Computer Vision System Based for Personal Protective Equipment Detection, by Using Convolutional Neural Network
Faishal Zhafran, Endah Suryawati Ningrum, Mohamad Nasyir Tamara, Eny Kusumawati · 2019
The number of work accidents happened in Indonesia does not decrease, in this moment. One of the cause of work accidents is human negligence in wearing personal protective equipment or what is called PPE. A system that can monitor the completeness of PPE worn by workers in an industrial environment automatically is needed. Computer vision technology one method known as convolutional neural network is used, this is a system to monitor directly and detect workers who do not wear PPE. The level system precision have seen by compare between the real condition and the system condition red which is shown the people without PPE completely. The resulted of the system resulted would compare with the examiner's. The result shown that this system is able to know and distinguish between workers who wear PPE in full and not. With the number of dataset 14512 images, the accuracy of this system is 79.14% with the precision of the detection resulted 80%. This system can give a warning if there are workers who do not wear PPE completely and also could be implemented into several workplaces that have the requirements for the detection. This system give the contribution to decrease work accidents and to increase the safeness of the workers.