Temperature is crucial for evaluating the effectiveness and understanding the mechanisms of microwave-induced rock weakening. However, the measurement of actual temperature for microwave-treated rock using real-time and non-real-time methods, as well as its influencing factors, have not been systematically studied. In this paper, the cooling processes of rock surface after multimode microwave heating was first investigated using the non-real-time temperature measurement method (with an infrared thermal imager). The effects of initial temperature, wind speed, and cooling time on rock surface temperature were preliminarily investigated. On this basis, the numerical model was calibrated using the experimental results and then numerical simulations were conducted for rock samples under multimode microwave heating and microwave surface irradiation. The experimental and numerical results indicate that, with increasing cooling time, the surface temperature continuously decreases, and the cooling rate gradually diminishes. Within the same period, the higher the initial surface temperature or the greater the wind speed, the larger the surface temperature drop. Subsequently, the applicable conditions for non-real-time temperature measurement were clarified, and a correction model for the highest surface temperature of rock samples under multimode microwave heating and microwave surface irradiation was established, respectively. The cooling mechanisms of rock surface under the two types of microwave irradiation were also revealed. Finally, for the real-time temperature measurement with infrared sensors, the deviation range between the highest surface temperature and the temperatures at the centers of side and top surfaces of rock samples was provided, with heating by multimode microwave oven in this study as an example. The results show that the temperatures at the centers of side and top surfaces of the sample treated by multimode microwave were 20-30% lower than the highest surface temperature.