缘是图象特征的重要属性之一。边缘检测技术一直是数字图象处理中一个极为重要的研究课题。边缘精确检测技术具有相当难度,近十多年来,国内外不少作者在这方面做了大量研究工作,提出了许多算法。本文简扼回顾十多年来边缘检测算法,对其中典型边缘检测算法性能进行分析比较;指出了如何正确处理“定位精度与抑制噪声”的矛盾是当前边缘检测技术研究的核心;概述了当前边缘检测的重要进展及今后继续探讨的方向。 Edge is one of the important attributes of image features. Edge detection technology has been an extremely important research topic in digital image processing. Accurate edge detection technology is quite difficult, nearly a decade, many authors at home and abroad have done a lot of research in this area, put forward many algorithms. This paper briefly reviews the edge detection algorithms for more than a decade, and analyzes and compares the performance of the typical edge detection algorithms. It also points out how to correctly handle the contradiction between “positioning accuracy and noise suppression” at the core of current edge detection techniques. The important progress of detection and the direction to continue to explore in the future.