PREDIKSI TOTAL PADATAN TERLARUT BUAH MELON GOLDEN MENGGUNAKAN VIS-SWNIRS DAN ANALISIS MULTIVARIAT
| Main Authors: | Hadiwijaya, Yuda, Kusumiyati, Kusumiyati, Munawar, Agus Arip |
|---|---|
| Format: | Article info application/pdf Journal |
| Bahasa: | eng |
| Terbitan: |
Institute of Research and Community Services, Universitas Negeri Yogyakarta
, 2020
|
| Subjects: | |
| Online Access: |
https://journal.uny.ac.id/index.php/saintek/article/view/34487 https://journal.uny.ac.id/index.php/saintek/article/view/34487/pdf |
Daftar Isi:
- Penelitian ini bertujuan untuk memprediksi total padatan terlarut buah melon golden (Cucumis melo L.) menggunakan Vis-SWNIRS dan analisis multivariat. Terdapat 82 sampel buah melon golden dipilih untuk dianalisis di Laboratorium Hortikultura, Fakultas Pertanian, Universitas Padjadjaran. Nirvana AG410 spectrometer dengan rentang panjang gelombang 300 sampai 1050 nm digunakan untuk pengambilan data spektra pada sampel buah melon utuh. Metode koreksi spektra yang digunakan yaitu Standard Normal Variate (SNV), Multiplicative Scatter Correction (MSC), dan Orthogonal Signal Correction (OSC). Pemodelan kalibrasi dilakukan menggunakan Partial Least Squares Regression (PLSR). Hasil penelitian menunjukkan bahwa penggunaan metode koreksi spektra OSC menampikan model kalibrasi terbaik dibandingkan spektra original dan 2 spektra lainnya yang telah dikoreksi. Koefisien determinasi pada spektra OSC memperlihatkan nilai R2 tertinggi yaitu 0,99. Di samping itu, nilai ratio performance to deviation (RPD) yang diperoleh sebesar 3,40. Hal ini membuktikan total padatan terlarut buah melon golden dapat diprediksi dengan akurasi yang tinggi menggunakan Vis-SWNIRS dan analisis multivariat.PREDICTION OF TOTAL SOLUBLE SOLIDS OF GOLDEN MELON USING Vis-SWNIRS AND MULTIVARIATE ANALYSISThis study was aimed at predicting the total dissolved solids of golden melon (Cucumis melo L.) using Vis-SWNIRS and multivariate analysis. There were 82 golden melon fruit samples selected for analysis at the Horticulture Laboratory, Faculty of Agriculture, Padjadjaran University. Nirvana AG410 spectrometer with a wavelength range of 300 to 1050 nm was used to collect spectral data on intact melon fruit samples. The spectra correction methods used were Standard Normal Variate (SNV), Multiplicative Scatter Correction (MSC), and Orthogonal Signal Correction (OSC). Calibration modeling was carried out using Partial Least Squares Regression (PLSR). The results show that the use of the OSC spectra correction method presents the best calibration model compared to the original spectra and 2 other corrected spectra. The coefficient of determination on the OSC spectra shows the highest R2 value, namely 0.99, besides that the ratio performance to deviation (RPD) value obtained is 3.40. This proves that the total dissolved solids of golden melon can be predicted with high accuracy using Vis-SWNIRS and multivariate analysis.