Non-destructive detection of tea stalk and insect foreign bodies based on THz-TDS combination of electromagnetic vibration feeder

Abstract

Background and objectives: Current technology of X-ray imaging can recognize hard foreign materials (FMs) such as metal and high-density plastic. However, low-density foreign bodies are still a challenge for food quality and safety assessment. Materials and methods: An electromagnetic vibration feeder aided by terahertz time-domain spectroscopy (THz-TDS) and imaging was investigated for non-destructively detecting tea stalk and insect FMs mixed with tea leaves. Results: THz time-domain signals were employed directly to develop the K-nearest neighbor model with a precision of 100%, accuracy of 95.6% and recall of 98.7% in predicting the unknown samples. High contrast THz-TDS images were obtained by the separation method for the samples using electromagnetic vibration feeder. The characteristic parameters of the ratio of maximum length (L) to maximum width (W) and hue extracted from THz-TDS images indicated significant difference between tea leaves and FMs. Conclusions: The results suggested that electromagnetic vibration feeder combination with THz-TDS was feasible for detecting FMs in finishing tea products.