A vehicle detection algorithm is an indispensable component of intelligent traffic management and control systems, influencing the efficiency and functionality of the system. In this paper, we propose a lightweight improvement method for the YOLOv5 algorithm based on integrated perceptual attention, with few parameters and high detection accuracy. First, we propose a lightweight module IPA with a Transformer encoder based on integrated perceptual attention, which leads to a reduction in the number of parameters while capturing global dependencies for richer contextual information. Second, we propose a lightweight and efficient multiscale spatial channel reconstruction (MSCCR) module that does not increase parameter and computational complexity and facilitates representative feature learning. Finally, we incorporate the IPA module and the MSCCR module into the YOLOv5s backbone network to reduce model parameters and improve accuracy. The test results show that, compared with the original model, the model parameters decrease by about 9%, the average accuracy (mAP@50) increases by 3.1%, and the FLOPS does not increase.

Vehicle detection algorithms are essential for intelligent traffic management and autonomous driving systems. Current vehicle detection algorithms largely rely on deep learning techniques, enabling the automatic extraction of vehicle image features through convolutional neural networks (CNNs). However, in real traffic scenarios, relying only on a single feature extraction unit makes it difficult to fully understand the vehicle information in the traffic scenario, thus affecting the vehicle detection effect. To address this issue, we propose a lightweight vehicle detection algorithm based on Mamba_ViT. First, we introduce a new feature extraction architecture (Mamba_ViT) that separates shallow and deep features and processes them independently to obtain a more complete contextual representation, ensuring comprehensive and accurate feature extraction. Additionally, a multi-scale feature fusion mechanism is employed to enhance the integration of shallow and deep features, leading to the development of a vehicle detection algorithm named Mamba_ViT_YOLO. The experimental results on the UA-DETRAC dataset show that our proposed algorithm improves mAP@50 by 3.2% compared to the latest YOLOv8 algorithm, while using only 60% of the model parameters.

Vehicle detection algorithms are integral to intelligent traffic management and AI-assisted driving systems. However, the complexity and variability of traffic scenarios present significant challenges. Spatial pyramid pooling enhances a model’s ability to capture multi-scale contextual information by offering different levels of feature representation, which is particularly beneficial in complex traffic scenarios. In this study, we extend spatial pyramid pooling to Transformer-based models and introduce a linear spatial pyramid attention (LSPA) mechanism. This approach allows the model to focus on both global information and local details when processing vehicle image data. Additionally, we propose a spatial coordination filter (SCF) module to explicitly summarize spatial global dependencies and reduce spatial feature redundancy. To address computational costs, we employ partial convolution (PConv) in the model’s backbone network to minimize redundant computation. Experiments conducted on the BDD100K and KITTI datasets demonstrate that our model improves the mean average precision (mAP@50) by 2.7% and reduces model parameters by approximately 16% compared to the existing RT-DETR model. These findings confirm the efficacy and advantages of our proposed method.