Abstract: Several intelligent technologies used in the automatic grading of flue-cured tobacco were studied. Image processing techniques were used to extract the characteristics of tobacco. Fuzzy pattern recognition statistics were applied to calculate the degree of membership for grading tobacco according to its characteristics, and comprehensive judgement was utilized to determine which group and grade a sample tobacco leaf belongs to. Hundreds of tobacco samples produced in Yunnan Province were graded, which proves that the grading system has met the standards of tobacco experts and possesses intelligence capabilities for accumulating experience.