An action evaluation method, device, mobile terminal and readable storage medium

Abstract

The invention discloses an action evaluation method, device, mobile terminal and readable storage medium. The method includes acquiring an action image of an object to be detected; wherein, each part of the object to be detected is divided into several parts according to the rotation angle of the corresponding joint. Regions, and each region is marked with a different color; the action image is transmitted to the preset joint recognition model to identify each joint of the object to be detected in the action image; the color of each part of the object to be detected in the action image is recognized , to determine the rotation angle of the joint corresponding to each part of the object to be detected; according to each joint of the object to be detected in the action image and the rotation angle of each joint, the motion joint map of the object to be detected is obtained; the motion joint graph and Compare the pre-stored standard motion joint graphs, and evaluate whether the motion of the object to be tested is standard according to the comparison results. By implementing the embodiments of the present invention, it is possible to evaluate the local motion of the object to be detected and improve the accuracy of motion evaluation.

Description

technical field [0001] The invention relates to the field of image recognition, in particular to an action discrimination method, device, mobile terminal and readable storage medium. Background technique [0002] In reality, whether it is a human or a machine, the evaluation of actions generally relies on the naked eye and experience judgment of relevant personnel. Because the evaluation cannot meet the strict requirements of precision, standardization, and perfection, the level of performance of the movements is often unsatisfactory. In addition, the evaluation of actions depends on human on-site supervision, which cannot realize intelligent automation and is difficult to popularize in society. [0003] With the continuous development of computer technology, based on the above problems, a series of motion evaluation products have emerged on the market. However, the action evaluation method in the prior art is based on the overall assessment of the action, and it is imposs...

AI Technical Summary

Problems solved by technology

However, the action evaluation method in the prior art is based on the overall assessment of the action, and it is impossible to evaluate a specific implementation part. Due to the different placement of the bones and joints, the implementation action is different from the standard action. For example, the previous Action evaluation can identify whether the hand is "clenching a fist" or "opening the palm", but it cannot identify the orientation of the palm of the hand according to the rotation angle of the wrist joint. For example, when "opening the palm", whether the palm of the hand is facing up or facing , resulting in inaccurate action evaluation

Images