Screening of pork freshness feature wavelengths and optimisation of nondestructive intelligent detection device

In modern culture, food safety and quality control have become increasingly important. The safety and quality of pork is closely tied to the physical well-being and overall quality of life of individuals. When it comes to food safety supervision, the traditional method frequently suffers from issues such as low efficiency, lengthy cycles, and expensive operating costs. Based on the experimental platform of fresh pork supernatant, the characteristic wavelengths of each index of pork freshness were extracted for the purpose of this study. The six characteristic wavelengths that had the highest association with pork freshness were chosen for further investigation. We analysed the changes in freshness index TVB-N concentration, pH value, and colour (L*, a*, b*) of fresh pork as a function of storage duration, as well as the spectrum information of the supernatant. Using CARS and SPA methods, we were able to identify the characteristic wavelengths that are associated with fresh pork in the spectrum of fresh pork supernatant. There was a significant improvement in the performance of the PLS model that was based on unique wavelengths. When the SNV-CARS-SPA preprocessing is complete, The values of the parameters TVB-N content, pH value, Rp of L*, a*, and b* were 0.996, 0.961, 0.731, 0.922, and 0.839, respectively. The RMSEP values were 0.388 mg/100g, 0.067, 2.377, 0.832, and 0.567 with respect to the RMSEP. 540 nm, 580 nm, 680 nm, 730 nm, 760 nm, and 860 nm were the six feature wavelengths that had the best connection with the freshness of pork. Based on the characteristic wavelength that was discovered in the research described above, as well as the software and hardware that needed to be improved in the laboratory handheld device, the miniature spectrometer of the handheld device was replaced with a multi-spectral sensor. Additionally, the light source module and control module were redesigned, and the structure of the device, including the cooling unit, power supply unit, acquisition probe, and shell, was improved. We have changed the software of the gadget such that it is based on the Arduino system. The stability of the gadget is superior to that of the enhanced device, which brings us to our final point. There were established the prediction models of the fresh pork freshness indices, which included the TVB-N content, pH value, L*, a*, and b*. The optimal models were obtained following SNV preprocessing, and the prediction models were higher for TVB-N content, pH value, a* and b* indexes. Following this, the optimal models were obtained. The best model of TVB-N content, pH value, L*, a*, and b* had Rp values of 0.970, 0.937, 0.805, 0.908, and 0.915, respectively, when measured under the six distinctive wavelengths. In that order, the RMSEP values were as follows: 0.874 mg/100g, 0.097, 1.972, 1.514, and 0.729. Before the enhancement, the portable fresh pork freshness detection device was superior to other indexes, except for the index brightness L*, which has an impact on the ability to differentiate between freshness and other factors. According to the findings, the upgraded device had the capability of determining whether fresh pork was fresh or not, and it predicted the freshness index TVB-N with a higher degree of precision than the device that had not been advanced.