Laser paper used for exquisite packaging has physical effects such as scattering, diffraction, and interference on its surface due to its optical microstructure, so that its color information is difficult to be measured objectively, accurately and repeatedly with traditional color measuring instruments such as spectrophotometer. spectrophotometers. To solve the problems of large color detection error and difficult control of color stability of printed laser paper, the work aims to propose an image-based color stability detection method of printed laser paper. Firstly, the samples were visually calibrated, and the acceptable color difference tolerance sample set of standard prints was selected. Secondly, the digital images of the standard prints and the selected samples were captured in a uniform illuminated imaging system, and the spectral reflectance of the samples was reconstructed by the spectral reconstruction method, and the color data of the captured sample were calculated according to chromaticity theory. Then, the convex hull algorithm was used to construct the color gamut of the acceptable color difference tolerance and the corresponding ellipsoid and ellipse. Finally, the digital image of any sample to be tested was captured, and its color data were calculated. Whether the color gamut of the print to be tested was qualified or not was judged by the positional relationship between the color gamut and the print to be tested. The experimental results showed that the method proposed could accurately identify the color of printed laser paper, with a detection accuracy of 100% for qualified samples and more than 90% for unqualified samples. It can achieve the stability detection of printed laser paper with consistent visual perception, providing a solution for the color control of printed laser paper.