Recognition method and system for hand raising in classroom based on machine vision

Abstract

The invention relates to a recognition method and system for hand raising in a classroom based on machine vision, belonging to the technical field of education information. The method comprises the following steps: collecting a to-be-detected image; according to the to-be-detected image, carrying out personnel identification by using a deep learning target detection model, and intercepting a region of interest of each person, wherein interferences of indoor seats, machinery equipment and the like can be effectively removed, high robustness is realized, and a calculation amount is reduced; based on a human body instance segmentation model, segmenting each region of interest to obtain a contour map of each part of the human body, wherein the contour map comprises an upper limb contour and a head contour, and the human body instance segmentation model is a model established by training a convolutional neural network by adopting a sample image in advance; and according to the upper limb contour and the head contour, determining whether a hand raising action exists or not. According to the method, multiple targets can be calculated at the same time, multi-target recognition is achieved, key part positions are obtained through an efficient region segmentation method, and hand raising recognition efficiency is improved.

Description

technical field [0001] The invention relates to the field of educational information technology, in particular to a method and system for recognizing raised hands in class based on machine vision. Background technique [0002] With the development of visual technology, automated classroom observation has become an important issue in the field of educational information technology. By observing the behavior and interaction characteristics of teachers and students in the classroom, it is helpful to objectively evaluate students, make different teaching plans for teachers, improve the quality of teachers' teaching, and raise hands as the most typical student action Behavior, its detection and recognition is an important supporting technology for automated classroom observation. [0003] The recognition of hand-raising movements currently used in teaching classrooms mainly includes the following methods, methods based on specific motion displacement, methods based on body shape...

AI Technical Summary

Problems solved by technology

In this method, each person needs to wear this terminal device, and the accuracy rate is not high, the cost is high, and it is not suitable for application in actual scenarios

Images