基于Canny算子的建筑裂缝边缘检测改进算法

Improved Algorithm for Edge Detection of Building Cracks Based on Canny Operator.

针对Canny算法在建筑物裂缝图像边缘检测中去除伪边缘和保护边缘细节方面的不足,提出了一种改进Canny算子的建筑物裂缝边缘检测算法。采用一种将双边滤波器和中值滤波器相级联的混合滤波器对建筑物裂缝图像进行处理;采用细分划分方法改进传统Canny算法中2×2邻域来计算梯度幅值;采用最大类间方差法与梯度直方图结合算法计算得到高低阈值。实验结果表明,改进算法能够解决建筑物裂缝图像中噪声干扰等问题,提高了Canny算子在提取建筑物裂缝图像边缘细节信息和抑制假边缘方面的性能,具有较好的鲁棒性。

To address the shortcomings of Canny algorithm in removing false edges and protecting edge details in the edge detection of building crack images,an algorithm for edge detection of building cracks based on an improved Canny operator is proposed.A hybrid filter cascading a bilateral filter and a median filter is used to process the image of building cracks.To improve the 2×2 neighborhood in the traditional Canny algorithm,the subdivision method is adopted to calculate the gradient amplitude.Otsu and gradient histogram algorithms are applied to calculate high and low thresholds.Experimental results show that the proposed algorithm can solve the problems of noise interference in the building crack image,which improves the performance of the Canny operator in both extracting the details of the edge of the building crack image and suppressing false edges.This algorithm has good robustness.