Performance Analysis of SVM and Naïve Bayes for Mango Image Classification Based on Ripeness Level and Variety
Keywords:
Classification, SVM, Naïve Bayes, Image ProcessingAbstract
Mango is a seasonal fruit that is commonly known by people in Indonesia. This fruit that comes from India has many types that are widely favored by the people of Indonesia. However, in the case of the ripeness of this mango fruit, there are mangoes that have a fairly ripe color but after being purchased the mango is not fully ripe and there are also colors that we see ripe are actually fully ripe. The purpose of this study is to classify mangoes based on their level of ripeness using the color feature. Classification is carried out using the SVM and Naïve Bayes methods. The dataset used is 3 types of mangoes with 180 data samples, namely 20 unripe mangoes, 20 half-ripe, and 20 ripe. The accuracy results obtained by the Naïve Bayes method are 69% while the accuracy results obtained by the SVM method are 86%. The results of this study indicate that the SVM method is more efficient and effective in classifying mangoes based on their level of ripeness.
References
Costaner, L., Lisnawita, Guntoro, & Abdullah. (2024). Feature Extraction Analysis for Diabetic Retinopathy Detection Using Machine Learning Techniques. SISTEMASI:Jurnal Sistem Informasi, 13(5).
Harahap, K. M., Pangestu, D., Alfarizdi, D., & Iqbal, F. (2025). Application of MATLAB-Based Digital Image Processing for Avocado Fruit Size and Color Sorting. Jurnal Pustaka Galeri Mandiri, 5(1), 22–26.
Laia, F. H., Rosnelly, R., Buulolo, K., Lase, M. C., & Naswar, A. (2023). Classification of Madani Mango Fruit Ripeness Based on Shape Using Artificial Neural Network Perceptron Method. Jurnal Ilmiah Komputer Dan Teknologi, 13(1), 14–20.
Muchtar, M., & Muchtar, R. A. (2024). Comparison of KNN and SVM Methods in Mango Fruit Ripeness Classification Based on HSV Images and Statistical Features. Jurnal Informatika Dan Teknik Elektro Terapan, 12(2), 876–884.
Mukhti, M. F. I. (2022). Image Processing Application for Avocado Ripeness Identification Based on Fruit Skin Texture Using Statistical Feature Extraction Method. Jurnal Ilmu Komputer Dan Informatika, 5(1), 37–46.
Rahmaniar, C., Darmawati, E., & Widayanti, S. M. (2024). Delaying the Ripening of Arumanis Mango Fruit at Various Picking Ages Using Ethylene Adsorber. JURNAL KETEKNIKAN PERTANIAN, 12(102–116).
Salim, M. R., Daulay, A. A. A., & Sinaga, G. V. (2023). Palm Line Pattern Recognition Using Fuzzy K-Nearest Neighbor Method. Bulletin of Computer Science Research, 4(3), 270–275.
Saputra, A. J., & Andriyani, W. (2025). Fruit Image Classification Using Naïve Bayes Algorithm with Histogram of Oriented Gradients (HOG) Feature Extraction. Journal of Artificial Intelligence and Software Engineering, 5(1), 215–227.
Sugiyono, S., & Dionta, D. (2023). Image Processing Modeling of Mango Type Classification Using the K-Nearest Neighbor Method. Jurnal Sains Dan Teknologi, 5(1), 347–352.
Wardani, L. A., Wijaya, I. G. P. S., & Bimantoro, F. (2021). Classification of Types and Ripeness Levels of Papaya Fruit Based on Color, Texture and Shape Features Using Support Vector Machine. Jurnal Teknologi Informasi, Komputer Dan Aplikasinya, 4(1), 75–87.
Zain, L. H., Setiawan, E., & Fitriyah, H. (2021). Mango Fruit Ripeness Detection System based on NH3, C2H5OH and VOCs Gas Content using the K-Nearest Neighbor (K-NN) method. Jurnal Pengembangan Teknologi Informasi Dan Ilmu Komputer, 5(9), 3792–3798.
