摘要
在基于机器视觉的测量中,由于影像设备的性能限制与对图像的精度要求,需要对拍摄的图片进行拼接。对于一般灰度图像进行旋转操作会对图像造成失真的情况,文中针对中小模数齿轮的灰度图像拼接提出了基于边界的齿轮拼接方法,对滤波后的齿轮图像进行边缘提取,对提取到的数据点采用半径约束的最小二乘法拟合圆心,圆心配准后对齿廓数据点进行配准以求得旋转角度完成拼接。仿真结果表明该方法可以将拼接误差控制在一定范围内。
In the measurement of machine vision, it is necessary to splice the captured picture due to the performance limitation of the image equipment and the accuracy requirement of the image. Normal gray image rotation operation will cause the distortion of the image. In this paper, a gear-based splicing method based on boundary is proposed for the gray-scale image splicing of small and medium-modulus gears. The edge of the filtered gear image is extracted, and the extracted image is extracted. The data points are fitted by the least squares method of radius constraint. After the center registration, the data points of the tooth profile are registered to obtain the rotation angle to complete the splicing. The simulation results show that the method can control the splicing error in a certain range.
出处
《机械工程师》
2017年第11期37-39,共3页
Mechanical Engineer
关键词
图像拼接
齿轮图像
边缘提取
图像配准
image splicing
gear image
edge extraction
image registration
