基于小波变换和Canny算子的齿轮边缘检测

Gear Edge Detection Based on Wavelet Transform and Canny Operator

齿轮表面的缺陷检测是保证齿轮使用安全的一个重要手段。本研究对象为整个缺陷检测系统中的一部分,对输入的图像进行边缘检测,得到图像的边缘信息,是缺陷检测前的准备工作。基于小波变换多尺度多分辨率的特性,提出了将小波变换和Canny算子相结合的方法,对齿轮边缘进行检测。首先采用中值滤波去除图像噪声,然后根据小波变换理论,选取Haar小波提取图像的低频系数并重构,再用Canny算法检测齿轮的边缘,并对实景拍摄图片进行检测实验。结果表明:当小波尺度为1时,与单纯的Canny算法检测相比,改进算法在检测精度上较Canny算子有所改进,能够较好地检测边缘。

Surface defect detection is considered as an important means to ensure the gear safety.A part of the whole surface defect detection system was studied in the present work.As preparation for the next step of defect detection,edge detection of an inputted image was performed to obtain the edge information of the image.Based on the multi-scale and multi-resolution characteristics of wavelet transform,an edge detection method by combining wavelet transform with Canny operator was proposed to detect the edge of the gear.Firstly,the image noise was removed by median filtering.Then the haar wavelet was used to extract the low frequency coefficient and reconstruct the image.Finally,the gear edge was detected by using the Canny operator.The experiment was conducted based on the real image.Compared with the Canny operator method,this modified method can obtain more precise results.