VELIE: A Vehicle-Based Efficient Low-Light Image Enhancement Method for Intelligent Vehicles

Abstract

In Advanced Driving Assistance Systems (ADAS), Automated Driving Systems (ADS), and Driver Assistance Systems (DAS), RGB camera sensors are extensively utilized for object detection, semantic segmentation, and object tracking. Despite their popularity due to low costs, RGB cameras exhibit weak robustness in complex environments, particularly underperforming in low-light conditions, which raises a significant concern. To address these challenges, multi-sensor fusion systems or specialized low-light cameras have been proposed, but their high costs render them unsuitable for widespread deployment. On the other hand, improvements in post-processing algorithms offer a more economical and effective solution. However, current research in low-light image enhancement still shows substantial gaps in detail enhancement on nighttime driving datasets and is characterized by high deployment costs, failing to achieve real-time inference and edge deployment. Therefore, this paper leverages the Swin Vision Transformer combined with a gamma transformation integrated U-Net for the decoupled enhancement of initial low-light inputs, proposing a deep learning enhancement network named Vehicle-based Efficient Low-light Image Enhancement (VELIE). VELIE achieves state-of-the-art performance on various driving datasets with a processing time of only 0.19 s, significantly enhancing high-dimensional environmental perception tasks in low-light conditions.

Optimized Design of EdgeBoard Intelligent Vehicle Based on PP-YOLOE+

“…Addressing traffic congestion in Macau, Lam et al [26] developed a low-cost, real-time traffic monitoring system using free online images, YOLOv3, and the mIOU algorithm, which showed high accuracy and adaptability in diverse conditions. Ye [27] presents the vehicle-based efficient low-light image enhancement (VELIE) network, utilizing the Swin Vision Transformer combined with a gamma transformation enhanced U-Net. This approach aims to improve low-light images, overcoming the challenges faced by RGB cameras in advanced driving assistance systems (ADAS).…”

“…Advantages Disadvantages CNN-SSD [13] Introduce variability convolution Complexity of degree calculation YOLOv4 [23] Hollow convolution; ULSAM; soft-NMS High computational resource YOLO [24] Combining R-FCN and histograms Low parameter detection accuracy AP-SSD [25] Gabor feature extraction; SSD enhancement Computational complexity YOLOv3 [26] Lightweight object detection framework Poor visual effect MAP [14] Fading memory estimation Low robustness complexity CNN [15] Multiclass object detection classifier Low detection rate VELIE [27] Combining the integrated U-Net of Swin Vision Transformer and gamma transform Gaps in detail enhancement IDOD-YOLOv7 [28] Combined AOD and SAIP; high accuracy Poor practice results Range-layer CNN [16] High detection speed and low cost Lack safety and reliability in autonomous driving EYOLOv3 [29] Kalman filter and particle filter; high efficiency Large amount of data SSD [30] Structure, training method, and loss function Suboptimal detection performance…”