Image multi-label classification method, system and device and readable storage medium

Abstract

The invention belongs to the field of image classification, and discloses an image multi-label classification method, system and device, and a readable storage medium. The method comprises the steps:S1, obtaining original images corresponding to multiple labels to be classified, and generating a sample set; s2, obtaining an initial deep convolutional neural network for image multi-label classification, and adding a gate activation function layer to obtain a deep convolutional neural network; s3, iteratively training a deep convolutional neural network through the sample set to obtain a saliency map corresponding to each label; s4, selecting a saliency map corresponding to a label with a preset requirement, and extracting a target from the corresponding original image through the saliencymap; s5, setting a to-be-generated image, and migrating the target area to the to-be-generated image to obtain a target image; s6, adding the target image into the sample set, iterating S3 to S5 for preset times, and performing image multi-label classification of the input image through the final deep convolutional neural network. According to the method, the precision and the accuracy of image multi-label classification are effectively improved.

Description

technical field [0001] The invention belongs to the field of image classification, and relates to an image multi-label classification method, system, equipment and readable storage medium. Background technique [0002] Image multi-label classification has always been a hot issue in the field of computer vision. With the rise and development of artificial intelligence technology, image multi-label classification methods based on deep learning, especially image multi-label classification using deep convolutional neural networks, have made significant progress. A breakthrough, its classification accuracy far exceeds the traditional machine learning method. However, due to the huge amount of model parameters based on deep convolutional neural networks, the results generally lack interpretability, which makes the results of image multi-label classification based on this method unreliable, which has become a bottleneck for the further development and application of deep learning. ...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the shortcomings of insufficient high precision and low accuracy of image multi-label cl

Images