Underwater optical imaging has the problem of “not seeing far” and “not seeing clearly” due to the absorption and scattering of the water body. Underwater laser range-gated imaging technology can improve the underwater optical imaging distance and image contrast. The paper presents the research of underwater long-range target intelligent identification system based on underwater laser range-gated imaging technology. Laboratory test results show imaging distances in excess of 7 times the attenuation length. The study combines deep learning algorithms to achieve quasi-real-time detection of targets in power-constrained hardware conditions. The combination of underwater laser range-gated imaging technology and deep learning algorithms is expected to enable underwater optical imaging to “seeing far” and “seeing clearly”, while “seeing fast” and “seeing accurately”.