Pemindai Tingkat Kematangan Tandan Buah Sawit Menggunakan Kamera Handphone Berbasis Android

Abstract

Maturity level is a very important parameter of palm fruit bunch quality. There are two levels of maturity that are often used, ie unripe ripe, which are determined based on the number of loose fruit or the color of fruit. This method has several drawbacks, such as the quantity of oil that will decrease if many fruits fall from the bunch and the subjective assessment of fruit color. This research will discuss about making an application for scanning the maturity level of palm fruit bunches using the Oriented fast and Rotated Brief (ORB) method which is able to recognize the image contained in palm fruit bunches based on keypoint features. Matching will be done using the template matching method, so that it can find similarities between the images on palm fruit bunches. The result show that the scanner application can recognize every image being tested, but the success rate is only 50%, so this application cannot be fully implemented in the field to replace the method used so far. Thus, this research still needs further development.