Image recognition method and device, computer equipment and storage medium

Abstract

The invention relates to an image recognition method and device, computer equipment and a storage medium. The method relates to the field of computer vision of artificial intelligence, and mainly relates to an image recognition technology in the field of computer vision. The method comprises the steps of obtaining a to-be-recognized image which comprises a to-be-recognized traffic sign; performing feature extraction on the to-be-identified image to obtain a feature vector of the to-be-identified image; matching the feature vectors with aggregation vectors of various traffic signs to obtain matching degrees of the to-be-identified image and the various traffic signs, wherein the aggregation vector corresponding to each type of traffic signs is obtained by aggregating feature vectors of a plurality of images including the corresponding types of traffic signs; and determining the type of the traffic sign to be identified in the image to be identified according to the matching degree. By adopting the method, the accuracy of identifying the traffic sign category from the image can be improved, and the identified traffic sign category can be applied to map construction or vehicle navigation.

Description

technical field [0001] The present application relates to the field of computer technology, in particular to an image recognition method, device, computer equipment and storage medium. Background technique [0002] With the rapid development of computer technology and image processing technology, more and more fields involve identifying the category of the target object from the image, so as to perform subsequent applications based on the category of the target object. For example, in the fields of autonomous driving and vehicle navigation, it is necessary to recognize the category of traffic signs from images. [0003] In the traditional recognition method, the classification network based on deep learning is usually used to identify the category of traffic signs in the image. However, due to the large number of traffic sign categories, the sample image categories used by the training classification network are unevenly distributed and the category coverage is not complete....

AI Technical Summary

Problems solved by technology

[0003]In the traditional recognition method, the classification network based on deep learning is usually used to recognize the category of traffic signs in the image, but due to the large number of traffic sign categories, it is difficult to train the classification network Factors such as uneven distribution of sample image categories and incomplete category coverage lead to low recognition accuracy of the classification network

Images