An accurate pattern classification for empty fruit bunch based on the age profile of oil palm tree using neural network

Wafi Aziz (Universiti Teknikal Malaysia Melaka)
Afif Kasno (Universiti Teknikal Malaysia Melaka)
Nurkamilia Kamarudin (Universiti Teknikal Malaysia Melaka)
Zaidi Tumari (Universiti Teknikal Malaysia Melaka)
Shahrieel Aras (Universiti Teknikal Malaysia Melaka)
Herdy Rusnandi (Universiti Teknikal Malaysia Melaka)
Kamal Musa (Universiti Teknikal Malaysia Melaka)

International Journal of Electrical and Computer Engineering
Vol 9, No 6: December 2019

This paper proposes an efficient method for pattern classification system of empty fruit bunch (EFB) by using a neural network technique. The main advantage of this method is the accuracy and speed of algorithm such that it can be computed rapidly with the proposed system. To test the effectiveness of the proposed method, 120 of EFB’s data with different ages and length that been obtained from Malaysian Palm Oil Board (MPOB) are use in the pattern classification process. In addition, there  are three classes of EFB in this system, which are Class 1 (less than 7 year old), Class 2 (8 to 17 year old) and Class 3 (more than 17 year old). It is envisaged that the proposed method is very useful in classifying the EFB and  90% of the sample parameters are successfully classified to its class.