End-to-end semi-supervised target detection method based on improved yolov5

Abstract

The invention discloses an end-to-end semi-supervised target detection method based on improved yolov5, relates to the technical field of target detection, and solves the technical problems that the existing target detection is relatively complex and the precision improvement of a model is limited by a pseudo tag. The labeled data and the unlabeled data are loaded at the same time in the training process; taking the model updated by using the ema algorithm as a teacher model, and generating a pseudo tag by using a dynamic threshold mode; the loss of the yolov5 is modified, the loss is divided into supervised loss and unsupervised loss, and the unsupervised loss part is optimized by using a difficult case mining strategy. The whole process is simpler, the pseudo labels can be more and more accurate in the training process, the detector is helped to achieve better performance, the reasoning speed is higher, and the real-time requirement can be met.

Description

technical field [0001] The present application relates to the technical field of target detection, in particular to an end-to-end semi-supervised target detection method based on improved yolov5. Background technique [0002] Most of the current mainstream semi-supervised target detection methods are multi-stage. Multi-stage refers to first using labeled data to train an initial fully supervised model, and then processing unlabeled data through this model to obtain high-confidence results as unspecified models. Pseudo-labels of labeled data, and finally input labeled data and unlabeled data with pseudo-labels to the model for training. The whole process is relatively complicated, and the pseudo-label is generated offline, that is, it will not be updated during the training process. A problem with such a model is that when the accuracy of the trained model gradually improves and exceeds the original model, continuing to use the pseudo-labels generated by the original model w...

AI Technical Summary

Problems solved by technology

The whole process is more complicated, and the pseudo-label is generated offline, that is, it will not be updated during the training process

A problem with such a model is that when the accuracy of the trained model gradually improves and exceeds the original model, continuing to use the pseudo-labels generated by the original model will limit the further improvement of the model accuracy; and most of the current semi-supervised target detection methods are based on Faster- RCNN, which is a two-stage target detection method, has the problem of slow detection speed and cannot meet the real-time requirements

Images