改进的Canny算子在裂缝检测中的应用

Application of improved canny operator in crack detection

结合OpenCV函数库提出一种新的Canny边缘检测方法,对视频数据进行处理,从而保障和提高路面裂缝检测的实时性和准确性。通过将多尺度形态学和双边滤波结合的去噪方法引入Canny算法中,不仅加强了降噪的效果,还能保留图像的边缘特征;在此基础上,用3×3的Scharr模板计算梯度幅值,以提高边缘检测精度,用最大熵自适应获取阈值、形态学滤波消除小的空洞,填补轮廓线的裂痕。实验结果表明,改进的Canny算法应用于不同路面裂缝检测时,都能够得到较好的结果,较高的检测精度。

A new Canny edge detection method based on OpenCV library is proposed to process the video data to protect and improve the real-time and accuracy of pavement crack detection. The introduction of the denoising method combined with multi-scale morphological and bilateral filtering into Canny not only enhances the effect of noise reduction, but also preserves the edge features of the image. On the basis of this, using the 3 × 3 Scharr template to calculate the gradient Amplitude to improve the edge detection accuracy, maximum entropy adaptive threshold, morphological filtering to eliminate small holes to fill the contour lines of the cracks. The experimental results show that the improved Canny algorithm can get better results and higher detection accuracy when applied to crack detection of different pavement.