Eggshell surfaces are highly susceptible to Escherichia coli and Salmonella enteritidis, demanding rapid, non-destructive detection. This study evaluated hyperspectral imaging (HSI) for pathogen discrimination. Visible-near-infrared (Vis-NIR) and short-wave infrared systems acquired data from eggs contaminated at 10³–10⁵ CFU/mL. Raw Vis-NIR spectra with competitive adaptive reweighted sampling provided optimal features. The pathogens showed opposite reflectance trends, attributed to biofilm structural differences affecting light scattering and absorption. Classification models achieved up to 91.11% accuracy in pathogen identification and 88.89% accuracy in differentiating E. coli levels. Vis-NIR-HSI shows great potential for rapid, non-destructive eggshell pathogen detection, with a clarified mechanistic framework.

This study was conceived in order to clarify the effect of the freshness of raw eggs on the gel quality and the ultrastructure of eggshell membrane during the curing of preserved eggs. Duck eggs with different freshness were used to make preserved eggs in the same way. The ultrastructure of eggshell membrane was observed by scanning electron microscopy (SEM), and Image J software was used to process the image data to obtain eggshell membrane status. Besides, eggshell membrane porosity, and gel pH, water content, hardness and elasticity were measured and the correlation between them was analyzed. The results showed that during the curing process, the pH of preserved egg white decreased first, then slowly increased, and finally decreased sharply; the hardness and elasticity of preserved egg gels showed a fluctuating trend. The lower the freshness of raw eggs, the higher the porosity of eggshell membrane, and the shorter the curing cycle of preserved eggs. The porosity of eggshell membrane had a significantly negative correlation with the freshness of raw eggs (P ＜ 0.01), and a significantly positive correlation with gel hardness (P ＜ 0.01) and elasticity (P ＜ 0.05). As a result, a decrease in the freshness of raw eggs could affect eggshell membrane as the permeation channel for the brine, which could in turn lead to variability in the quality of preserved eggs. Moreover, the yield of preserved eggs decreased with a decrease in the freshness of raw eggs. Therefore, the freshness of raw eggs is closely related to the quality of preserved eggs. This study provides a new idea for controlling the quality change of preserved eggs during the pickling process.

In this study, hyperspectral imaging was used to visualize the water content and springiness of preserved egg gels and to predict different quality grades. First, the hyperspectral information of qualified and high quality preserved eggs was collected, and their water content and springiness were measured. Then, the original spectral data were transformed by Savitzky-Golay (S-G), first derivative (FD) or Savitzky-Golay and first derivative (S-G-FD) to analyze their correlation with water content and springiness values. We identified and excluded outliers by Monte Carlo-partial least squares (MCPLS), and partitioned the sample sets by sample set partitioning based on joint X-Y distance (SPXY). The characteristic wavelengths were selected using successive projection algorithm (SPA) and the uninformative variable elimination (UVE) method, and a multiple stepwise regression model (MSR) was established to predict the water content and springiness of preserved eggs. The results showed that UVE-MSR was the optimal model for predicting water content. Its determination coefficient and root-mean-square error (RMSE) were 0.882 and 0.583, respectively, and its relative percent deviation (RPD) was 2.1. The optimal model for predicting springiness was SPA-MSR, whose determination coefficient and RMSE were 0.903 and 0.348, respectively, and whose RPD was 2.2. Then, the models were used to calculate the water content and springiness for each pixel in the hyperspectral image, and a visual distribution map was generated for the visual detection of the water content and springiness of preserved eggs. Finally, the competitive adaptive weight sampling method was used to select the characteristic wavelengths, and a back propagation (BP) neural network model was established for quality prediction. The total recognition accuracy of the model was 98.3%.

The near-infrared spectral data of salted duck eggs, prepared from Gaoyou Ma duck eggs, were collected during the whole curing period and based on them, a model for nondestructive and rapid detection of the key quality indicators of salted duck eggs (yolk moisture content, yolk sodium chloride concentration and salted egg yolk index). In order to reduce the influence of other external factors on spectrum acquisition, various spectral preprocessing methods such as multiplicative scatter correction and normalization combined with three feature selection algorithms including competitive adaptive reweighted sampling (CARS), successive projections algorithm (SPA) and uninformative variables elimination (UVE) were used to establish partial least squares regression (PLSR) models. Furthermore, CARS or UVE combined with SPA was used to establish a more robust model. It was found that the optimal band selection method for the three quality indicators of salted duck eggs was UVE combined with SPA, which had the best overall performance. Comparative analysis showed that the optimal model structures for egg yolk moisture content, egg yolk sodium chloride concentration, and salted egg yolk index were standardization-UVE + SPA-PLSR, Savitzky-Golay-UVE + SPA-PLSR, and mean centering-UVE + SPA-PLSR, respectively. The correlation coefficients were 0.9334, 0.8978 and 0.9286 for the training set (R_c), and 0.9276, 0.9085 and 0.9163 for the prediction set (R_p), respectively. The spectral model established in this study can allow the non-destructive detection of the physical and chemical indicators of salted duck eggs during the salting period.

In this study, hyperspectral imaging was used for nondestructive detection of preserved eggs at different maturity levels during the pickling period. First, the optimal waveband was determined based on the one-dimensional spectra and two-dimensional correlation spectra in the time-series mode, separately. Then, the modeling effects of traditional machine learning and the improved ResNet20_SE model in the optimal waveband were compared, and the results showed that the improved ResNet20_SE model was better; the overall recognition accuracy was 97.29% for the synchronous spectral dataset, and the average detection speed for a single image was 24.62 ms. Finally, the better synchronous spectral dataset ResNet20_SE model was applied to the hyperspectral pixel spectral image to calculate the value of each pixel point, and a pseudo-color technique was used for the visual detection of the spatial distribution of preserved egg maturity during the pickling process. The results of this study showed that hyperspectral imaging combined with deep learning is useful for nondestructive detection of preserved egg maturity during curing, which can lay a theoretical foundation for high-throughput online sorting of preserved egg maturity in the future.

To accurately and efficiently detect dead caged laying ducks, thereby reducing reliance on manual inspection, this study proposes a method that integrates infrared thermography with deep learning technology. A lightweight object detection algorithm is developed, utilizing YOLO v8n as the baseline model. The backbone network is replaced with StarNet, which is based on “Star Operate”. Additionally, the C2f-Star structure is designed by combining the Star Block from StarNet with the C2f module, and it is inserted into the Neck structure of the baseline model. Lightweight module L-SPPF replaces the SPPF module in the baseline model to enhance feature augmentation. Furthermore, a lightweight shared convolutional detection head, termed SCSB-Head, is introduced to reduce computational complexity. These improvements collectively form a lightweight object detection algorithm named SLSS-YOLO. Experimental results show that SLSS-YOLO achieves mAP@50%-95%, precision, and recall scores of 80.50%, 99.44%, and 98.46%, respectively. Compared to the baseline model, these metrics improve by 1%, 1.98%, and 0.26%, respectively. In terms of model size and detection speed, SLSS-YOLO has 1.44 M parameters and 4.6 G FLOPs, achieving an FPS rate of 134.9 f/s. This represents a reduction of 52.16% and 43.90% in parameters and FLOPs, respectively, while increasing FPS by 5.4 f/s compared to the baseline model. Moreover, an object tracking model is constructed using SLSS-YOLO and Hybrid-SORT. Tracking tests demonstrate that Hybrid-SORT achieves zero ID-Switches, with a detection speed of 10.9 ms/f. It outperforms Bot-SORT, ByteTrack, Deep OC-SORT, and OC-SORT in terms of tracking performance. Therefore, the proposed thermal infrared detection method can effectively identify and track dead ducks in complex cage environments, providing a reference for automated inspection in caged duck farms.