Tomato Classification Based On Ripeness With Rgb And Hsv Feature Extraction Using Naïve Bayes Algorithm

Sari

Tomato is a vegetable commodity that is also categorized as a fruit and is easy to cultivate in various regions. Differences in the level of tomato ripeness often become a challenge in the accurate classification process. Although many studies have been conducted related to the shape, disease, and varieties of tomatoes, classification based on the level of ripeness is still rarely done. This study aims to develop a classification of tomatoes using the level of tomato ripeness based on the color extracted through the RGB (Red, Green, Blue) and HSV (Hue, Saturation, Value) channels, using the Naïve Bayes algorithm. This research was conducted by collecting 150 tomato images that had similar shapes but with varying levels of ripeness, with a total of 135 training data and 15 test data. The research stages include the extraction of tomato color image features in RGB and HSV features, data simplification, separation between training data and testing data with a ratio of 90:10, and the application of the Naïve Bayes algorithm for the classification process. The results of the study showed that the RGB and HSV feature extraction methods combined with the Naïve Bayes algorithm were able to classify tomato ripeness levels with an accuracy of 80%. RGB and HSV color attributes together contributed to the classification accuracy, by producing a significant effect on certain ripeness categories