Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Abstract. In order to realize the remote real-time non-contact detection of freshness changes of pork in the cold room (0~4℃), this study designed a remote intelligent sensing system based on multispectral technology. By comparing and analyzing the variation law of pork reflection spectrum of ring probe and condensing probe at different detection heights, it was finally determined that the multispectral system with condensing probe combined with 5 mm detection height was the most suitable for non-contact acquisition of pork's diffuse reflectance spectrum. Then, the hardware system is designed in combination with the Internet of Things module, and the real-time control, data analysis and transmission software of the device is developed by using C language. The constructed system was used to collect the diffuse reflectance spectra of pork samples in the cold room, and two modeling analysis methods, PLSR and MLR, were compared. The results show that the results of the two modeling methods are not very different, and SNV preprocessing can improve the performance of the model. Among them, the SNV-MLR model had the best prediction effect on pork TVB-N content, the prediction correlation coefficient was 0.8103, and the prediction root mean square error was 1.7240 mg/100g. Through the Internet of Things module, the detection results can be updated and displayed in real time on the remote control terminal. The experimental verification shows that the intelligent sensing system designed in this paper can meet the monitoring requirements of the change of pork freshness in the cold room, and provides a reference for the realization of remote real-time monitoring of livestock and poultry meat quality in the cold room.