<p>With the popularization of sports training, the training effect of athletes is closely related to environmental factors. Thermal environment monitoring directly affects the performance and health of athletes in sports training, but the traditional environmental monitoring methods are not accurate and real-time. This study aims to propose a thermal environment monitoring system based on image denoising algorithm and sensor to improve the accuracy and timeliness of environmental monitoring during sports training, so as to provide better training conditions for athletes. A monitoring system integrating image processing and sensor technology is studied and designed. The use of high-performance sensors to collect environmental temperature, humidity and other heat related data; Then the environment image is processed by image denoising algorithm to reduce the influence of noise on data acquisition. Finally, based on the sensor data and image analysis results, the thermal energy state of the training environment is evaluated in real time. The experimental results show that the monitoring system based on image denoising algorithm significantly improves the data accuracy compared with the traditional method, and can maintain good performance under extreme environmental conditions. The system can feedback the change of environment in real time and help the coaches and athletes adjust the training plan.</p>