College English Teachers’ Classroom Behavior and Improvement of Digital Literacy Based on Image Recognition

This study investigates the utilization of classroom behavior analysis technology, using image recognition to enhance college English teachers’ digital literacy. An algorithm for classroom behavior analysis has been devised to detect and interpret teachers’ classroom conduct automatically. The data are instrumental in evaluating and fortifying their digital proficiency. Methodologically, this study integrates behavioral analysis, image recognition, and deep learning techniques, employing the Convolutional Neural Network (CNN) model for behavior recognition. Experimental validations confirm the algorithm’s effectiveness. Results reveal that the CNN model demonstrates superior recognition accuracy and efficiency compared with the Long Short-Term Memory model. Image recognition-based classroom behavior analysis technology offers teachers unbiased and instantaneous teaching feedback. Such feedback aids them in refining their pedagogical approach, thereby elevating the quality of instruction. The implementation of this technology underscores the significance of digital educational aids in teachers’ career advancement. It carves out a fresh avenue for college English instructors to elevate their digital literacy.