WORKERS SAFETY TRACKING IN FACTORY/SITE

Neural networks

In construction area accidents have been rapidly growing throughout years in many countries. Due to various dangers and disasters, the construction site is becoming most endangered jobs. The helmet is the main safety equipment for workers but many workers do not use it. If a worker without helmet an accident can be fatal. This paper presented an automatic method for helmet detection, ANN classification on construction area and a system for automatic helmet detection of workers without helmet. For processing, in first step, we detect helmet that moving real-time by extracting back ground out from fore ground using back subtraction then enhancing it using threshold and mathematical morphology method. In the second step, we classify between wearing helmet, wearing clothes in head and other without wearing helmet. Area is applied for statistical feature extraction and artificial neural network is applied for classification. In the final step, Fuzzy C-means is applied for detecting a helmet. From the experimental results, the accuracy rates of the ANN classification and helmet detection were 96.12% and 85.23%, respectively.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.