PENERAPAN ADAPTIVE NEURO FUZZY INFERENCE SYSTEM (ANFIS) DALAM MENGELOMPOKKAN JENIS KALENG BERDASARKAN CITRA RED GREEN BLUE (RGB)
Daftar Isi:
- The many uses of cans as a food and beverage container can provide waste problems that disrupt the environment. To minimize canned waste, it can be done recycling canned waste where the initial steps of sorting cans are based on the type of can. The purpose of this study is to classify the types of cans and determine the level of accuracy of grouping using the Adaptive Neuro Fuzzy Inference System (ANFIS) method. The data of this study consisted of 250 cans which were divided into training data and testing data with a composition of 30%: 70%, 50%: 50%, 70%: 30%. The membership functions used are Triangular, Trapezoidal, and Gauss. The results of this study indicate that the highest level of accuracy for training data is 56% in the composition of 30%: 70%, 30% : 70% with the Gauss membership function and 30% : 70 % with the Trapezoidal membership function. While for testing data is 44,80% in the composition of 50%: 50% with the Trapezoidal membership function. Keywords: Canned Type, ANFIS Method, Membership Function