Vehicle identification method based on deep learning and reinforcement learning

Abstract

The present invention discloses a vehicle identification method based on deep learning and reinforcement learning. On the structure features of the deep network, a deep learning and reinforcement learning combination method is provided, the Q learning algorithm in the reinforcement learning is applied to the deep learning network, and the training process still uses a random gradient descent algorithm so as to improve the vehicle identification capability of the depth network; and the reinforcement learning technology based on the missed classed sample learning is added so as to overcome the current technology deficiency of the deep learning network in the vehicle identification field and improve the network training efficiency while improving the vehicle identification performance.

Description

technical field [0001] The invention belongs to the technical field of pattern recognition, and more specifically, relates to a vehicle recognition method based on deep learning and reinforcement learning. Background technique [0002] Vehicle recognition is an important topic in the field of intelligent transportation. The main difficulty in designing a reliable vehicle recognition system lies in the differences between vehicles. Compared with the traditional method, the deep learning network has stronger recognition robustness to the partial occlusion of the vehicle due to the simulation of the human brain. [0003] In the prior art, the common method is to extract abstract features through the deep structure, and then identify the vehicle through the abstract features. However, in the training process of the deep learning network, there are often the following problems: Compared with training the traditional three-layer neural network, the deep learning network has an in...

AI Technical Summary

Problems solved by technology

However, in the training process of the deep learning network, there are often the following problems: Compared with training the traditional three-layer neural network, the deep learning network has an increased time cost due to the large amount of calculation and more parameters that need to be updated.

Secondly, when the output error of the deep learning network does not change much, the training process will slow down and take too long

Images