The Research On Critical Technique Of Quality Grading Of Flue-Cured Tobacco Leaves Based On Image Processing

The quality of flue-cured tobacco leaves have a direct relation with the quality of the tobacco products and smokers'health. For the strong subjectivity, low efficiency and accuracy in the tobacco manual detection and grading, the computer image processing technology is applied to grade the quality of flue-cured tobacco leaves in this paper.The flue-cured tobacco leaves are studied in this paper. In order to automate the tobacco leaves'grading process, tasks of this paper are elaborated as follows:1. Preprocessing the tobacco image: the improved vector median filter method is chose to smooth the color tobacco image and avoid the noise in it. The iteration threshold value method is adopted to realize the background segmentation of the tobacco image. The simple contour extraction method is used to acquire the contour information of the tobacco leaves.2. Extracting the feature parameters of the tobacco: Firstly, appropriate tobacco leaves'features which are graded according to the tobacco classification standards are chose as the computer grading judgment. The hue, saturation and brightness mean value are extracted as the color feature parameters of tobacco using the HIS color model which is close to human vision. The Least-square elliptic fitting method is applied to extract the length, width, ratio of length and width, area, perimeter and damaged degree of tobacco leaves as the shape characteristic parameters.3. Implementing the classification algorithm: The required characteristics parameters are chose by analyzing their correlation. Gaussian function is introduced as the membership function of tobacco classification according to the distribution of characteristics parameters. And then the weight of various characteristics is got according the experts'experience and the mathematical model of fuzzy comprehensive evaluation for tobacco grading is established. At last, the final tobacco grading is achieved according to the maximum membership degree principle.