【摘要】 针对传统Canny算子在边缘检测中的不足,提出一种基于Canny算子图像边缘检测的改进方法。传统Canny算子的高低阈值一般是人为地设定固定值,这容易造成虚假边缘。对此,利用最小交叉熵计算Canny算子的高低阈值,得到理想的边缘后,利用数学形态学对结果进行后处理。实验结果表明,改进后的算法具有自适应性,抑制噪声能力较强,有效地提高了边缘检测的鲁棒性。更多还原

【Abstract】 The traditional Canny edge detection algorithm has the defects of losing details of the edge. In this paper an improved Canny-based edge detection algorithm is proposed.High and low threshold are given as an fixed value in traditional Canny operator, which may cause false edge information when the image’s all gray level is concentrated on one region. A desired threshold value is given based on minimum cross entropy, and the obtained edge will be handled by mathematical morphology.The experiment results indicate that the algorithm improves the effect of edge detection in the case of noise disturbance, and has a good robustness and also a strong self-adaptability.更多还原