<![CDATA[Multispectral imaging (MSI) is one of the optical methods used for the classification of fruits and vegetables based on ripeness levels. MSI is simpler than hyperspectral imaging due to fewer wavelength bands used hence less processing time. In this study, MSI is used to classify the ripeness of oil palm fresh fruit bunch (FFB). The MSI system consists of three main components, namely a VIS-NIR camera, a camera lens, an LED array, and a current control unit. The use of the LED array as a light source in the MSI system aims to minimize the use of bandwidth filters. The LEDs used are arranged in a circular pattern with 8 wavelengths, namely 680, 700, 750, 780, 810, 850, 880, and 900 nm. FFB samples were recorded using the MSI system and then processed using Python language to obtain relative reflectance intensity values. The purposes of this research are to analyze the relationship between relative reflectance intensity and wavelength and to classify the ripeness level of oil palm FFB using principal component analysis (PCA). We used two categories of ripeness, unripe and ripe FFBs.The results of the PCA analysis showed that the classification carried out was able to group into two levels of ripenesses with a total variant percentage value for PC1 and PC2 of 90.95%.]]>
Copyrights © 2024
