Research on safety helmet detection method based on convolutional neural network

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

doi:10.1117/12.2564896

17 April 2020 Research on safety helmet detection method based on convolutional neural network

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

Proceedings Volume 11455, Sixth Symposium on Novel Optoelectronic Detection Technology and Applications; 114554C (2020) https://doi.org/10.1117/12.2564896
Event: Sixth Symposium on Novel Photoelectronic Detection Technology and Application, 2019, Beijing, China

Abstract

By combining artificial neural network with deep learning technology, convolution neural network is characterized by local perception, adaptive feature extraction and end-to-end application, etc., and it has been used in image recognition and target detection more and more in recent years. Problems are existing widely in the traditional safety helmet detection algorithm generally such as the severe background interference, complex computing, high time-complexity and largely fluctuant accuracy. A detective method for safety helmet based on deep convolution network was proposed in this paper, which first decoded the acquired video monitoring data for a number of YUV images, then to determine the detecting area in the image, and transfer the YUV component image in the detecting area to the RGB image data; then in which to determine the training set and detecting set; finally, based on the constructed convolution neural network model to compute and process to acquire the ultimate detective results.

Citation Download Citation

Qiong Li, Fulun Peng, Zhibin Ru, Shuai Yu, Qinglin Zhao, Qiongjun Shang, Yue Cao, and Jie Liu "Research on safety helmet detection method based on convolutional neural network", Proc. SPIE 11455, Sixth Symposium on Novel Optoelectronic Detection Technology and Applications, 114554C (17 April 2020); https://doi.org/10.1117/12.2564896

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available