Evaluation of Mutton Adulteration under the Effect of Mutton Flavour Essence Using Hyperspectral Imaging Combined with Machine Learning and Sparrow Search Algorithm

Abstract

The evaluation of mutton adulteration faces new challenges because of mutton flavour essence, which achieves a similar flavour between the adulterant and mutton. Hence, methods for classifying and quantifying the adulterated mutton under the effect of mutton flavour essence, based on near-infrared hyperspectral imaging (NIR-HSI, 1000–2500 nm) combined with machine learning (ML) and sparrow search algorithm (SSA), were proposed in this study. After spectral preprocessing via first derivative combined with multiple scattering correction (1D + MSC), classification and quantification models were established using back propagation neural network (BP), extreme learning machine (ELM) and support vector machine/regression (SVM/SVR). SSA was further used to explore the global optimal parameters of these models. Results showed that the performance of models improves after optimisation via the SSA. SSA-SVM achieved the optimal discrimination result, with an accuracy of 99.79% in the prediction set; SSA-SVR achieved the optimal prediction result, with an RP2 of 0.9304 and an RMSEP of 0.0458 g·g−1. Hence, NIR-HSI combined with ML and SSA is feasible for classification and quantification of mutton adulteration under the effect of mutton flavour essence. This study can provide a theoretical and practical reference for the evaluation and supervision of food quality under complex conditions.

Adulteration Detection of Pork in Mutton Using Smart Phone with the CBAM-Invert-ResNet and Multiple Parts Feature Fusion

“…It strengthens the feature differences of the images and effectively improves the accuracy of the model through the adaptive allocation of feature weights. With the effects of additives such as mutton flavor essence and colorant, the characteristics of adulterated mutton meat with different pork content show little difference [26]. Therefore, the feature differences between adulterated mutton with different pork content can be strengthened by adding a CBAM attention mechanism to the ResNet50 network.…”

“…Although the inverted residual structure meets the requirements of a lightweight model, its ability to learn features with small differences is limited. There is little difference in the characteristics of adulterated mutton with different contents of pork under the influence of additives such as mutton flavor essence and coloring agent, and it is still difficult to accurately predict its content by existing models [26]. The convolutional block attention module (CBAM) [27] can effectively improve the accuracy of the model by using the spatial and channel features of the images to redistribute the feature weights and strengthen the feature differences of the image.…”

Research on Automatic Classification and Detection of Mutton Multi-Parts Based on Swin-Transformer

“…Common technical methods used for mutton detection in recent years mainly include image processing, spectroscopic techniques combined with machine learning classification, and regressors. By extracting and analyzing multi-dimensional feature information such as color, texture, contour, protein, water content, and total volatile basic nitrogen (TVB-N) in mutton sample images, and establishing the relationship with mutton freshness [ 6 , 7 , 8 ], tenderness [ 9 , 10 , 11 ], authenticity [ 12 , 13 ], pH [ 14 , 15 ], storage time [ 16 , 17 ], and other indicators, these methods allow effective and nondestructive detection of mutton quality. Although the aforementioned technical methods can achieve high detection accuracy, they also have shortcomings such as cumbersome artificial extraction of sample features, poor generalization of models, and low adaptability, which are not suitable for the classification and detection of mutton with multiple categories, large quantities, and complex natural feature expression.…”

Online Machine Vision-Based Modeling during Cantaloupe Microwave Drying Utilizing Extreme Learning Machine and Artificial Neural Network

“…It is less prone to being trapped in local optima. In recent years, ELM has also been gradually applied in the food-processing field and has shown good performance in some studies [ 35 , 36 ]. However, there have been no studies found that utilize ELM with material shrinkage as input to model and predict water content during microwave drying processes.…”