【摘要】 针对自动生产线上的机械零件图像几何特征,提出一种基于Canny算子的分块自适应边缘检测方法。主要包括基于鼠标框选的图像不均匀分块和基于Otsu算法的检测阈值自适应生成等。实验表明,该法有较强的抑制噪声能力,检测准确,是一种有效的机械零件图像边缘检测方法。更多还原

【Abstract】 A blocking adaptive edge detection method based on the Canny operator is proposed in this paper,which is adapt to inspect the machinery parts machining on the automatic product line.It includes the non-uniform blocking process of the machinery parts images using the mouse pane selection and the method of obtaining each sub-block’s detection threshold adaptively based on Otsu arithmetic.The experimental results show that the method enables to control the noise and it is a correct and effective edge detection method for the machinery parts images.更多还原