Key Technology Research On Quality And Safety Detection For Liquid Food Containing Fat Based On Photoelectric Technology

The quality and safety detection of fat-containing liquid foods is a key and frontier topic in the field of food testing.However,the current food measurement technology is insufficient in terms of miniaturization,instrument cost and measurement speed.The detection method based on photoelectric technology and sensors has the advantage of fast response,high accuracy and low cost,and can realize non-contact measurement.In recent years,the systems based on photoelectric sensors tend to be miniaturized and automated.Therefore,it is of great significance to study the quality and safety of fat-containing liquid foods based on photoelectric technology.Based on photoelectric sensing technology and image processing technology,the main components,food additives and biological contamination of fat-containing liquid food are studied in this paper,which aims to provide reliable technical support for the establishment and improvement of quality and safety detection system for lipid-containing liquid foods.The main innovation research contents are shown as follows:(1)The measurement system based on ring-shaped light sensor is designed for the detection of fat content in fat-containing liquid food.In order to realize the near-field real-time detection,a ringshaped photoelectric sensor system with uniformly distributed light source is designed.The measurement method is to establish the model between the absorbance of the sample and the fat content in the fat-containing liquid food,load the model into the microprocessor to complete the calibration of the system,and the fat content can be finally displayed in real time.Meanwhile,the system is evaluated by measurement system analysis,which shows the reliability of the measurementsystem.(2)The measurement system based on Y-type optical fiber sensor is designed for the detection of fat content in fat-containing liquid food.In order to achieve real-time detection with long distance,a measurement system based on Y-type optical fiber sensor is designed.The single optical probe allows vertical incidence and vertical reception of the optical signal,which effectively separating the exiting light from the incident light,and reducing the stray light.Meanwhile,the system is evaluated by measurement system analysis,which shows the reliability of the measurementsystems.(3)The W-type optical fiber sensor system for measuring fat content in fat-containing liquid food is designed and proposed.Based on some defects of the Y-type optical fiber sensor and ringshaped light sensor,a W-type optical fiber sensor with a dual probe structure is proposed for transmission of light.It can not only achieve the long-distance measurement and reduce the interference of stray light signals,but also eliminate the specular reflection signal,thereby improving the measurement accuracy,and finally achieving the measurement of fat content in lipidcontaining liquid foods.(4)Based on two-dimensional and three-dimensional fluorescence spectroscopy,the content of FWA(0-1?g/ml)in the fat-containing liquid food is quantitatively determined.In this chapter we make soybean milk as the sample.For the two-dimensional fluorescence spectra,the model is established between the fluorescence intensity and the FWA content,and the detection of the fluorescent whitening agent in the soybean milk can be realized.For the three-dimensional fluorescence spectra,we combine the wavelet analysis image processing method with support vector regression(SVR)machine to realize the detection of FWA in soybean milk.The results show that the proposed methods have high sensitivity,and the detection results of three-dimensional fluorescence spectroscopy is better than those of two-dimensional fluorescence spectroscopy.(5)Based on the weak light photoelectric sensor system and the image processing method combined with wavelet analysis method,the content of FWA(0.02-0.5mg/ml)in the liquid food containing fat is quantitatively determined.In this study we also make soybean milk as the sample.For the weak fluorescent images of the samples containing FWA,the wavelet moments are extracted.In addition,the distance classification method is used to realize the accurate measurement of the FWA in soybean milk.(6)Based on the measurement system of BARDOT laser sensor,a quick method for effectively detecting Listeria monocytogenes in fat-containing liquid food is proposed.In this study we make milk as the object.First,the image database of Listeria and non-Listeria is established by the BARDOT system.The scattering images of colonies from milk samples are identified by the BARDOT system for the detection of Listeria monocytogenes.In order to verify the reliability of the method,the detected colonies are verified by PCR and qPCR.The results show that the proposed method can effectively and accurately detect Listeria in milk and realize non-destructive measurement.The BARDOT system can detect Listeriain milk samples in 34-40 h,whereas the standard USDA Food Safety and Inspection Service method requires about 72 h.