A method and system for judging image consistency based on optical flow method

Abstract

The invention discloses an image consistency judging method and system based on an optical flow method. The method comprises steps as follows: performing zooming processing and gray processing on a to-be-detected first contrast image and a to-be-detected second contrast image to acquire a first gray image and a second gray image, calculating the first gray image with the optical flow method on the basis of the second gray image to obtain an offset value, then performing threshold calculation and preliminary consistency judgment on the offset value, ending the process if the offset value is not consistent with a threshold, otherwise, further calculating the difference between the color value of each pixel of the first contrast image and the color value of each corresponding pixel of the second contrast image after optical flow according to the obtained offset value, and finally, performing difference threshold calculation and consistency re-judgment on the absolute value of the difference. With the adoption of the method and the system, the processing speed is increased, the automatic test efficiency is improved, and the accuracy of consistency test results is guaranteed through dual judgment.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method for judging image consistency based on an optical flow method and a system for applying the method. Background technique [0002] In recent years, the application of software automation testing in software quality control has become more and more common, especially with the development of GUI software, the advancement of multimedia technology has made images account for a considerable proportion of software output, which requires automated testing The system supports the image verification function, which verifies the correctness of the image and evaluates the image quality; in the existing image processing application, due to the difference in hardware and the size of the user's shooting, each mobile phone uses the same image for processing , the size of the obtained effect image may be different, which will interfere with the judgment of whether the effect of ...

AI Technical Summary

Problems solved by technology

[0002] In recent years, the application of software automation testing in software quality control has become more and more common, especially with the development of GUI software, the advancement of multimedia technology has made images account for a considerable proportion of software output, which requires automated testing The system supports the image verification function, which verifies the correctness of the image and evaluates the image quality; in the existing image processing application, due to the difference in hardware and the size of the user's shooting, each mobile phone uses the same image for processing , the size of the obtained effect image may be different, which will interfere with the judgment of whether the effect of the effect image is correct during the image verification process of the automated test

For example, the traditional image verification method is mainly to scale the effect image saved by a certain model and the preset second comparison image, and then calculate the difference according to the color value of the pixel at the same position, but we found that due to Different interpolation algorithms may lead to row or column offsets, resulting in a relatively large difference in color values, which leads to inconsistent judgment results, but in fact it is consistent

Images