<![CDATA[Tomato is a fruit-category vegetable plant that is easy to cultivate in various locations. The diversity of tomato varieties, such as Red Zebra Tomato, Green Zebra Tomato, and Kumato, often makes rapid and accurate variety identification challenging. Misclassification can impact the selection of environmental conditions and pest or disease management, ultimately leading to suboptimal cultivation results. Currently, research primarily focuses on tomato shape, diseases, and ripeness levels, while cultivar classification based on color characteristics remains limited. This study aims to develop a method for classifying tomato cultivars based on RGB color features using the Naïve Bayes algorithm. The research was conducted by collecting 45 tomato images with similar shapes but different colors (red, green, and dark red). The research stages include RGB feature extraction, data rounding, splitting training and test data with a 70:30 ratio, and classification using Naïve Bayes. A re-evaluation was performed by removing specific color attributes to assess their impact on accuracy. This study is expected to support rapid and accurate tomato variety identification, enhance efficiency in modern agriculture, and expand the application of technology in the agricultural industry to achieve advanced, self-sufficient, and modern farming. The results show that the RGB feature extraction method and the Naïve Bayes algorithm can classify tomato cultivars with an accuracy of up to 78.57%. The RG color attributes have the most significant impact on accuracy, reaching 85.71%.]]>
