To realize the pre-perception and accurate recognition of coal-rock interface and provide an accurate cutting trajectory for a shearer before mining， a coal-rock interface identification method based on active infrared and coupling optimization of multiple influencing factors was proposed. Considering the influence of the coupling effect of illumination time， distance and intensity on the accuracy of coal-rock interface identification， we tested the active excitation infrared images of coal-rock specimens with different parameters. Then， an optimal combination of multiple factor parameters for achieving high-precision identification of coal-rock interface was determined by orthogonal experimental method. Moreover， combined with the iterative optimization method， the optimal parameters of each influencing factor were searched near the optimal combination to overcome the problem of local optimal parameter and realize the high precision identification of coal-rock interface. Experimental results show that the proposed method can realize the coal-rock interface rapidly and accurately， and the minimum identification accuracy is over 97.96%， which provides fine universality， as well as an effective technical means for realizing the underground intelligent and unmanned mining.