Image classification method, device, computer equipment and storage medium

Abstract

The application provides an image classification method, device, computer equipment and storage medium, belonging to the technical field of image processing. The method includes: inputting a target image into an image classification model, classifying the target image based on the image classification model, and outputting a plurality of first confidence levels; determining at least two Second confidence levels of the at least two object types corresponding to the image cluster; according to the second confidence levels of the at least two object types, at least one object type corresponding to the target image is determined. The above technical solution avoids the problem of low accuracy caused by directly outputting the confidence that the target image belongs to the object type, and determines the second confidence of at least two object types through the first confidence of at least two image clusters that meet the target conditions. Confidence, and then determine at least one object type, so that the accuracy of the at least one object type is relatively high.

Description

technical field [0001] The present application relates to the field of image processing, in particular to an image classification method, device, computer equipment and storage medium. Background technique [0002] With the rapid development of computer technology, image processing technology can be applied in a wider range. For example, when a user wants to retrieve an item, he can take and upload an item image including the item, and the server can classify the item image to determine the category to which the item belongs, and then display related items under this category. How to quickly predict the category of items provided by users is a problem that needs to be solved. [0003] At present, the common image classification method is to use the method of machine learning to extract the features of the input image, and then obtain the prediction result through the classifier to obtain the object type corresponding to the input image, that is, to obtain the object type of...

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present application provides an image classification method, device, computer equipment and storage medium. By inputting the target image into the image classification model, the first confidence degree that the target image belongs to multiple image clusters can be obtained, which avoids directly outputting the object The second confidence level of the type leads to the problem of low accuracy. Since one image cluster corresponds to at least one object type, the at least two image clusters can be determined through the first confidence level of at least two image clusters whose first confidence level meets the target condition. The second confidence degree of at least two object types corresponding to each image cluster, and then according to the second confidence degree of the at least two object types, determine at least one object type corresponding to the target image, thereby determining at least one object type of the target image higher accuracy

Images