Image matching method and device, computer equipment and storage medium

Abstract

The invention discloses an image matching method and apparatus, a computer device and a storage medium. The method comprises the steps of obtaining a to-be-matched image and a template image; obtaining a target neural network model, wherein the target neural network model comprises a convolutional neural network and a mutual information response network; respectively inputting the to-be-matched image and the template image into a convolutional neural network of a target neural network model to obtain a first feature map and a second feature map; respectively inputting the first feature map andthe second feature map into a mutual information response network of a target neural network model, and calculating a mutual information response between the first feature map and the second featuremap to obtain a mutual information response area in the first feature map; based on the first mutual information response value, screening a mutual information response region in the first feature mapby adopting a non-maximum suppression method to obtain a target region; therefore, the problems of long time consumption, low efficiency and inaccurate result of image matching are solved, and the efficiency and accuracy of image matching are improved.

Description

technical field [0001] The invention relates to the field of intelligent decision-making, in particular to an image matching method, device, computer equipment and storage medium. Background technique [0002] Searching for a desired object within a given area is a common task in the field of computer vision. Currently, there are many solutions to this task, and template matching is one of the most commonly used methods. Template matching is to study the position of the template image of a specific object in the image to be matched, and then identify the object. It is the most basic and commonly used matching method in image processing. However, most of the existing template matching techniques need to be based on a given model, and after traversing the entire search graph, return the most similar subgraph as the matching result. In this way, the entire matching process takes a lot of time and the matching efficiency is low; In addition, some template matching uses graysca...

AI Technical Summary

Problems solved by technology

[0003] Embodiments of the present invention provide an image matching method, device, computer equipment, and storage medium to solve the problems of time-consuming image matching, low efficiency, and inaccurate results

Images