Efficient Gait Analysis Using Deep Learning Techniques

Human Activity Recognition (HAR) has always been a difficult task to tackle. It is mainly used in security surveillance, human-computer interaction, and health care as an assistive or diagnostic technology combination with other technologies such the Internet of Things (IoT). data can be recorded help sensors, images, smartphones. Recognizing daily routine-based human activities walking, standing, sitting, etc., could statistical classify into categories hence 2-dimensional Convolutional Neural Network (2D CNN) MODEL, Long Short Term Memory (LSTM) Model, Bidirectional long short-term memory (Bi-LSTM) are for classification. demonstrated that recognizing on extremely accurate, almost all accurately getting recognized over 90% time. Furthermore, because examples generated from only 20 s data, these actions recognised fast. Apart classification, work extended verify investigate need wearable sensing devices individually walking patients Cerebral Palsy (CP) evaluation chosen Spatio-temporal features based 3D foot trajectory. Case-control research was conducted 35 persons CP ranging weight 25 65 kg. Optical Motion Capture (OMC) equipment referral method assess functionality quality foot-worn device. The average accuracy precision stride length, cadence, step length 3.5 ± 4.3, 4.1 3.8, 0.6 2.7 cm respectively. For swing, people had considerably high inter-stride variables. Foot-worn made it easier examine Gait even without laboratory set up about gait abnormalities who have during linear walking.