A Fast Alignment Algorithm for Posture Measurement of Autistic Children in Swimming using Computer Vision

Motor control ability is directly related to the stereotype and repetitive behavior of autistic children. Posture control refers to the control of the body's center of gravity, which is the most basic control ability. In order to solve the shortcomings of the existing research on the fast alignment algorithm of autistic children's swimming motion attitude measurement, this paper discusses the basic methods of attitude description, strap down inertial navigation system and binocular camera calibration algorithm, and proposes a multi-information aided strapdown inertial navigation system swimming motion attitude measurement alignment based on neural network. The specific experiment shows that the minimum j oint position error estimation of the proposed method reaches 0.0071 in the 200 second swimming posture alignment process. Therefore, the advantages of this method in rapid alignment of autistic children's swimming posture measurement are verified.