The invention discloses a hyperspectral imaging technology-based fish fat content distribution detection method. The method comprises the following steps of (1) acquiring single-band spectral images of fish to be detected at 12 characteristic wavelengths, wherein the 12 characteristic wavelengths are 956nm, 975nm, 1,033nm, 1,081nm, 1,129nm, 1,153nm, 1,205nm, 1,234nm, 1,282nm, 1,354nm, 1,522nm and 1,627nm respectively; (2) converting the single-band spectral images in the step (1) into reflectivity images according to a linear relationship between gray values of the single-band spectral images and reflectivity; (3) calculating fish fat content corresponding to each pixel in the reflectivity images of the fish according to a formula. According to the fish fat content distribution detection method, the detection accuracy is high and the detection time is short; not only is environmental pollution alleviated, but also the detection cost is lowered to a certain extent.