Potential of multispectral imaging technology for rapid and

non-destructive determination of the microbiological quality of beef

filets during aerobic storage

在不同温度条件下(0, 4, 8, 12和16 °C)监控无氧包装肉片腐烂实验中，我们对多光谱成像系统（Videometer Lab）的性能进行了评估。从肉样表面采集了可见光和短波近红外区域(405–970 nm)的光谱数据以及相关的微生物学数据（对数计数），包括总生菌数(TVCs），假单胞菌属（Pseudomonas spp.）和热死环丝菌Brochothrix thermosphacta.）。基于总生菌落数，我们采用了质量定性分析法(PLS-DA) 来区分在3个微生物质量级别的肉样，分别为Class 1 (TVC b 5.5  log10 CFU/g), Class 2 (5.5  log10 CFU/g b TVC b 7.0  log10 CFU/g)和 Class 3 (TVC N 7.0  log10 CFU/g).另外，建立了PLS回归方法模型对在储存中的肉品微生物菌落数进行定量估算。两种情况下，在中间存储温度区间（2和10°C），利用不同批次的肉品，采用独立实验对模型进行验证。结果显示多光谱性能表型先进， 3个质量级别的肉样分级模型校准和模型验证的总正确率范围为91.8%-80%。为进行定量估算，用于建模和验证的所计算的观测和预估菌落计数的回归系数在0.90-0.93到0.78–0.86之间，这与微生物相关。另外，对于Pseudomonas spp., B. thermosphacta, 和 TVC的观测值和预估值之间的平均偏差相应为11.6%, 13.6%和 16.7%。结果表明多光谱成像技术是较有潜力的快速、无损肉品微生物品质检验技术，Videometer Lab多光谱测量系统是根据此技术开发出的科研和应用级研究利器，请详见产品链接。全文Potential of multispectral imaging technology.pdf。

The performance of a multispectral imaging system has been evaluated in monitoring aerobically packaged beef filet spoilage at different storage temperatures (0, 4, 8, 12, and 16 °C). Spectral data in the visible and  short wave near infrared area (405–970 nm) were collected from the surface of meat samples and  correlated with microbi- ological data (log counts), for total viable counts (TVCs), Pseudomonas spp., and  Brochothrix thermosphacta. Qual- itative analysis (PLS-DA)  was  employed for  the discrimination of meat samples in  three microbiological quality classes based on the  values of total viable counts, namely Class 1 (TVC b 5.5  log10 CFU/g), Class 2 (5.5  log10 CFU/g b TVC b 7.0  log10 CFU/g), and  Class 3 (TVC N 7.0  log10 CFU/g). Furthermore, PLS regres- sion  models were developed to provide quantitative estimations of microbial counts during meat storage. In both cases model validation was  implemented with independent experiments at intermediate storage tem- peratures (2 and  10 °C) using different batches of meat. Results demonstrated good performance in classifying meat samples with overall correct classification rate for the  three quality classes ranging from 91.8% to 80.0% for model calibration and  validation, respectively. For quantitative estimation, the calculated regression coeffi- cients between observed and  estimated counts ranged within 0.90–0.93 and  0.78–0.86 for model development and  validation, respectively, depending on the microorganism.Moreover, the  calculated average deviation be- tween observations and estimations was  11.6%, 13.6%, and  16.7% for Pseudomonas spp., B. thermosphacta, and TVC, respectively. The results indicated that multispectral vision technology has  significant potential as a rapid and  non-destructive technique in assessing the microbiological quality of beef  fillets.