Analisis Kinerja Instrumen Uji Formalin pada Tahu Berbasis Elektronik Nose

Sariayu Wulandari (Universitas Dian Nuswantoro)

Jurnal Nasional Teknik Elektro dan Teknologi Informasi
Vol 2 No 4: November 2013

Enose is an instrument that be able to discriminated odorant. This instrument is designed to detect formalin content in the tofu product. The aim of artificial olfaction machine design is to find the recommended method to detect content formalin with specified instrument. The instrument consisted of electronic module and pattern recognition systems. The electronic module is a typical application system, consisting of Arduino Uno R3 board, with ATMega 328 chip. In the pattern recognition system, is programmed in Matlab language using LRMA (Linear Regression Moving Average) for pre processing, wavelet and PCA (Principle Component Analysis) for feature extraction, and SVM (Support Vector Machines) for analysis method. Result of pattern recognition that value highest level of recognition is show with SVM method with a 100% of accuracy. Whereas, pattern recognition of unlearn data, the obtained accuracy is 98.33%.