Research on safety helmet detection method based on convolutional neural network

Li, Qiong; Fulun, Peng; Ru, Zhibin; Yu, Shuai; Zhao, Qinglin; Shang, Qiongjun; Cao, Yue; Liu, Jie

doi:10.1117/12.2564896

Cited by 4 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…P. Sridhar and colleagues (2022) [15] have proposed a model which is mainly trained with Yolov2, they have used their custom dataset in order to obtain exceptional results and their model achieve good accuracy at detecting motorbike riders with and without helmet. YueJing Qian and team (2023) [16] have proposed an approach to optimize the BottleneckCSP structure in yolov5 backbone network, they mainly aimed to propose a robust sytem to detect helmets which turned out to reduce the complexity of model with no changes in size of inputs and output, their method is proved to be better compared with existing methods with fastest interference speed. A. Vandana Peter and colleagues (2023) [17] have proposed an application to detect helmets and read license plates using deep learning algorithms, they have used Yolov3 to detect motorcycles and another CNN to determine whether bicyclist wears helmet or not, TesseractOCR is used to read the license plate number of motorcycles which are already captured using Yolov3.…”

Section: Literature Reviewmentioning

confidence: 99%

Automated Helmet Monitoring System Using Deep Learning

Kavuri.K.S.V.A.Satheesh,

Nandam Sai Akhila,

Dondapati Amarnadh

et al. 2024

EPRA

View full text Add to dashboard Cite

Safety and compliance are the uppermost and fundamental concerns in the modern transport subsystems. As a result, the project is essentially designed to come up with an advanced solution by combining YOLOv8 for precise identification of objects and, on the other hand, Easy OCR for reading characters. The key goals are to detect helmets, those without helmets, and identify number plates of the respective motor vehicle. With YOLOv8, we start training the model to identify not only helmets but the lack of helmets, which is necessary for compliance monitoring based on the law. Further, YOLOv8 is also designed to determine the Regions of Interest . Regarding vehicles, the model focuses mainly on license plates which are key objects. After finding the appropriate areas, Easy OCR is designed for applying optical character recognition, helping to obtain vehicle numbers of any type in the most organized, quick way. Therefore, combining YOLOv8 at the stage of object detection and Easy OCR for the recognition of characters creates a novel but, at the same time susceptible system for a vehicle control company. This integrated system is a sophisticated device for monitoring helmeted and un helmeted riders, promoting a safe and stable journey gadget. By leveraging real-time records, our answers provide precious insights into protection compliance. In summary, the aggregate of YOLOv8 and Easy OCR presents a effective answer for item popularity and conduct reputation, so that our system contributes to the development of secure and green urban mobility by means of preserving rider protection and safety. s. Index Terms - Helmet, Deep Learning, Object Detection, Character Recognition, ROI

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Automated Helmet Monitoring System Using Deep Learning

Kavuri.K.S.V.A.Satheesh,

Nandam Sai Akhila,

Dondapati Amarnadh

et al. 2024

EPRA

View full text Add to dashboard Cite

show abstract

“…Their objective was to achieve high accuracy in detecting helmets at construction sites, even under low-light conditions. Li et al [20] proposed a safety helmet detection method based on a deep convolutional network. Their approach involved decoding video monitoring data, extracting YUV images, and applying a carefully designed convolutional neural network model for detection purposes.…”

Section: Related Workmentioning

confidence: 99%

“…Li et al [20] This method consists of multiple stages and has high algorithmic complexity. It does not prioritize real-time requirements in its design.…”

Section: Hayat Et Al [19]mentioning

confidence: 99%

An Improved YOLOv5s-Based Helmet Recognition Method for Electric Bikes

Huang,

Wu,

Xiang

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

This paper proposes an improved model based on YOLOv5s, specifically designed to overcome the challenges faced by current target detection algorithms in the field of electric bike helmet detection. In order to enhance the model’s ability to detect small targets and densely populated scenes, a specialized layer dedicated to small target detection and a novel loss function called Normalized Wasserstein Distance (NWD) are introduced. In order to solve the problem of increasing model parameters and complexity due to the inclusion of a small target detection layer, a Cross-Stage Partial Channel Mixing (CSPCM) on top of Convmix is designed. The collaborative fusion of CSPCM and the Deep Feature Consistency (DFC) attention mechanism makes it more suitable for hardware devices. In addition, the conventional Nearest Upsample technology is replaced with the advanced CARAFE Upsample module, further improving the accuracy of the model. Through rigorous experiments on carefully constructed datasets, the results show significant improvements in various evaluation indicators such as precision, recall, mAP.5, and mAP.95. Compared with the unmodified YOLOv5s algorithm, the proposed enhanced model achieves significant improvements of 1.1%, 8.4%, 5.2%, and 8.6% on these indicators, respectively, and these enhancements are accompanied by a reduction of 778,924 parameters. The experimental results on our constructed dataset demonstrate the superiority of the improved model and elucidate its potential applications. Furthermore, promising improvements for future research are suggested. This study introduces an efficient approach for improving the detection of electric bike helmets and verifies the effectiveness and practicality of the model through experiments. Importantly, the proposed scheme has implications for other target detection algorithms, especially in the field of small target detection.

show abstract

“…According to recent reports of mine safety accidents, occasional casualties caused by not wearing helmets have occurred due to a lack of awareness of safety precautions. 1 Therefore, to protect the life safety of mine personnel, mine units must supervise them wearing helmets. In early supervision tasks, manual supervision was often used to determine whether mine personnel were wearing helmets, which was prone to safety hazards and added high labor costs due to the subjectivity of the supervisors and the difficulty in understanding the mine conditions.…”

Section: Introductionmentioning

confidence: 99%

FM-YOLOv7: an improved detection method for mine personnel helmet

Shao

Liu

et al. 2023

J. Electron. Imag.

View full text Add to dashboard Cite

.Due to the complex underground environment and the small object of the helmet, the detection accuracy is low when the original YOLOv7 algorithm is used to detect whether the mine personnel wears the helmet, which cannot be applied to the actual operation site. In response to this problem, we proposed an FM-YOLOv7 mine personnel helmet detection. First, to improve the feature extraction ability of the shallow network and enhance the representation ability of the model on the helmet, we propose the fused-MBCA (fused-MBConv with the coordinate attention) module. Second, to improve the detection ability of small objects, enable the fused features to obtain high-level semantic information and low-level details from different scales, and have more extensive receptive fields, we propose the multi-scale feature fusion efficient layer aggregation networks. Finally, to accelerate the convergence of the model and improve the regression accuracy, we use efficient intersection over union as the bounding box regression loss function. These experiments are based on the self-built mine personnel safety helmet dataset. The results show that the FM-YOLOv7 model outperformed the other six algorithms. The mAP@0.5 of the proposed model can reach 85.7%, which is 1.4% higher than the original YOLOv7 model. Also, the improved YOLOv7 model achieves 91 frames per second in detection speed, which detects whether the mine personnel wears a safety helmet in real time.

show abstract

Research on safety helmet detection method based on convolutional neural network

Cited by 4 publications

References 0 publications

Automated Helmet Monitoring System Using Deep Learning

Automated Helmet Monitoring System Using Deep Learning

An Improved YOLOv5s-Based Helmet Recognition Method for Electric Bikes

FM-YOLOv7: an improved detection method for mine personnel helmet

Contact Info

Product

Resources

About