Browsing by Author "1632409042"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Open Access
    Real Time Safety Measurement Protocol System for Construction Sites Using Machine Learning in Bangladesh
    (North South University, 2022) Tanjila Islam; Tanzila Islam; Sourav Biswas; Md. Abir Ahmed; Shuvo Bhowmick; Shahnewaz Siddique; 1811017042; 1811027042; 1721288642; 1722322042; 1632409042
    Safety in the construction industry is one of the main concerns in Bangladesh. It is one of the most unpredictable and danger-filled industry sectors. Most developed countries endeavor to reduce the tragic damages and losses resulting from construction accidents by preventing, eliminating, and bypassing the probable occurrences. Unfortunately, Bangladesh is one of the countries most at risk of construction accidents because they lack a robust safety system. Both authorities and employees do not have a clear understanding of construction safety. Safety negligence tends to cause most accidents. Thousands of Bangladeshi workers are injured or die from accidents on construction sites every year. Lack of training and knowledge about the equipment are the top five causes of these misfortunes, followed by lack of personal protective equipment, lack of safety eliminating/avoiding design, unfit equipment, and a lack of knowledge about the equipment. [1] The last decade has seen numerous studies conducted to introduce effective protection systems within the construction industry using machine learning and computer vision. To achieve this goal, in this study we proposed a model to actively monitoring the proper wearing of Safety Equipment (hard-hat, gloves, face masks, vests, harnesses and boots) of the construction workers in real-time. Based on the results of experimental tests, the model proved to have 86.93% mean average precision, which was effective for identifying safety equipment correctly. In combination with YOLOv4 and Darknet, these pieces of equipment can be registered and classified simultaneously. In future, we want to develop a system that monitors the wear of safety equipment to determine if workers are wearing it properly based on our model. Workers will not be able to access certain construction areas if one of these pieces of equipment is missing