Determination of Quality Grade of Tobacco Leaf by Image Processing on Correlated Color Temperature

Abstract

The most important factor in tobacco is quality. It is the sum of physical, chemical, and smoking properties that characterize quality. Post-harvest curing of leaves plays an important role in quality. In this study, an approach based on correlated color temperature (CCT) is proposed to determine the ideal curing time of tobacco leaves. While photographing leaf samples from different days during the curing of tobacco leaves, color correction was made using X-rite color checker to eliminate the negative effect of ambient light. CCT values were calculated from the repaired photographs using the McCammy method. The inclusion rate of pixels during the calculation is 97%. The equation of CCT variation depending on the day is fitted as CCT=(-0.1889xday3)+(4.8347xday2)-(115.59xday)+4919.2$$ \mathrm{CCT}=\left(-0.1889\times {\mathrm{day}}<^>3\right)+\left(4.8347\times {\mathrm{day}}<^>2\right)-\left(115.59\times \mathrm{day}\right)+4919.2 $$. RMSE (26.77 K) and high R2$$ {R}<^>2 $$ value were calculated as 98.8% in the equation. CCT values have been converted to quality grade (QG) values to be more understandable. As a result of the study, CCT values ranged between 5000 and 2500 K and corresponding QG values ranged between 79% and 105%. The ideal curing time was calculated as the 17th day after harvesting, the QG value for that day was 100% and the average CCT value was 3000 K.