Image processing for mango ripening stage detection: RGB and HSV method

Mango is one of the commercial fruits produced and consumed throughout the world. Mangos are needed to be classified according to their ripening stage for commercial use. Currently this classification is being performed manually which is not accurate and prone to human errors. Also it adds up the manpower and thereby overall cost and effectiveness of the mango processing industries. In this paper we have proposed two methods for classification of mangos based on changes in their visual features. We have considered Alphonso mango in our work as it is one of the most famous and traded mango type. But the proposed methods can be applied to any other species of mango which changes color with ripening process. In RGB method, the ripening stage is detected based on the red, green and blue components of the image pixels whereas in HSV method the hue-saturation-value map is analyzed for the detection. Algorithms are proposed and implemented using MATLAB. Results are compared with the manual results and found to be 90.4 and 84.2% accurate in case of RGB and HSV respectively. It is also demonstrated that the proposed methods are insensitive to amount of ambient light, provided that the image is taken in the natural light or under the white light.