In order to precise recognition the yellowing degree of tobacco leaves in the curing process and serve the development needs of the digital intelligent baking mode of tobacco leaves.The yellowing images of middle and upper tobacco leaves of K326 during curing were collected.The optimal color threshold foryellowing area identification of tobacco leaves was determined by three methods:general method,numerical verification method and empirical discriminant method,and the yellowing degree of tobacco leaves was calculated,and the yellowing characteristics of tobacco leaves with different qualities were analyzed.The results showed that the color thresholds obtained by the general method and the empirical discriminant method were not fine enough,and the recognition effect on the yellowing degree of tobacco leaves during the baking process was poor.The color threshold(Hmin=17,Hmax=32,Smin=138,Smax=255,Vmin=157,Vmax=255)obtained by the numerical verification method can accurately identify the yellowing degree of tobacco leaves with different qualities.The change process of yellowing degree of middle tobacco leaves is divided into slow start yellowing,rapid yellowing and slow termination yellowing stage.The change process of yellowing degree of upper tobacco leaves is divided into rapid yellowing and slow termination yellowing stage.With the increase of maturity of tobacco leaves,the maximum yellowing rate and the time of yellowing of tobacco leaves in different parts gradually decreased.The color threshold obtained by the numerical verification method can be used for tobacco leaf curing process,which provides a methodfor the realization of digitalization of tobacco leaves.
万方数据