Image classification method based on semi-supervised self-paced learning cross-task deep network

Abstract

The invention discloses an image classification method based on a semi-supervised self-paced learning cross-task deep network. The method includes the steps of randomly selecting a small amount of labeling samples from the whole image data set, reserving the labels, and remaining all the samples as unlabelled samples having the real labels to be unknown in the whole process, wherein the weight ofthe labeled samples is constant to be one in the training process, the weight of the unlabelled samples is initialized to be zero, and only the labeled samples are used as a training set in the initial process; S2, training a cross-task deep network by the training set; S3, according to the trained cross-task deep network, predicting the pseudo labels of all the unlabelled samples, and giving a corresponding weight of each unlabelled sample; S4, according to a self-paced learning normal form, selecting an unlabelled sample with a high confidence degree, and adding to the training set; and S5,repeating the steps S2-S4 until the cross-task deep network performance is saturated or reaches a preset cycle number. According to the method, the human design feature is not needed to be input, andthe classification can be realized by directly inputting the original image.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to an image classification method based on a semi-supervised self-paced learning cross-task deep network. Background technique [0002] Image classification has been a challenging task in the field of computer vision in the past few decades, because the category information of images reflects human's high-level semantic cognition of these images. The traditional method is generally to extract some low-level features from the picture, and according to the label of the picture, supervised training to obtain a model to predict the image label. However, with the development of the mobile Internet world, the number of pictures is increasing rapidly every day. Labeling these images has become a very labor-intensive and time-consuming task. Therefore, how to reduce the workload of manual labeling as much as possible while maintaining the performance of the classifier is of great significan...

AI Technical Summary

Problems solved by technology

Most of the traditional image classification methods are shallow models, which cannot handle large-scale data sets. However, for semi-supervised methods, we often only have a small part of training data at the beginning, and it is easy to cause network overfitting during training.

Images