Abstract: With the rapid development of the cigarette industry, the quality of cigarettes has become higher and higher in the production process. Traditional length measurement method usually requires manual operation, which is not only low-efficiency, but also easily affected by subjective factors. The length of the cigarette segment (the length ratio of the cigarettes and the length ratio of the filter part) has received industry concern as an important quality indicator. In order to improve the recognition accuracy of cigarette segment length, the cigarette segment length recognition method based on machine vision and watershed algorithm is proposed. The method distinguishes different regions according to the brightness variation of the image, which can effectively segment the length of each segment of the cigarette and realize the length measurement. The experimental results show that the method can effectively identify and measure the length of cigarette. It ensures the quality stability of cigarette products and provides a feasible solution for the quality control in the process of cigarette production.