Automatic image mosaicking method for high-accuracy image measuring apparatus of super-view field part

Abstract

The invention relates to an automatic image mosaicking method for a high-accuracy image measuring apparatus of a super-view field part. A template matching algorithm based on similarity measure is adopted and an index template is built through a run length encoding method to appoint pixels participating in operation, wherein the template is selected by adopting a maximal overlap region so as to contain image characteristics as much as possible, and the index template only contains pixels with edge characteristics or other pixels with obvious gray level change. The invention has the advantages that: the template is automatically determined without man-machine interaction selection, thereby enhancing the efficiency; the determined template has most image characteristics of objects to be measured; the pixels participating in the operation have obvious characteristics, and the other pixels do not participate in the operation, therefore matching accuracy is enhanced; the pixels participating in the operation, which is appointed through the index template, only occupies a great small part of the integral template, therefore operation efficiency is enhanced; the index template is built through the run length encoding method, therefore storage efficiency is high, and computer processing is convenient.

Description

technical field [0001] The invention belongs to the technical field of automatic image acquisition of ultra-field parts in a high-precision image measurement system, and in particular relates to an automatic image splicing method of a high-precision image measuring instrument for ultra-field parts based on optimal template extraction. Background technique [0002] The precision of machining depends largely on the development of measurement technology. Image measurement technology based on machine vision is a new technology with the most development potential in the field of precision testing. It uses electronics, photoelectric detection, image processing and computer technology to introduce machine vision into industrial inspection, and realizes the measurement of part size, The rapid measurement of geometric parameters such as shape and relative position has the outstanding advantages of non-contact, high speed, high precision and good flexibility. [0003] Traditional mac...

AI Technical Summary

Problems solved by technology

Traditional template selection is often realized by human-computer interaction, and the user selects a gray area with obvious characteristics as a template, which reduces efficiency and cannot realize splicing automation

Other automatic template selection algorithms cannot extract the optimal gray area, or the algorithm is too complex to improve the real-time performance of the system

Images