The work aims to improve the generality of threshold method of paper disease detection system. Firstly, the image gray-scale data were obtained by experiment and were classified according to different time periods. Secondly, the maximum average value, the minimum average value, the average value of overall gray-scale and the standard deviation of image grey-scale data were calculated. Finally, the graph analysis method was used to compare and analyze data on grey-scale of image, summarize the regular pattern of the change of image gray-scale value and complete programming and verification by Visual studio 2015. The experiments showed that the proposed method can significantly adapt to the changes of external light, and improve the robustness of paper disease detection system. The dynamic threshold setting scheme of paper disease detection based on fuzzy logic can effectively enhance the universality of the paper disease detection system.