Recognition of Foodborne Pathogens Terahertz Spectrum Based on Convolutional Neural Network

Abstract

Foodborne pathogenic bacteria detection is an important work, the traditional laboratory identification method, will cause irreversible damage to the sample, and the detection cycle is long. Compared with traditional biological detection methods, the greatest advantage of terahertz spectroscopy detection is that it can do non-destructive detection of samples, and terahertz spectroscopy technology, combined with computer systems, can reduce analysis time, improve recognition accuracy, and maximize the extraction and mining of effective information in the spectrum. This paper establishes an intelligent foodborne pathogenic bacteria terahertz spectroscopic recognition model, The results showed that the recognition accuracy of pathogens was 99.76%.