In this study, confocal Raman microspectroscopy was used to detect lipids in microalgae rapidly and non-destructively. Microalgae cells were cultured under nitrogen deficiency. The accumulation of lipids in Scenedesmus obliquus was observed by Nile red staining, and the total amount of lipids accumulated in the cells was measured by gravimetric method. The signals from different microalgae cells were collected by confocal Raman microspectroscopy to establish a prediction model of intracellular lipid content, and surface scanning signals for drawing pseudo color images of lipids distribution. The images can show the location of pyrenoid and lipid accumulation in cells. Analyze Raman spectrum data and build PCA-LDA model using four different bands (full bands, pigments, lipids, and mixed features). Models of full bands or pigment characteristic bands were capable of identifying S. obliquus cells under different nitrogen stress culture time. The prediction accuracy of model of lipid characteristic bands is relatively low. The correlation between the fatty acid content measured by the gravimetric method and the integral Raman intensity of the oil characteristic peak (1445 cm−1) measured by Raman spectroscopy was analyzed. There was significant correlation (R 2 = 0.83), which means that Raman spectroscopy is applicable to semi-quantitative detection of microalgal lipid content.
Hyperspectral images are widely used in the food industry as a fast and non-destructive analytical technique. Cantonese sausage has a long history and is a very old food production and meat preservation technology. According to the physical and chemical characteristics of the sausage, the Chinese business industry standard SB/t10003-92 divides the sausage into three grades, which are called excellent grade, level 1, and level 2. In this paper, k-means is adopted first to separate two parts of the meat adaptively to improve the discriminant rate. The hyperspectral information of the near-infrared band is extracted by successive projections algorithm (SPA). The multiple linear regression (MLR) and partial least squares regression (PLSR) algorithms are used to classify the sausage grade. The experimental results show that the lean meat and fat of the sausage have different characteristics in the near-infrared band, and the modeling results have higher accuracy and anti-interference after separating lean meat and fat meat. The best model of sausage classification is using SPA-MLR method to model the fat region of Cantonese sausage; the prediction accuracy of which is 100%. It was found that the modeling results of fat were better than lean meat in both PLSR and SPA-MLR, which indicated that there were obvious differences in fat composition among different grades of sausage, and the fat of sausage was more suitable for classification.