Prediction of The Maturity Level of Pontianak Oranges (Citrus Suhuniensis CV Pontianak) Using Color and Texture Features of Digital Reflectance-Fluorescence Images and Partial Least Square (PLS) Model
- DOI
- 10.2991/978-94-6463-525-6_14How to use a DOI?
- Keywords
- Computer vision; Fluorescence; Partial Least Square Regression; Reflectance
- Abstract
Pontianak sour orange (Citrus suhuniensis CV Pontianak) is one of the citrus varieties that is widely found among the Indonesian population. This fruit has a sweet taste with a hint of sourness. Often, there are mistakes in determining whether the fruit tends to be sweet or sour. The taste of the fruit is influenced by different composition contents within the fruit. This also applies to Pontianak sour orange, where the taste is determined by specific contents. The factors that influence the taste of this fruit are the total soluble solids (brix) and total acidity. During the ripening process of Pontianak sour orange, there is a rearrangement of compounds within the fruit. This process involves an increase in total soluble solids (brix), a decrease in total acidity, and a decrease in fruit hardness. During the ripening process of Pontianak sour orange, there are changes in the fruit’s composition involving an increase in total soluble solids (brix), a decrease in total acidity, and a decrease in fruit hardness. The aim of this research is to develop a prediction model for the ripeness level of Pontianak sour orange based on digital reflectance and fluorescence image analysis using computer vision with color and texture features. The experimental laboratory method was conducted in two stages: destructive and non-destructive tests. Destructive testing was performed by measuring fruit firmness, total soluble solids (brix), and total acidity. Non-destructive testing was carried out by capturing fruit images using computer vision based on reflectance fluorescence. The classification modeling process used machine learning algorithms, including K-Nearest Neighbors (KNN), Linear Discriminant Analysis (LDA), Logistic Regression (LR), and Support Vector Machines (SVM). Partial Least Square Regression (PLSR) was used for predicting the ripeness level of the fruit, including fruit firmness, total soluble solids (brix), total acidity, and brix-acid ratio. The best model for classifying the ripeness level of the fruit was obtained using Support Vector Machine (SVM). In the prediction modeling of the physicochemical characteristics of Pontianak sour orange fruit, the results showed that for the all dataset with min-max scaling, the training accuracy was 0.97, and the testing accuracy was 0.97. In the regression model, the results showed that for the firmness parameter using the reflectance dataset, the training R2 value was 0.82 and the testing R2 value was 0.65. For the brix parameter using the fluorescence dataset, the training R2 value was 0.55 and the testing R2 value was 0.56. For the acidity parameter using the all dataset, which was a combination of reflectance and fluorescence datasets, the training R2 value was 0.83 and the testing R2 value was 0.87. And for the brix-acid ratio parameter using the reflectance dataset, the training R2 value was 0.72 and the testing R2 value was 0.57.
- Copyright
- © 2024 The Author(s)
- Open Access
- Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
Cite this article
TY - CONF AU - Dimas Firmanda Al Riza AU - Salsabil Lazuardi Nugroho AU - Ahmad Avatar Tulsi AU - Mochamad Bagus Hermanto AU - Yusuf Hendrawan AU - Naoshi Kondo PY - 2024 DA - 2024/10/29 TI - Prediction of The Maturity Level of Pontianak Oranges (Citrus Suhuniensis CV Pontianak) Using Color and Texture Features of Digital Reflectance-Fluorescence Images and Partial Least Square (PLS) Model BT - Proceedings of the 2023 Brawijaya International Conference (BIC 2023) PB - Atlantis Press SP - 125 EP - 132 SN - 2352-5428 UR - https://doi.org/10.2991/978-94-6463-525-6_14 DO - 10.2991/978-94-6463-525-6_14 ID - AlRiza2024 ER -