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

Liang, Siyuan; Wu, Hao; Zhen, Li; Hua, Qiaozhi; Garg, Sahil; Kaddoum, Georges; Hassan, Mohammad Mehedi; Yu, Keping

doi:10.1109/tits.2022.3158253

Cited by 110 publications

(50 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is an interface with graphical components, allowing users to communicate with electronic devices such as telematics units and infotainment displays. Here, the YOLO algorithm is employed for vehicle detection and counting [47][48][49]. The YOLO algorithm must run on CPU-and GPU-based systems like NVIDIA boards.…”

Section: Validation and Resultsmentioning

confidence: 99%

EVATL: A novel framework for emergency vehicle communication with adaptive traffic lights for smart cities

Dodia,

Kumar,

Rani

et al. 2023

IET Smart Cities

View full text Add to dashboard Cite

Fixed cycle traffic lights primarily regulate road traffic, in which traffic light control systems are for specific lanes or crossings in urban areas. Also, not being appropriately installed can prolong the congestion delay and unnecessarily long wait times for crossing intersections, which can cause emergency vehicles to become stuck at intersections. Adaptive signal timing management technique that is more computationally viable than current fixed cycle signal control systems and can improve network‐wide traffic operations by reducing traffic delay and energy consumption. Even though specific adaptive control systems exist, there is no mechanism to communicate with emergency vehicles, which is crucial for smart cities. Motivated by this problem, a novel framework, Emergency Vehicle Adaptive Traffic Light (EVATL), is proposed for smart cities where an adaptive mode of operation for traffic lights is employed with emergency vehicle communication, improving their functioning and reducing overall congestion delay. EVATL detects emergency vehicle location using GPS with the Internet of Things(IoT), which integrates with traffic signals and works adaptively according to vehicle density at the traffic signal using YOLOv8. So, the primary goal of the proposed EVATL is to prioritise an emergency vehicle while simultaneously integrating adaptive traffic signals for smart cities. A GUI is developed for evaluating the proposed model by creating different scenarios for an adaptive traffic light and emergency vehicle communication. While analysing the simulation results of the proposed model EVATL, a clear improvement can be seen in the wait time of vehicles at a traffic light with the timely detection of an emergency vehicle at a set distance.

show abstract

Section: Validation and Resultsmentioning

confidence: 99%

EVATL: A novel framework for emergency vehicle communication with adaptive traffic lights for smart cities

Dodia,

Kumar,

Rani

et al. 2023

IET Smart Cities

View full text Add to dashboard Cite

show abstract

“…In such scenarios, deploying lightweight models capable of making inferences on edge devices is desirable. The YOLO family has also demonstrated its potential for edge computing applications [ 27 , 28 ].…”

Section: Proposalmentioning

confidence: 99%

POSEIDON: A Data Augmentation Tool for Small Object Detection Datasets in Maritime Environments

Ruiz-Ponce

Ortiz-Perez

García-Rodríguez

et al. 2023

Sensors

View full text Add to dashboard Cite

Certain fields present significant challenges when attempting to train complex Deep Learning architectures, particularly when the available datasets are limited and imbalanced. Real-time object detection in maritime environments using aerial images is a notable example. Although SeaDronesSee is the most extensive and complete dataset for this task, it suffers from significant class imbalance. To address this issue, we present POSEIDON, a data augmentation tool specifically designed for object detection datasets. Our approach generates new training samples by combining objects and samples from the original training set while utilizing the image metadata to make informed decisions. We evaluate our method using YOLOv5 and YOLOv8 and demonstrate its superiority over other balancing techniques, such as error weighting, by an overall improvement of 2.33% and 4.6%, respectively.

show abstract

“…The YOLO algorithm has successfully struck an optimal balance between real-time performance, speed, and accuracy, positioning it as an excellent candidate for real-time detection applications 6 . YOLOv5 has demonstrated efficacy in tasks such as weld surface inspection and 7 surface defect detection 8 .…”

Section: Introductionmentioning

confidence: 99%

“…5 The YOLO algorithm has successfully struck an optimal balance between real-time performance, speed, and accuracy, positioning it as an excellent candidate for real-time detection applications. 6 YOLOv5 has demonstrated efficacy in tasks such as weld surface inspection and 7 surface defect detection. 8 YOLOv7, an advanced iteration of the YOLO algorithm, boasts impressive accuracy and detection speed, 9 particularly well suited for engineering applications.…”

Section: Introductionmentioning

confidence: 99%

Lightweight target detection algorithm based on partial convolution

Chen,

Liu

2024

J. Electron. Imag.

View full text Add to dashboard Cite

You only look once (YOLO) target detection algorithms have garnered substantial attention in engineering inspection tasks due to their remarkable accuracy and swift detection speed. However, adapting the YOLOv7 algorithm to engineering construction presents challenges in balancing lightweight design and high precision. This work introduces a YOLOv7 lightweight detection algorithm, leveraging partial convolution to efficiently reduce network parameters without compromising detection speed. Integrating partial convolution into the network's backbone is a highly effective parameter reduction strategy. Additionally, we propose the SE_Conv module following quantitative analyses through experiments, thereby significantly enhancing the algorithm's detection accuracy. To address the impact of low-quality data on detection effectiveness in engineering tasks, the advanced bounding box loss function WIoU is employed. In a practical application scenario involving helmet detection, the algorithm achieves a notable 23.4% reduction in parameters and a substantial 28.2% decrease in operations through partial convolution. Subsequent experiments with the SE_Conv module yield a noteworthy 1.4% improvement in results. Significantly, in head detection, our algorithm surpasses YOLOv8 algorithm by 2.6% and single shot multibox detector algorithm by 9.4%. These findings underscore the efficacy of our proposed algorithm in engineering applications.

show abstract

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

Cited by 110 publications

References 49 publications

EVATL: A novel framework for emergency vehicle communication with adaptive traffic lights for smart cities

EVATL: A novel framework for emergency vehicle communication with adaptive traffic lights for smart cities

POSEIDON: A Data Augmentation Tool for Small Object Detection Datasets in Maritime Environments

Lightweight target detection algorithm based on partial convolution

Contact Info

Product

Resources

About