Based on the improved YOLOv9 laboratory personnel counting system

The laboratory, as a research base for university students, is becoming increasingly important to manage. With the development of technology, intelligent laboratory management systems have become a trend. Therefore, how to implement a fast and accurate laboratory management system has become a key issue. To solve these problems, this paper constructs a preliminary laboratory personnel counting system based on the improved YOLOv9 target detection algorithm. The following research work has been done in this paper: The important downsampling part ADown in YOLOv9 is improved, and the convolution module SPD-Conv for low-resolution image detection is combined, which improves the detection effect of YOLOv9 in low-resolution scenes. A laboratory personnel entry and exit counting system is designed. By dividing the entrance line, the statistics of the laboratory entry and exit personnel are achieved. When the number of laboratory personnel returns to 0, an alarm reminder can be given.