基于计算机视觉的课堂学生行为图像分析系统的设计与实现  

Design and Implementation of Classroom Student Behavior Image Analysis System Based on Computer Vision

作　　者：刘幸福 LIU Xingfu(Inner Mongolia Hospital Beijing Hospital Of Traditional Chinese Medicine,BayannurInner Mongolia 015000)

机构地区：[1]北京中医医院内蒙古医院,内蒙古巴彦淖尔015000

出　　处：《软件》2024年第7期144-149,共6页Software

摘　　要：本文设计并实现了一种基于计算机视觉的课堂学生行为图像分析系统。该系统采用最新的物体检测算法YOLOv5s,通过智能摄像头采集课堂图像,实时检测学生行为并进行分析,以提高课堂教学效果和学生的学习质量。本文汇集了众多的学生行为图像资料,经过数据预处理和数据增强,建立了适合YOLOv5s的数据库。其后利用收集到的课堂学生行为数据集训练了一个基于YOLOv5s的课堂学生行为检测模型,可以快速准确地识别学生在课堂中的各种行为,如举手、听讲、看书等。This article designs and implements a classroom student behavior image analysis system based on computer vision.The system adopts the latest object detection algorithm YOLOv5s,collects classroom images through intelligent cameras,detects student behavior in real time and analyzes it to improve classroom teaching effectiveness and student learning quality.This article collects numerous student behavior image materials,and after data preprocessing and enhancement,establishes a database suitable for YOLOv5s.Afterwards,a classroom student behavior detection model based on YOLOv5s was trained using the collected classroom student behavior dataset,which can quickly and accurately identify various behaviors of students in the classroom,such as raising hands,listening,reading,etc..

关 键 词：计算机视觉 学生行为 图像检测系统 深度学习 监督式学习 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]