Research on Vehicle Detection Model based on Attention Mechanism
Hai‐Tao Zhang, Jianmin Bao, Fei Ding, Guanyu Mi · 2021
With the continuous development of artificial intelligence technology, innovative car technology has become an important direction for a new round of technological changes in the driverless scene. As an essential part of uncrewed vehicles, vehicle detection is of great significance for improving the reliability and safety of intelligent vehicles in the unmanned environment. Therefore, achieving efficient and real-time vehicle detection has become one of the most popular research contents. However, traditional deep learning-based vehicle detection algorithms still have the problems of losing high-resolution features of the target object and insufficient feature fusion. Because of the above issues, this paper proposes a multiscale parallel vehicle detection model based on an attention mechanism. The algorithm adopts a multi-channel neural network structure. On this basis, a weighted feature fusion method is used. The weighted fusion is carried out according to the contribution of the multi-channel features to the network. The simulation results show that, compared with mainstream vehicle detection models, the model proposed in this paper achieves better detection results.