Scholarlink

<p>Students classroom behaviours are complex and variable, involving multiple aspects such as students personality traits, learning attitudes, thinking styles and learning abilities, but traditional classroom behavioural assessment cannot comprehensively and reasonably assess students classroom learning status. The study adopts the improved Yolov5 behavioural detection method to enhance the ability of small-target feature extraction for typical behaviours in the classroom by increasing the detection layer, in order to improve the algorithm’s image recognition accuracy in dense scenarios. The results of the study show that by adopting the task-oriented teaching method based on image recognition technology, teachers are able to dynamically adjust the teaching tasks and methods through students’ behavioural data, and accurately match students’ individual needs and learning status with data-driven decision-making, so that the frequency of students’ positive behaviours, such as listening, writing_reading, and raising hands, in the experimental class increases by 6%, 4%, and 1% respectively, compared with that of the control class. The study shows that image recognition can effectively support teachers to improve their classroom control ability, and combined with the task-oriented approach, it can promote the classroom concentration and participation of students with differentiated teaching needs, effectively stimulate students’ interest and motivation in learning, and help teachers and students to achieve positive feedback and promote the improvement of teaching quality. The study successfully applies image recognition technology to educational scenarios, solves the problems of insufficient recognition refinement, training scale and diversity limitations of image recognition in complex classroom environments, verifies the potential of image recognition integration into the teaching framework system, expands the boundaries of its application, and provides a reference paradigm for the application of image recognition technology in various complex teaching scenarios.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Research on improving students concentration by task-oriented method based on image recognition technology

  • 期刊论文

  • 发表日期：

    2025年5月9日

查看全文
  • Xiaohong Ji,
  • Xin Liu,
  • Xin Chen,
  • Rong Li

摘要

Students classroom behaviours are complex and variable, involving multiple aspects such as students personality traits, learning attitudes, thinking styles and learning abilities, but traditional classroom behavioural assessment cannot comprehensively and reasonably assess students classroom learning status. The study adopts the improved Yolov5 behavioural detection method to enhance the ability of small-target feature extraction for typical behaviours in the classroom by increasing the detection layer, in order to improve the algorithm’s image recognition accuracy in dense scenarios. The results of the study show that by adopting the task-oriented teaching method based on image recognition technology, teachers are able to dynamically adjust the teaching tasks and methods through students’ behavioural data, and accurately match students’ individual needs and learning status with data-driven decision-making, so that the frequency of students’ positive behaviours, such as listening, writing_reading, and raising hands, in the experimental class increases by 6%, 4%, and 1% respectively, compared with that of the control class. The study shows that image recognition can effectively support teachers to improve their classroom control ability, and combined with the task-oriented approach, it can promote the classroom concentration and participation of students with differentiated teaching needs, effectively stimulate students’ interest and motivation in learning, and help teachers and students to achieve positive feedback and promote the improvement of teaching quality. The study successfully applies image recognition technology to educational scenarios, solves the problems of insufficient recognition refinement, training scale and diversity limitations of image recognition in complex classroom environments, verifies the potential of image recognition integration into the teaching framework system, expands the boundaries of its application, and provides a reference paradigm for the application of image recognition technology in various complex teaching scenarios.