In this study, an optical structure design and optical radiometric-calibration method for a hyperspectral-imaging lidar system are proposed. Using the proposed method, the central wavelength and bandwidth of each receiving channel were directly determined using a scanning monochromator. The system-calibration coefficients of each channel were measured in the laboratory using the hyperspectral-image lidar equation and excluding the influence of atmosphere extinction. Simulation results show that the synthetic uncertainty is 0.87% and the extended uncertainty (k=2) is 1.73%. Furthermore, a novel technique was used to monitor the emitted laser-pulse power in real time during the flight experiment. Based on echo-signal intensity information from the detector, the hyperspectral reflectance of the earth's surface can be accurately retrieved, and the high-precision terrain acquisition and super-fine surface classification can be realized.