基于数学形态学和Canny算子的音圈马达磁体边缘检测

Edge Detection of VCM Magnets Based on the Mathematical Morphology and Canny Operator

针对音圈马达(VCM)边缘检测中容易出现虚假边缘及精度低等问题,文中根据VCM图像的特性,运用不同尺度的结构元素对图像进行膨胀、腐蚀交叉预处理,提取磁体的轮廓,并利用Canny算子提取图像的边缘。实验结果表明,文中方法能够有效地提取VCM磁体图像的边缘特征,为工业生产中VCM磁体的自动检测奠定了基础。

According to the voice coil motor（VCM） in the edge detection is easy to cause the false edge and low accuracy. In this paper, according to the characteristics of the VCM image,we use the different scales of structure element to the image expansion, corrosion cross pretreatment and extract the outline of magnets,and use the Canny operator to extract the edge of the image. The experimental results show that this method can effectively extract the VCM magnets image edge character for industrial production in the VCM magnets.