Determination of Pineapple Ripeness Using Support Vector Machine for Philippine Standards

Abstract

The determination of the ripeness of pineapple has always relied on the visual judgment of a person. While there may be research papers about pineapple maturity grading which can be found online, there is still a lack of pineapple maturity grading research papers that use the Philippine National Standards (PNS) using automated means. In this study, we will automate the process of determining the ripeness of pineapples based on the Philippine Standard using Support Vector Machine (SVM) and HSV Color Space. Using 100 sample images, we trained the system to identify the pineapple inside a container. It is then processed for segmentation where only the body of the fruit remains on the photo. Using HSV Color Space we detect the colors yellow and green and count their respective pixels. These values were used to determine the maturity of the pineapple. The results yielded a 100% accurate prediction with the Unripe and Overripe classes. However, the system only predicted 86% for Ripe classes. This error can be solved by increasing the lighting on the pineapple for the color to be seen by the camera. The researchers have successfully created a device that can determine the maturity of a pineapple with the use of image processing techniques such as HSV and segmentation.