Interval Estimation of Motion Intensity Variation Using the Improved Inception-V3 Model

In the process of acute resistance exercise, repeated variation in motion intensity can lead to muscle fatigue and heart failure. Therefore, acquiring interval time training pattern effect be improved by variation. order achieve this goal, an Inception-V3 model is proposed for estimation. The MIVIE(Motion Intensity Variation Interval Estimation) dataset consisting Strong, Moderate, Weak groups centralized uninterrupted collection. Then, multi-modal fusion vectors time-frequency eigenvalues are stacked up $227\times227$ grayscale images fed into inception-CNN. Finally, manipulator’s trajectory optimization completed under guidance ATO-DQN (Adaptive Trajectory Optimization-Deep Q Network) algorithm based on This work improve non-stationary motor speed caused changes during rehabilitation, which better guarantee safety patients.