The identification of the coal and gangue are essential in the coal preparation process. Combining with imaging processing technologies (like de-noising, enhancement, and feature extraction), the imaging-based technology accomplishes the coal-gangue identification while avoiding the shortcomings of traditional methods (radiation, pollution, etc). However, environment factors (light, dust, temperature, etc) would greatly influence the identification results. A millimeter imaging combining with the optical photo technique is proposed for the recognition of coal and gangue. First, the millimeter wave (MMW) imaging results of the ores are achieved by the Range Migration Algorithm (RMA). Second, the profiles of the ores in the corresponding optical photo are adopted by the Seeded Region Growing (SRG), which are transferred to the MMW images for further identification. Furthermore, a simple threshold-based strategy is applied to compute the coal ratio of each ore to achieve the identification. The experimental results demonstrate that the proposed strategy is of great potential for nondestructive identification of coal from the gangue.