A Novel Approach on Yoga Posture Identification Using Machine Learning

The era of globalization has increased beyond expectations and to fulfil it, every individual has to face many challenges and stressful situations. To keep stress away from us many doctors and health experts suggest suitable yoga, pranayama, diet along with appropriate medicine. In recent times yoga exercise with postures estimation analysis of is followed everywhere because Practicing yoga helps to rise flexibility, improves muscle strength and boost respiratory system. In addition to fostering spiritual growth, yoga has several positive physiological impacts on the body. Yoga strengthens the body and the psyche when practiced regularly. Yoga is not only a physical exercise but also keeps our mental health intact. Yoga helps to maintain self-discipline and self-awareness. But because it is hard to verify the precision of the yoga poses, practitioners won’t gain fully from the exercises. In this research work, we are presenting a scientific analysis related to yoga posture recognition using systematic approach of artificial intelligence and machine learning. The objective is to verify the accuracy of the poses when practicing yoga. We are developing embedded wearable sensor-based device to acquire the time series data from sensors and by using suitable algorithm different postures get recognized. In addition to this we have also studied the benefits of practicing yoga in the improvements of various health parameters.