Method for classifying images on terminal equipment, medium and system

Abstract

The embodiment of the invention relates to the field of image processing, and discloses a method for classifying images on terminal equipment, a medium and a system which method can increase the number of categories supported by image classification on the terminal equipment by combining the image classification capability of an image classification model in the terminal equipment and the image classification capability of an image classification model on a cloud server, improve the accuracy of image classification, and avoid privacy leakage of the image in the terminal equipment. The method comprises the following steps that: the terminal equipment classifies a plurality of images on the terminal equipment into a first type through an image classification model in the terminal equipment; the terminal equipment determines that a condition of reclassifying the plurality of images is satisfied; the terminal equipment obtains a first image subdivision model corresponding to the first type from the server; the terminal equipment classifies the plurality of images into a plurality of first sub-types of the first type through a first image subdivision model; and the terminal equipment deletes the first image subdivision model. The method is specifically applied to a scene for classifying images on terminal equipment.

Description

technical field [0001] The present application relates to the field of image processing, and in particular to a method, medium and system for classifying images on a terminal device. Background technique [0002] Terminal devices (such as mobile phones, tablet computers, etc.) often store a large number of image files, such as photos taken by users every day. For a large number of image files in the terminal device, the terminal device generally uses image classification technology to classify the images, so as to support users to search for images of a certain category through the image search function, such as images of landscapes, flowers, animals, cars, etc. [0003] Currently, the image classification technology generally uses a neural network model (specifically an image classification model) to recognize images, and uses the recognition results as image labels. For example, the recognition result of an image can be one or more labels, that is, an image can have one o...

AI Technical Summary

Problems solved by technology

[0004] However, due to the limited computing power, memory and storage space of the terminal device, the storage space for storing the image classification model on the terminal device cannot be very large (generally less than 100 megabytes (MB)), so the computing power of the image classification model (ie Classification ability) is relatively limited, generally only supports hundreds of label categories, and the accuracy of image classification is relatively low

Images