YOLOv5-LC: Enhancing Vehicle Detection for Evening Rushing Hour

Abstract

In order to address the challenges of low lighting and overlapping vehicles in the detection of vehicles during the evening rush hour, this article proposes a new model YOLOv5-LC based on YOLOv5s. First, to solve the problem caused by the insufficient ambient light in the detection process, CycleGAN is introduced to perform feature domain swapping between the source domain and target domain. The conversion of object features in images is realized by image panning of unpaired images, successfully converting some daytime images in the data set into nighttime images. This enables the model to learn richer feature representations of low-light conditions, thereby improving the model’s ability to detect vehicles under low-light conditions and, in practical applications, reducing the cost of manually annotating nighttime data sets. Second, to improve the model’s ability to detect vehicles under overlapping conditions in the image, the article uses the CrowdDet-V algorithm to supplement the anchor-based object detector. The enhanced detector uses a candidate anchor frame to generate multiple presets to improve the detection ability of the model under the condition of overlapping vehicles in the image, which enables the model to better detect highly overlapping instances in crowded scenes. Finally, the performance of CycleGAN in this task is evaluated qualitatively through image comparisons and quantitatively using grayscale histograms. The proposed model is quantitatively validated on the private COTRS data set for crowded vehicle counting, achieving a 4.3% increase in vehicle detection accuracy compared to YOLOv5s at the cost of little time. The experimental results show that our proposed method can improve the accuracy of congested vehicle detection at night, effectively resolving the challenges of vehicle detection during evening rush hour. On the relatively sparse UA-DETRAC data set, our approach can still achieve moderate improvement, suggesting that the proposed method is robust to different levels of congestion. The model proposed in this paper, YOLOv5-LC, is a vehicle detection model designed for full-time operation based on roadside surveillance videos. It can be deployed in the cloud and enables real-time vehicle detection by processing surveillance videos obtained from roadside cameras, thereby providing traffic flow data for traffic administrative departments or transmitting through vehicle-to-infrastructure (V2I) communication systems to connected and autonomous vehicles (CAVs). The model demonstrates superior performance in congested and low-light road scenarios, particularly addressing the significant decline in vehicle detection capability observed in existing models during the evening rush hour, which represents the peak period for vehicular traffic throughout the day. The output of the model will be further used to calculate fundamental parameters of traffic flow such as vehicle flow, density, and speed. (1) For traffic management, these real-time parameters facilitate dynamic traffic management and control, such as intelligent real-time adjustment of signal light timings and manual traffic diversion to alleviate congestion. (2) For CAVs, through V2I technology, the real-time detected traffic flow data can be transmitted to CAVs via vehicle-to-infrastructure communication systems, helping CAVs select the optimal driving routes and adjust driving speeds to adapt to real-time traffic conditions.