This study aimed to determine the integrated freshness index (IFI) of eggs using Vis-NIR spectroscopy and optimized support vector regression, which gave the first insight into the freshness quality of eggs from the biochemical essence of quality changes. In this work, Vis-NIR transmission spectra of brown-shell and pink-shell egg samples were analyzed between 500 nm and 900 nm. Standard normal variables (SNV) were used to normalize the spectral data, and the Shuffled Frog Leaping Algorithm (SFLA) and Competitive Adaptive Reweighted Sampling (CARS) were used to choose the optimal wavelengths. The quantitative analysis model of IFI was developed using a support vector regression (SVR) that was optimized using Grid Search (GS), Genetic Algorithm (GA), and Particle Swarm Optimization (PSO). After conducting a comparative analysis, it was determined that the GA-SVR model based on 63 wavelengths screened by the SFLA best predicted IFI with a training set coefficient of determination (R_c²) of 0.900, root means square error (RMSEC) of 0.005, a prediction set coefficient of determination (R_p²) of 0.816, root mean square error (RMSEP) of 0.012 and relative analysis error (RPD) of 2.077. The results demonstrate that the model can be used to simultaneously perform nondestructive detection of two distinct egg IFI variants, suggesting broader applicability and enhanced model reliability.

The hardness of passion fruit is a critical feature to consider when determining maturity during post-harvest storage. The capacity of near-infrared diffuse reflectance spectroscopy (NIRS) for non-destructive detection of outer and inner hardness of passion fruit epicarp was investigated in this work. The passion fruits’ spectra were obtained using a near-infrared spectrometer with a wavelength range of 10,000–4000 cm⁻¹. The hardness of passion fruit’s outer epicarp (F1) and inner epicarp (F2) was then measured using a texture analyzer. Moving average (MA) and mean-centering (MC) techniques were used to preprocess the collected spectral data. Competitive adaptive reweighted sampling (CARS), successive projection algorithm (SPA), and uninformative variable elimination (UVE) were used to pick feature wavelengths. Grid-search-optimized random forest (Grids-RF) models and genetic-algorithm-optimized support vector regression (GA-SVR) models were created as part of the modeling process. After MC preprocessing and CARS selection, MC-CARS-Grids-RF model with 7 feature wavelengths had the greatest prediction ability for F1. The mean square error of prediction set (RMSE_P) was 0.166 gN. Similarly, following MA preprocessing, the MA-Grids-RF model displayed the greatest predictive performance for F2, with an RMSE_P of 0.101 gN. When compared to models produced using the original spectra, the R²_P for models formed after preprocessing and wavelength selection improved. The findings showed that near-infrared spectroscopy may predict the hardness of passion fruit epicarp, which can be used to identify quality during post-harvest storage.

In order to avoid damaging viscera during poultry evisceration and enhance the economic value of poultry products, this paper proposes a predictive method for poultry carcass visceral dimensions based on 3D point cloud and a Genetic Algorithm-based Wavelet Neural Network (GA-WNN). In this study, a data set of poultry carcasses was obtained through the use of 3D point cloud scanning equipment combined with reverse engineering software. The inputs and predicted targets of the model were determined through correlation analysis of various carcass dimensions. Then, a prediction model of poultry visceral size (GA-WNN) was built by K-fold cross validation method, Genetic Algorithm and Wavelet Neural Network (WNN). By comparing the prediction results and analyzing Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) of the six models, it was determined that the GA-WNN model had the best prediction results. Finally, in order to verify the generalizability of the method, generalizability experiments were conducted on different breeds of poultry, which proved that the method of this study had superior generalizability ability. In the comparative analysis of the six models, the MAPE and RMSE of the GA-WNN model for the prediction of the three visceral dimensions were the lowest except for the RMSE for the prediction of visceral length. Compared with the largest of the two kinds of errors, the MAPE and RMSE for the prediction of the position of the upper end of the left liver by the method of this study were lower by 5.56% and 0.915 cm, respectively, and the prediction effect had a significant advantage. The experimental results showed that the model built in this paper based on 3D point cloud and GA-WNN network can accurately predict the size of the viscera of poultry carcasses, thus providing theoretical references for the automated evisceration technology without damaging the viscera.

As a key component connecting the fuel injection pump and the fuel injector, the engine high—pressure fuel pipe often suffers from problems such as scratches, pit defects, and dimension deviation of the end—head bulge during the production process. However, although traditional manual inspection is simple and convenient, it is slow and not conducive to the needs of information integration and automated production. To address the issue of difficulties in detecting the end—head defects and bulge dimensions of automotive high—pressure fuel pipes, this study developed an end—head defect detection system based on LabVIEW software, which adopts a rectangular rake instead of the traditional linear rake. The LabVIEW graphical programming language was used to construct the control program, and Vision Assistant was used for image processing and analysis. This system can accomplish the detection of scratches and pit defects on the outer surface of the fuel pipe and the measurement of the end—head bulge dimensions. Experimental results indicate that the detection rates of scratches and pits are 97% and 96% respectively. Compared with manual measurement, the dimension measurement of long end—heads and short end—heads saves 86% and 90% of the working time respectively. Compared with traditional edge—point detection, the detection speed of long end—heads and short end—heads has increased by 1.5 times. This provides support and a theoretical basis for the defect detection of automotive high—pressure fuel pipes and the research of relevant systems.