Implementasi Logika Fuzzy Mamdani dalam Penentuan Tingkat Kematangan Buah Tomat berdasarkan Ruang Warna Hue Saturation

Abstract

Indonesia is an agricultural country and one of the countries that has high potential in agriculture. Most of the Indonesian population earns their livelihood as farmers, one of which is tomato farmers. Tomatoes have a relatively fast maturity level, so farmers must be able to identify the fruit to select fruits that are ready to harvest according to the maturity level of the fruit. The maturity level of tomato fruit can be seen from several aspects, one of which is the color of the fruit skin. The process of identifying the level of fruit maturity is mostly done by farmers manually by looking at it directly which causes differences and non-uniformity from farmers and inaccuracy in the process of identifying the maturity of tomato fruit. These problems can be overcome with the use of digital image technology. Digital image features that are often used include the Hue, Saturation, Value (HSV) color space, and others. The use of digital images can be processed effectively using the Mamdani fuzzy logic method to help classify maturity levels based on criteria that are not binary, so that the assessment results are more flexible and closer to human perception. This research uses Hue, Saturation, and Value values as input values. While the output of this research is the level of ripeness of tomatoes in the categories of unripe, half-ripe, and ripe. The implementation of Mamdani fuzzy logic uses MATLAB software. Based on the research that has been done, the accuracy of the program and Kaggle data in determining the ripeness level of tomatoes based on HSV color space reaches 85.07%.