Edge detection for image measurement based on nonlinear diffusion filtering

Image measurement technique is a new developing method in the non-contact measurement and instrument fields. An adaptive nonlinear flow equation process is presented to improve the measurement precision. First, an inverse diffusion is performed to sharpen edges along the normal directions to the isophote lines (edges), while a normal diffusion is done to remove noise and artifacts ('jaggies') along the tangent directions. Then, classical differential operator is employed to detect image edges. With a better edge location, experimental results show that sharper and smoother edges with one pixel are obtained in a relative error of 0.03. The method also shows its advantages in the minuteness measurement for a better edge location. In the case of blurry edges and additional noise in the image, the measurement results would be worse.