This project is focused on analyzes squat performing video through image processing by considering key points taken from front and side view. There are conditions to check the performance of squat correct or not. If the Squat is correct, three angles are analyzed reference to five positions. The OpenCV is used to identify the five positions. Also, the Vector function is used to determine the reference angles by using the Cosine rule. These angles are calculated at the knee, waist, and ankle. All these angles are calculated on the free-weight back squat exercise. The ultimate purpose of this project is to minimize injuries that occur due to technical errors because most of the armature players get injured due to the difficulty in posing of correct technique. The mobile application is developed to identify user mistakes and can get instructions.

Keywords : squat, image processing, OpenCV.