Abstract
With the development of sports training simulation technology, wearable sensors have been widely used to monitor the physiological signals of athletes. However, in indoor sports training, sensors are affected by infrared light interference, leading to a decrease in sensor signal quality and thus affecting training effectiveness. A research has proposed an infrared image compensation filtering algorithm based on wearable sensors, which performs a series of preprocessing steps on infrared images to improve image quality. Signal contrast enhancement technology is used to enhance the visual effect of the images. In order to solve the problem of infrared interference, an infrared image compensation model was established, and filtering algorithms were applied to process the compensated images to extract features related to motion posture. Filtering algorithms can improve image noise suppression and contour feature extraction by performing spatial or frequency domain filtering operations on the image. Through experimental verification, this algorithm effectively improves the accuracy and stability of sensor signals while reducing infrared light interference, and can more accurately capture the physiological changes of athletes.