Human movement intelligent measurement and digital training system

A system for intelligent measurement and digital training of human motion, comprising N inertial navigation wearable devices, M cameras, a data comprehensive analysis device and a terminal, wherein both N and M are greater than or equal to 1;

• wherein,
• a total field of view of the M cameras is configured to cover a motion scene of an athlete, and each camera is configured to capture an image in the field of view to form an image data frame and send the image data frame to the data comprehensive analysis device;
• each inertial navigation wearable device is secured on a limb of the athlete in a wearable manner, and each inertial navigation wearable device is configured to measure a three-axis linear acceleration of the limb of the athlete and a three-axis angular velocity of the limb of the athlete in an inertial coordinate system by taking the limb of the athlete as a carrier, and send the three-axis linear acceleration and the three-axis angular velocity to a data comprehensive analysis device;
• and
• the data comprehensive analysis device is configured to store basic information of the athlete, and establish and maintain an association relationship between the athlete and an inertial navigation wearable device the athlete wears;
• perform, based on the three-axis linear acceleration of the limb of the athlete and the three-axis angular velocity of the limb of the athlete in the inertial coordinate system, navigation solution and coordinate conversion to obtain and store a relative position and an attitude of the limb of the athlete in a body coordinate system of the athlete;
• collect and store the image captured by a respective camera, and perform target identification, tracking and coordinate conversion on the image captured by the respective camera to obtain and store a position and a speed of the athlete in a world coordinate system of the motion scene;
• and analyze the position and the speed of the athlete in the world coordinate system of the motion scene as well as the relative position and the attitude of the limb of the athlete in the body coordinate system of the athlete to determine and store motion parameters of the athlete,
• wherein the data comprehensive analysis device comprises an inertial navigation solution device, a motion target detection and tracking device, a motion target speed identification device and a motion parameter analysis device;
• the inertial navigation solution device is configured to perform, based on the three-axis linear acceleration of the limb of the athlete and the three-axis angular velocity of the limb of the athlete in the inertial coordinate system, the navigation solution to obtain attitude, speed and position information of the limb of the athlete in a navigation coordinate system;
• perform zero-velocity detection on motion of the limb of the athlete;
• and perform, in a case that the limb of the athlete is within a zero-velocity interval, zero-velocity error correction on the attitude, speed and position information of the limb of the athlete in the navigation coordinate system;
• and define the body coordinate system of the athlete, to convert the attitude, speed and position information of the limb of the athlete in the navigation coordinate system into corresponding information in the body coordinate system of the athlete;
• the motion target detection and tracking device is configured to collect the image captured by the respective camera, record time for collecting the image, perform distortion correction on the image captured by the respective camera;
• perform target detection on each corrected image captured at a same time by using a you only look once YOLO model to obtain rough bounding boxes of all athletes in the image in a pixel coordinate system;
• obtain precise positions and precise bounding boxes of all athletes in the pixel coordinate system based on an edge detection method;
• and match precise bounding boxes of a respective athlete at different instants to track the precise bounding boxes of the respective athlete at different instants;
• convert coordinates of the respective athlete in the pixel coordinate system into coordinates in the world coordinate system corresponding to a coverage area of the field of view of the camera through a perspective projection matrix;
• calculate coordinates of the respective athlete in a global world coordinate system of the motion scene at different instants based on a position relationship among the coverage area of the field of view of the camera;
• and send the calculated coordinates to the motion target speed identification device;
• the motion target speed identification device is configured to filter and denoise a coordinate sequence of the respective athlete in the global world coordinate system of the motion scene at different instants, and perform differential processing on the filtered and denoised coordinate sequence to obtain the speed of the athlete in the world coordinate system of the motion scene;
• and
• the motion parameter analysis device is configured to analyze the relative position and the attitude of the limb of the athlete in the body coordinate system of the athlete to obtain the motion parameters;
• compare positions and speeds of athletes in the world coordinate system of the motion scene, analyze and sort these data, and rank the athletes based on a determined rule;
• and perform contrast and comparison based on the motion parameters of the athlete and standard parameters.