A Faster and Lightweight Lane Detection Method in Complex Scenarios

Shuaiqi Nie,Guiheng Zhang,Libo Yun,Shuxian Liu
DOI: https://doi.org/10.3390/electronics13132486
IF: 2.9
2024-06-26
Electronics
Abstract:Lane detection is a crucial visual perception task in the field of autonomous driving, serving as one of the core modules in advanced driver assistance systems (ADASs).To address the insufficient real-time performance of current segmentation-based models and the conflict between the demand for high inference speed and the excessive parameters in resource-constrained edge devices (such as onboard hardware, mobile terminals, etc.) in complex real-world scenarios, this paper proposes an efficient and lightweight auxiliary branch network (CBGA-Auxiliary) to tackle these issues. Firstly, to enhance the model's capability to extract feature information in complex scenarios, a row anchor-based feature extraction method based on global features was adopted. Secondly, employing ResNet as the backbone network and CBGA (Conv-Bn-GELU-SE Attention) as the fundamental module, we formed the auxiliary segmentation network, significantly enhancing the segmentation training speed of the model. Additionally, we replaced the standard convolutions in the branch network with lightweight GhostConv convolutions. This reduced the parameters and computational complexity while maintaining accuracy. Finally, an additional enhanced structural loss function was introduced to compensate for the structural defect loss issue inherent in the row anchor-based method, further improving the detection accuracy. The model underwent extensive experimentation on the Tusimple dataset and the CULane dataset, which encompass various road scenarios. The experimental results indicate that the model achieved the highest F1 scores of 96.1% and 71.0% on the Tusimple and CULane datasets, respectively. At a resolution of 288 × 800, the ResNet18 and ResNet34 models achieved maximum inference speeds of 410FPS and 280FPS, respectively. Compared to existing SOTA models, it demonstrates a significant advantage in terms of inference speed. The model achieved a good balance between accuracy and inference speed, making it suitable for deployment on edge devices and validates the effectiveness of the model.
engineering, electrical & electronic,computer science, information systems,physics, applied
What problem does this paper attempt to address?
The problems that this paper attempts to solve mainly focus on the insufficient real - time performance of lane detection models in complex scenarios and the contradiction between the high inference speed requirements and excessive parameters in resource - constrained edge devices (such as in - vehicle hardware, mobile terminals, etc.). Specifically: 1. **Challenges of Lane Detection in Complex Scenarios**: - In complex scenarios such as strong light irradiation, poor light at night, sight blocked by obstacles, and blurred lane lines, existing deep - learning lane detection models are prone to losing semantic information, resulting in poor detection performance. 2. **Requirements for Real - Time and Lightweight Design**: - When applying lane detection models on resource - constrained edge devices, there are strict requirements for real - time performance and lightweight design. Many existing state - of - the - art (SOTA) models, although having high precision, have a large number of parameters and long inference times, and it is difficult to meet the requirements of real - time and lightweight simultaneously. To solve these problems, this paper proposes an efficient and lightweight auxiliary branch network (CBGA - Auxiliary), aiming to enhance the feature extraction ability of the model in complex scenarios, increase the segmentation training speed, and reduce the parameters and computational complexity by introducing lightweight GhostConv convolution while maintaining accuracy. In addition, an additional structural loss function is introduced to compensate for the structural defect problems existing in the row - anchor - based method, further improving the detection accuracy. ### Summary of Main Contributions: 1. **Proposing an Efficient and Lightweight Auxiliary Segmentation Network (CBGA - Auxiliary)**: This network shares parameters and weights with the backbone network and collaboratively completes the segmentation task during the training phase, improving the overall generalization ability and inference speed. 2. **Replacing Standard Convolution with Lightweight GhostConv Convolution**: Reducing the number of parameters and computational complexity of the model while improving accuracy and inference speed. 3. **Introducing an Additional Structural Loss Function**: Aiming at the structural defects of the row - anchor - based method, the similarity loss between lanes is added, further improving the model's detection accuracy. 4. **Experimental Verification of the Model's Effectiveness**: Extensive experiments on the Tusimple and CULane datasets show that this model achieves good detection accuracy while maintaining a high inference speed, reaching a good balance between accuracy and inference speed. These improvements make this model more suitable for deployment on edge devices and verify its effectiveness in practical applications.