ISBN: 978-981-14-4787-7 DOI: 10.18178/wcse.2020.06.014
An Accident Analysis and Prevention Model based on Heterogeneous Convolution Kernel
Abstract— In order to reduce the incidence of safety accidents in factories, an accident analysis and prevention model based on heterogeneous convolution kernel is proposed. On the basis of the framework of convolutional neural network, the traditional fixed convolutional kernel optimization network was replaced by constructing heterogeneous convolutional kernel, which was applied to the industrial workshop to monitor the age and gender of workers in the workshop in real time, so as to reduce the calculation amount and the number of parameters while maintaining the efficiency of characterization. Through the comparative experiment in the standard database of Adience and SCUT-FBP 5500, the experimental results of various methods are analyzed. The results show that the model proposed in this paper can achieve 92% accuracy in gender classification of characters and 67% accuracy in age prediction. It can monitor the gender and age of workers in the workshop, real-time warn the potential safety hazards, and reduce the safety of the factory to a certain extent the occurrence rate of all accidents.
Index Terms— Heterogeneous Convolution Kernel; Convolutional Neural Network; Age Predicts; Gender Prediction; Accident analysis; Accident prevention
Zhipeng Yang, Shasha Wang, Yanling Li
School of Computer and Information Technology, Xinyang Normal University, CHINA
Cite: Zhipeng Yang, Shasha Wang, Yanling Li , " An Accident Analysis and Prevention Model based on Heterogeneous Convolution Kernel " Proceedings of 2020 the 10th International Workshop on Computer Science and Engineering (WCSE 2020), pp. 81-86, Shanghai, China, 19-21 June, 2020.