Quality Classification of Chili Sauce Using Electronic Nose with Principal Component Analysis

Abstract

Chili Sauce is food products made from tomatoes, chili, and additives. The quality classification of chili sauce is very important to help consumers choose the products by their qualities. In this research, Electronic Nose (E-nose) is used to classify few quality grade of chili sauce. E-nose has ability to analyze the sample based on its aroma. The data acquisition is sampled and Data processing is applied through few step including pre-signal processing using Baseline manipulation method, feature extraction using maximum value method and data analysis using Principal Component Analysis Method. The results showed that the electronic nose with 5 sensor array i.e. TGS 2620, TGS 813, TGS 822, TGS 2600 and TGS 2620 give the difference responses from different the quality based on its aroma. The similarity and dissimilarity of sensor responses due to the composition of the chili sauces explored by multivariate pattern recognition techniques based on principal component analysis (PCA). It can be concluded that by using chosen feature in sensor responses of various sauce samples, PCA can be successfully used to reveal the clusters existing in chili sauces according to their organic composition. PC1 give the accuracy of 92% and PC2 give 5.8%, so the PCA give accuracy value of 97.8%.