MODEL PENDETEKSI KEMATANGAN BUAH APEL DENGAN METODE NAIVE BAYES CLASSIFIER DAN SENSOR WARNA TCS3200 BERBASIS ARDUINO NANO

  • Nur Wikayati STMIK PPKIA Pradnya Paramita

Abstract

Abstract

The dominance of apple growth in East Java has reached the first position at the national level with productivity data from a total of 2,131,861 trees reaching 91,931 tons of apples produced. Apples are a type of fruit that can be processed into various foods, one of which is sticks. Many large companies in Indonesia produce apple sticks. Processed apples must have the appropriate maturity so that product quality is maintained. In maintaining the quality of the product, the ripeness of apples is grouped to separate raw, ripe and very ripe fruits. Assessment is subjective and inconsistent so that the results of measuring the ripeness of apples are not relevant between individuals. One way to solve this problem is to create an arduino-based model for detecting the ripeness of apples using the TCS3200 color sensor. This study aims to produce a detection model that can improve the accuracy in grouping the ripeness of apples. The color sensor readings in the form of R (Red), G (Green), B (Blue) values ​​from apples are used to classify the ripeness of apples using the naive bayes classifier method. There are 3 criteria for the level of maturity of apples, namely raw, ripe and very ripe. The results showed that the detection model can be used properly to detect the ripeness of apples. From the data used a number of 111 apples consisting of 36 training data and 75 test data, the results obtained accuracy of success reaching 90.67% with details of 7 unsuitable tests and 68 testing accordingly.

Keywords: Arduino, Color Sensor, Naive Bayes Classifier

Published
2022-07-31