Research on the Identification Model of Psychological Health Status of College Students Based on Improved YOLOv11

Accurately and automatically identifying the mental health status of college students is crucial for the development of modern students' mental health. However, existing manual recognition methods face challenges such as high labor intensity and difficulty in detection. Therefore, this study proposes a university mental health status classification model YOLO-PHSC based on YOLOv11. YOLO-PHSC is built on YOLOv11n and incorporates various improvement strategies to meet the high-precision requirements of practical applications. Specifically, the C3k2 module of the backbone network has been improved by adding cascaded diversified branch block (DEM) modules. By combining multi-scale convolution, sequential convolution, and average pooling operations, this model can learn more comprehensive and specific target features. After performing connection operations in the neck structure, an efficient network (MFA-KR) was introduced to achieve bidirectional fusion and weighted optimization of multi-scale features, further enhancing the model's ability to extract key features in object detection. In addition, CAFM attention mechanism was added after each C3k2 module in the neck to further improve the model's learning of key features. This study used a dataset based on 12 types of university mental health status materials, consisting of 13091 images. The experimental results indicate that the proposed YOLO-PHSC achieves the best recognition performance on the university's mental health status dataset. this one [email protected] The accuracy is 90.12%, which is 1.32% higher than YOLOv11n [email protected]. The experimental results show that YOLO-PHSC outperforms yolo series recognition models and achieves high-precision recognition of the mental health status of universities.