Squeeze-EnGAN: Memory Efficient and Unsupervised Low-Light Image Enhancement for Intelligent Vehicles
Open Access
Abstract
Intelligent vehicles, such as autonomous cars, drones, and robots, rely on sensors to gather environmental information and respond accordingly. RGB cameras are commonly used due to their low cost and high resolution but are limited in low-light conditions. While employing LiDAR or specialized cameras can address this issue, these solutions often incur high costs. Deep learning-based low-light image enhancement (LLIE) methods offer an alternative, but existing models struggle to adapt to road scenes. Furthermore, most LLIE models rely on supervised training but are heavily constrained by the lack of low-light and normal-light paired datasets. In particular, obtaining paired datasets for driving scenes is extremely challenging. To address these issues, this paper proposes Squeeze-EnGAN, a memory-efficient, GAN-based LLIE method capable of unsupervised learning without paired image datasets. Squeeze-EnGAN incorporates a fire module into a U-net architecture, substantially reducing the number of parameters and Multiply-Accumulate Operations (MACs) compared to its base model, EnlightenGAN. Additionally, Squeeze-EnGAN achieves real-time performance on devices like Jetson Xavier (0.061 s). Significantly, enhanced images improve object detection performance over original images, demonstrating the model’s potential to aid high-level vision tasks in intelligent vehicles.Keywords
Funding Information
- Korea Institute for Advancement of Technology (P0020536)
This publication has 40 references indexed in Scilit:
- Panoptic-FusionNet: Camera-LiDAR fusion-based point cloud panoptic segmentation for autonomous drivingExpert Systems with Applications, 2024
- Continual Learning for Real-World Autonomous Systems: Algorithms, Challenges and FrameworksJournal of Intelligent & Robotic Systems, 2022
- Research on Comparison of LiDAR and Camera in Autonomous DrivingJournal of Physics: Conference Series, 2021
- Computing Systems for Autonomous Driving: State of the Art and ChallengesIEEE Internet of Things Journal, 2021
- Making of Night Vision: Object Detection Under Low-IlluminationIEEE Access, 2020
- DN-GAN: Denoising generative adversarial networks for speckle noise reduction in optical coherence tomography imagesBiomedical Signal Processing and Control, 2020
- Edge Computing for Autonomous Driving: Opportunities and ChallengesProceedings of the IEEE, 2019
- LLNet: A deep autoencoder approach to natural low-light image enhancementPattern Recognition, 2017
- Perception, information processing and modeling: Critical stages for autonomous driving applicationsAnnual Reviews in Control, 2017
- Adaptive histogram equalization and its variationsComputer Vision, Graphics, and Image Processing, 1987