Istaqim: An Assistant Application to Correct Prayer for Arab Muslims

Prayer in Islamic community is the second pillar of Islam, a link between the servant and his Lord, it is an act of worship that Muslims must perform five times a day. There are many postures in the pillars of prayer and its duties that must be performed in a precise manner. However, many Muslims, young and old, do not perform prayer properly due to having learned to pray incorrectly, having no one to personally guide them, or being new to prayer. To address this issue, we propose to develop an Artificial Intelligence assistant work using deep learning to guide worshipers through detecting the wrong postures in their prayers and assessing their mistakes and showing corrections. The contribution of this paper lies in training the YOLOv5 neural network to recognize prayer postures. The results are shown only for the correct prayer positions via the pictures with the percentage of the error of each prayer position.