Abstract:With the rapid advancement of autonomous driving technology, efficient and accurate object detection capabilities have become crucial factors in ensuring the safety and reliability of autonomous driving systems. However, in low-visibility environments such as hazy conditions, the performance of traditional object detection algorithms often degrades significantly, failing to meet the demands of autonomous driving. To address this challenge, this paper proposes two innovative deep learning models: YOLO-Vehicle and YOLO-Vehicle-Pro. YOLO-Vehicle is an object detection model tailored specifically for autonomous driving scenarios, employing multimodal fusion techniques to combine image and textual information for object detection. YOLO-Vehicle-Pro builds upon this foundation by introducing an improved image dehazing algorithm, enhancing detection performance in low-visibility environments. In addition to model innovation, this paper also designs and implements a cloud-edge collaborative object detection system, deploying models on edge devices and offloading partial computational tasks to the cloud in complex situations. Experimental results demonstrate that on the KITTI dataset, the YOLO-Vehicle-v1s model achieved 92.1% accuracy while maintaining a detection speed of 226 FPS and an inference time of 12ms, meeting the real-time requirements of autonomous driving. When processing hazy images, the YOLO-Vehicle-Pro model achieved a high accuracy of 82.3% mAP@50 on the Foggy Cityscapes dataset while maintaining a detection speed of 43 FPS.

What problem does this paper attempt to address?

This paper attempts to solve the problem that the performance of existing object detection algorithms drops significantly in autonomous driving scenarios, especially in low - visibility environments (such as hazy weather). Specifically, the paper focuses on how to improve the accuracy and real - time performance of object detection to meet the requirements of autonomous driving systems under various weather conditions. To address this challenge, the paper proposes two innovative deep - learning models: YOLO - Vehicle and YOLO - Vehicle - Pro. YOLO - Vehicle is an object detection model specifically designed for autonomous driving scenarios. It adopts multi - modal fusion technology and combines image and text information for object detection. YOLO - Vehicle - Pro, on this basis, introduces an improved image defogging algorithm, which improves the detection performance in low - visibility environments. In addition, the paper also designs and implements a cloud - edge collaborative object detection system. By deploying the model on edge devices and offloading some computing tasks to the cloud in complex situations, an efficient and flexible computing architecture is achieved. The experimental results show that the YOLO - Vehicle - v1s model achieves an accuracy of 92.1% on the KITTI dataset, while maintaining a detection speed of 226 FPS and an inference time of 12 ms, meeting the real - time requirements of autonomous driving. When processing hazy images, the YOLO - Vehicle - Pro model achieves an mAP @50 accuracy of 82.3% on the Foggy Cityscapes dataset, while maintaining a detection speed of 43 FPS. In general, the main contributions of the paper are as follows: 1. Proposing a new object detector, YOLO - Vehicle, which includes an image processing module and a text processing module, can effectively handle vehicle targets of different sizes and distances, and enhances the understanding of complex traffic scenes through regional text feature extraction. 2. Proposing YOLO - Vehicle - Pro, an advanced version of YOLO - Vehicle, which is especially suitable for hazy driving scenarios. By introducing an improved image defogging algorithm and an adaptive feature extraction mechanism, the detection performance in low - visibility environments is significantly improved. 3. Implementing a cloud - edge collaborative object detection system, which can realize real - time image collection and preliminary processing on edge devices, and offload some computing tasks to the cloud when encountering hazy weather, thereby optimizing the utilization of computing resources and the response speed of the system.

YOLO-Vehicle-Pro: A Cloud-Edge Collaborative Framework for Object Detection in Autonomous Driving under Adverse Weather Conditions

Research on Autonomous Driving Image Recognition Based on a New Real-Time Object Detection Model YOLOv5st

DAN-YOLO: A Lightweight and Accurate Object Detector Using Dilated Aggregation Network for Autonomous Driving

Edge YOLO: Real-Time Intelligent Object Detection System Based on Edge-Cloud Cooperation in Autonomous Vehicles

YOLOv8n-FAWL: Object Detection for Autonomous Driving Using YOLOv8 Network on Edge Devices

YOLOv8-QSD: An Improved Small Object Detection Algorithm for Autonomous Vehicles Based on YOLOv8

MobileYOLO: Real-Time Object Detection Algorithm in Autonomous Driving Scenarios

YOLOv4-5D: An Effective and Efficient Object Detector for Autonomous Driving

Detection and Recognition for Autonomous Driving Based on Improved YOLO Algorithm

AIE-YOLO: Effective object detection method in extreme driving scenarios via adaptive image enhancement

Lightweight Object Detection Based on Autonomous Driving

YOLOv3-MT: A YOLOv3 using multi-target tracking for vehicle visual detection

YOLOv5s-Fog: An Improved Model Based on YOLOv5s for Object Detection in Foggy Weather Scenarios

A Real-Time Object Detector for Autonomous Vehicles Based on YOLOv4

DCW-YOLO: Road Object Detection Algorithms for Autonomous Driving

Analyzing Performance of YOLOx for Detecting Vehicles in Bad Weather Conditions

YOLOv5-Fog: A Multiobjective Visual Detection Algorithm for Fog Driving Scenes Based on Improved YOLOv5

Real-time vehicle target detection in inclement weather conditions based on YOLOv4

An improved YOLOv8 algorithm for small object detection in autonomous driving

Defog YOLO for road object detection in foggy weather