Deep learning technology has recently become increasingly prevalent in the field of education due to the rapid growth of artificial intelligence. Teachers’ teaching behavior is a crucial component of classroom teaching activities, and identifying and examining teachers’ classroom teaching behavior is an important way to assess teaching. However, the traditional teaching evaluation method involves evaluating by either listening to the class on-site or playing back the teaching video afterward, which is a time-consuming and inefficient manual method. Therefore, this paper obtained teaching behavior data from a real smart classroom scenario and observed and analyzed the teacher behavior characteristics in this scenario. Aiming at the problems of complex classroom environments and the high similarity between teaching behavior classes, a method to improve YOLOv7 for large target classroom behavior recognition in smart classroom scenarios is proposed. First, we constructed the Teacher Classroom Behavior Data Set (TCBDS), which contains 6660 images covering six types of teaching behaviors: facing the board (to_blackboard, tb), facing the students (to_student, ts), writing on the board (writing, w), teaching while facing the board (black_teach, bt), teaching while facing the students (student_teach, st), and interactive (interact, i). This research adds a large target detection layer to the backbone network so that teachers’ instructional behaviors can be efficiently identified in complex classroom circumstances. Second, the original model’s backbone was extended with an effective multiscale attention module (EMA) to construct cross-scale feature dependencies under various branches. Finally, the bounding box loss function of the original model was replaced with MPDIoU, and a bounding box scaling factor was introduced to propose the Inner_MPDIoU loss function. Experiments were conducted using the TCBDS dataset. The method proposed in this study achieved mAP@.50, mAP@.50:.95, and recall values of 96.2%, 82.5%, and 92.9%, respectively—improvements of 1.1%, 2.0%, and 2.3% over the original model. This method outperformed other mainstream models compared to the current state of the art. The experimental results demonstrate the method’s excellent performance, its ability to identify various classroom behaviors of teachers in realistic scenarios, and its potential to facilitate the analysis and visualization of teacher classroom behaviors.

Early fault vibration signals of rolling bearings are non-stationary and nonlinear, with weak fault signatures easily masked by noise. Traditional denoising methods (e.g., wavelet thresholding, empirical mode decomposition (EMD)) struggle to accurately extract effective features. Although variational mode decomposition (VMD) overcomes mode mixing, its core parameters rely on empirical selection, making it prone to local optima and limiting its denoising performance. To address this critical issue, this study aims to propose a hybrid model with adaptive parameter optimization and efficient denoising capabilities, enhancing the signal-to-noise ratio (SNR) and feature discriminability of early fault signals in rolling bearings. The novelty of this work is reflected in three aspects: (1) An improved edge strategy whale optimization algorithm (IEWOA) is proposed, incorporating six enhancements to balance global exploration and local exploitation. Using the minimum average envelope entropy as the objective function, the IEWOA achieves adaptive global optimization of VMD parameters. (2) A hybrid framework of “IEWOA-VMD + dataset-specific wavelet thresholding for secondary denoising” is constructed. The optimized VMD first decomposes signals to separate noise and effective components, followed by secondary denoising, ensuring both adaptable signal decomposition and precise denoising. (3) Comprehensive validation is conducted across five models using two public datasets (Case Western Reserve University (CWRU) and Paderborn Universität (PU)). Key findings demonstrate that the proposed method achieves a root-mean-square error (RMSE) as low as 0.00013–0.00041 and a Normalized Cross-Correlation (NCC) of 0.9689–0.9798, significantly outperforming EEMD, traditional VMD, and VMD optimized by single algorithms. The model effectively suppresses noise interference, preserves the fundamental and harmonic components of fault features, and exhibits strong robustness under different loads and fault types. This work provides an efficient and reliable signal preprocessing solution for early fault diagnosis of rolling bearings.