International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 177 - Number 28 |
Year of Publication: 2019 |
Authors: Ottopianus Mellolo, Toban Tiku Pairunan, Eliezer Mangoting Rongre |
10.5120/ijca2019919745 |
Ottopianus Mellolo, Toban Tiku Pairunan, Eliezer Mangoting Rongre . Clustering the Quality of Coconut Wood based on Digital Images and Compressive Test Values using the Fuzzy C-Mean Method. International Journal of Computer Applications. 177, 28 ( Dec 2019), 21-26. DOI=10.5120/ijca2019919745
Every part of coconut commodity can be useful. Coconut wood is a part of coconut trunk that is widely used as building materials, furniture products, souvenirs or fuel. Although coconut wood is useful, but most people are still difficult to recognize kind of shape or texture model of a good quality coconut wood. The method used to test the quality of conventional coconut wood still relies on the compressive test model. This study is directed to determine the quality of coconut wood by modeling the relationship between visualization of wood surface texture and the results of compressive strength testing. Samples of coconut wood in this research are taken from a variety of ages, the location of the trunk and the outer part and the inside of coconut wood. Coconut wood samples used contain less than 20% moisture content, then taken digital images. By using the Fuzzy C-Mean method for the two-measurement data, forming 3 centers of coconut wood quality cluster, from the results of this clustering it is concluded that coconut wood with high bundle density also has a high compressive quality value, bundle density above 30.83% classified as wood superior type coconut. Meanwhile, bundle density of less than 11.88% is classified as low-quality coconut wood, besides that it is a medium type.