Text
PEMUTUAN CRUDE PALM OIL (CPO) DENGAN METODE PENGOLAHAN CITRA DIGITAL BERBASIS JARINGAN SARAF TIRUAN
The research aims to determine the quality of Crude Palm Oil (CPO) non-destructively using a digital image processing method based on artificial neural networks build upon the color of Crude Palm Oil (CPO) using a digital camera. The treatment factors were analyzed using the Non-Factorial Completely Randomized Design method with 3 treatment factors at storage temperatures of 40±5°C, 50±5°C, and 60±5°C. The research used CPO quality parameters including free fatty acid content, DOBI index, moisture content, and impurities content. The results showed that storage temperature treatment had a significant effect on CPO quality parameters including free fatty acids, DOBI index, and moisture content and had no significant effect on quality parameters of impurities content. Storage temperature treatment of 50 ± 5°C resulted in the lowest increase in free fatty acids, the highest rate of decrease in water content, and the highest increase in DOBI index respectively of 0.36±0.10%/day, 0.11±0.01%/day, and 0.19±0.02/day. The artificial neural network model developed to determine the quality of crude palm oil (CPO) with RGB index input data produces a Mean Absolute Percentage Error (MAPE) value below 5% and the smallest Mean Square Error (MSE) of 3,71±10-4 Keywords: Artifical neural networks, CPO quality, image processing, storage tempetaure
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2307000968 | T88021 | T880212023 | Central Library (Referens) | Available but not for loan - Not for Loan |
No other version available