基于改进Sobel算子的多仪表图像边缘检测算法

为了解决Sobel算子在多仪表图像检测中存在的边缘定位精度不高、提取边缘较粗以及对噪声敏感等缺点,提出了一种改进算法。首先在原有水平和垂直模板的基础上新增6个方向模板,提高了边缘定位精度,使其适用于纹理结构复杂的图像。然后根据图像的具体特征自适应确定阈值,有效去除图像的伪边缘。最后结合数学形态学的灰度腐蚀理论,对检测到的边缘进行细化处理。不仅能获得细化效果较好的边缘,同时可以有效的抑制噪声。实验结果表明:改进后的算法对图像噪声干扰有较强的抑制能力,提取的边缘定位准确、结构细腻。

Edge detection algorithm for multi-meters image based on improved Sobel operator

In order to solve the shortcomings of low edge positioning accuracy and coarse edge and sensitive to the noise of the Sobel operator in the detection of multi-meters image. In this paper, an improved algorithm was proposed to overcome the disadvantages. Through adding six direction templates based on the original horizontal and vertical templates to improve the edge positioning accuracy, make it applicable to texture complex images. Then according to the specific characteristics of the image to determine the adaptive threshold, effectively eliminates the pseudo-edge of the image. Finally, the detected edge is thinned by combining with theory of gray erosion in mathematical morphology, the method can not only obtain edge in better refinement effect, but also can effectively restrain the influence of the noises. The experimental result shows that the improved algorithm has a better immunity to the noise jamming of images and the edge position extracted by the algorithm is accurate.