Fusion technique for honey purity estimation using artificial neural network

This paper presents the estimation of purity of honey using Artificial Neural Network (ANN). ANN method was used to automate the decision of estimation, replacing the manual human approximate method. A total of 21 honey purity samples from various concentrations of pure honey, adulterated honey and pure sugar were collected. The data were collected using electronic nose (E-nose) and Fourier Transform Infrared (FTIR). Fusion method was applied in this work by combining the variable from both E-nose and FTIR to produce new sets of data. This data were used as input parameters for ANN learning. The results show that ANN was able to estimate the concentration of honey purity in adulterated honey solution with less error using fusion data as compared to single modality data.