TY - JOUR AU - Laily, Milzamah Elvi AU - Fajri, Fathorazi Nur AU - Pratamasunu, Gulpi Qorik Oktagalu PY - 2022/12/01 Y2 - 2024/03/28 TI - Detection of the Use of Personal Protective Equipment (PPE) for Occupational Health and Safety Using the Mask Region Convolutional Neural Network (Mask R-CNN) Method JF - Jurnal Komputer Terapan JA - JKT VL - 8 IS - 2 SE - DO - 10.35143/jkt.v8i2.5732 UR - https://jurnal.pcr.ac.id/index.php/jkt/article/view/5732 SP - 279 - 288 AB - <p><em>The construction sector is one of the strongest sectors in supporting economic growth in Indonesia. In supporting the development and demands for the provision of services from the construction sector, as a state of law, the state of Indonesia has laws in the construction sector, one of which is on Occupational Health and Safety (K3). One of the efforts to minimize the consequences caused by work accidents, every worker is required to use Personal Protective Equipment (PPE). Lack of self-awareness and discipline of workers in the use of Personal Protective Equipment (PPE), can result in a fairly large risk of work accidents. So there needs to be automatic detection for workers in the use of good and correct PPE. This research is using the latest method from R-CNN, namely Mask Region Convolutional Neural Network (Mask R-CNN). The best model obtained is the epoch 35 parameter with a loss value of 0.1985 and a val_loss value of 0.1933 in 461s 922ms/step. Trial ith 250 images which produces an accuracy 0f 95%.</em></p> ER -