Endoscopes are commonly used to guide laser or radio frequency surgeries, which inevitably induce thermal effects on tissues. To avoid excessive damage to human tissues, it is necessary to monitor the temperature at the contact surface between the probe and the tissue. Currently, the main temperature measurement technologies rely on various temperature sensors. However, for disposable endoscopes, integrating sensors into the probe often increases the size of the endoscope and raises the overall cost. This study explores a temperature measurement method based on the dark noise of endoscopic image sensors. A mathematical model was developed to describe the relationship between dark noise and temperature, and the model was calibrated and validated. The temperature measurement accuracy of this method was maintained within approximately± 1.5℃. This research provides a theoretical foundation and technical support for temperature monitoring in disposable medical endoscope probes, demonstrating significant potential for practical application.