A Hybrid Mobile Application for Quality Grade of Tobacco Using Correlated Color Temperature

Abstract

Concerning tobacco producers, the size and color of the leaf are important factors in understanding the quality grade of tobacco leaves in the market. The color of tobacco leaves serves as an indicator of quality and is referred to as the maturity index when determining the optimal time for harvesting. In this study, a hybrid mobile application was developed to help determine the harvest time. CoLab was preferred as the backend. Python Imaging Library (Pillow) was used for image processing on the server side. Color correction was performed on the images taken with the help of X-rite. The correlated color temperature (CCT) value of the repaired images was calculated. The CCT values were calculated using the Ohno method. Quality grade (QG) was calculated using the mean CCT value. The data of the images obtained depending on the time were used in the application as a graphic. We present quality grade of tobacco automatically identifying the plant leaves in a given image with the help of the mobile application.