CD Skripsi
Pengaruh Warna Lampu Dan Filter Pada Kualitas Citra Untuk Klasifikasi Tingkat Kematangan Tandan Buah Segar (Tbs) Kelapa Sawit
The ripeness level of oil palm Fresh Fruit Bunches (FFB) is a determining factor for the quality of Crude Palm Oil (CPO). The method of sorting FFB before entering the boiling process is generally done manually which is destructive and subjective. Imaging methods have emerged to address this problem. This research aims to develop a computer imaging system based on LED lights and color filters for the effective classification of the ripeness level of palm oil FFB. The main purpose of this study is to determine the influence of the color of LED lamps and filters on image quality for the classification of the ripeness level of oil palm FFB. The LED lights used consist of 4 types of LED light colors, namely red, green, blue, and white LED lights and the color filters used consist of red, green, and blue color filters. The image of two parts of oil palm FFB, namely the front and back sides, were taken using a color camera. The classification process was carried out using Principal Component Analysis (PCA) based on RGB and HSV color spaces. The reflectance intensity of each FFB sample was analyzed to determine the relationship with the ripeness level of palm oil FFB. The classification method using PCA based on RGB color space for LED lamps managed to achieve a cumulative variance percentage of 96.77% and for color filters reached 99.6%. For classification based on the HSV color space, where only the Hue value is used. Separating the ripe and unripe categories reaches 100% accuracy. The results of this study show that computer imaging systems based on LED lights and color filters has potensial to the development of non-destructive methods efficiently.
Keywords: Computer imaging, LED light, color filter, color space, ripeness level classification, PCA.
Tidak tersedia versi lain