PENDEKATAN MULTIVARIAT UNTUK PENGUKURAN KUALITAS TOMAT (Lycopersicon esculentum) BERDASARKAN PARAMETER WARNA Multivariate Approach to The Measurement of Tomato (Lycopersicon esculentum) Quality Based on Color Parameters

Abstract

In this study, multivariate linear regression (MLR) was used to predict the content of Brix, total carotene, citric acid,and vitamin C of tomato based on RGB color parameters. Tomatoes were stored at 6 °C and 28 °C then their quality parameters were measured. R, G, and B values were measured non-destructively using computer vision system developed in the previous study. Brix, total carotene, citric acid, and vitamin C were determined by conventional procedures in the laboratory. Data analysis showed that the MLR calibration models could be used to predict Brix, total carotene, citric acid, and vitamin C with R2 of 0.77and 0.72, 0.902 and 0.85, 0.71 and 0.77, as well as 0.88 and 0.82 for temperature of 6 °C and 28 °C, respectively. ABSTRAK Pada penelitian ini, multivariate linier regression (MLR) digunakan untuk memprediksi kandungan Brix, karoten total,asam sitrat, dan vitamin C dari tomat berdasarkan parameter warna RGB. Tomat disimpan pada suhu 6 °C dan 28 °C kemudian diukur parameter kualitasnya. Nilai R, G, dan B diukur secara non-destructive dari computer vision system yang dikembangkan pada penelitian sebelumnya. Parameter kualitas Brix, karoten total, asam sitrat, dan vitamin C ditentukan secara destructive dengan prosedur konvensional di laboratorium. Analisis data menunjukkan bahwa model kalibrasi MLR dapat digunakan untuk memprediksi Brix, karoten total, asam sitrat, dan vitamin C dengan R2 sebesar0,77dan 0,72, 0,902 dan 0,85, 0,71 dan 0,77, serta 0,88 dan 0,82 untuk suhu 6 °C dan 28 °C secara berturutan.