template matching method based on edge features

In order to solve the problem that the identification process is easily interfered by environmental factors in a 2D visual task, the invention provides a template matching method based on edge characteristics, which comprises the following steps: multiplying Gaussian filters of multiple scales to obtain a new filter; Carrying out convolution on the collected picture by using the filter to obtain an edge feature image; Carrying out binarization on the edge feature image to obtain a binary image; Solving a minimum bounding rectangle for the feature image, and obtaining a deflection angle of thetarget; Establishing a template set according to the obtained deflection angle and the binary image; And processing the collected images to obtain a final binary image, matching the final binary imagewith a template set, and identifying a specific workpiece. The method is a template matching method based on edge characteristics, and the influence of illumination change, noise and rotation of a target object on the matchin