基于图像处理技术的尺寸测量中边缘定位算法

为了进一步提高图像的定位精度,设计了一种基于插值理论的精确定位细分算法.通过多尺度小波变换与三次样条插值相结合,实现了亚像素级的测量.基于多尺度边缘检测的小波边缘检测方法,既能对边缘进行准确定位,又可以有效去除噪声干扰,提高边缘检测的稳定性和准确性.三次样条插值函数是分段插值函数,算法复杂度低、段与段之间连接处平滑、具有快速收敛性和稳定性.将上述两种融合的边缘定位算法,可以对图像进行精确检测和测量,其精度达到0.001个像素.

Edge Location Algorithm of Size Measurement of Image Processing Technology

To improve the location accuracy of the image further,a precise location subdivision algorithm based on the interpolation theory was designed. A sub-pixel level measurement is achieved through multi-scale wavelet transform combining with cubic spline interpolation. It can not only accurately locate the edge,but also effectively remove the noise and improve the stability and accuracy of edge detection according to the wavelet edge detection methods of multiscale edge detection. Here,the cubic spline interpolation function is a piecewise interpolation function,algorithm complexity is low,the connection between paragraphs is smooth with fast convergence and stability. The two fused edges location algorithm mentioned above can accurately detect the image with the accuracy of 0. 001 pixels.