Recognition of the part of growth of flue-cured tobacco leaves based on support vector machine

It is the most important in flue-cured tobacco leaves grading to recognize the parts of tobacco plants where the flue-cured tobacco leaves grown. In this paper, the image processing and analysis are applied in extracting the feature parameters of the tobacco leaves quality, the fuzzy statistics and comprehensive judgment techniques are applied to judge the group membership of tobacco leaf samples preparatory to the further recognition by means of support vector machine (SVM). The samples which are wrongly classified form the working set of SVM, the rest samples make up the non-working set. By this method, a passel of flue-cured tobacco leaves from Qujing area, Yunnan province are classified into 3 groups according to the recognition of the grow parts of tobacco plants. The result indicates that near 95% of samples in the SVM grouping are consistent with those in the expert grouping.