Research and application of multi sensor data fusion technology in the electronic tongue

Multi- sensor data fusion technology based electronic tongue system was designed in this paper to avoid the disadvantages in traditional detection methods such as complex operation, long operating time and too many man- made factors.The system consists of potentiometric sensors and voltammetric sensors.The eigenvalue of the response curve detected by sensor array was calculated to analyze the freshness of pork. First, potentiometric sensor and voltammetric sensor feature fusion technology was applied to extract the eigenvector of pork.Then the extracted eigenvector was optimized by stepwise discriminant method.In the end, bayes discriminant method was used to analyze the freshness changes of pork. According to the results, the distinctions of pork of 5 groups with detect time interval of 12 hours was identified precisely, which proved that electronic tongue system provided an effective way to pork freshness intelligent detection.