Cao Zhihang, Wu Xiaohong, Wu Bin, Zhang Zexi, Sun Jun
Mengxi Honors College, Jiangsu University, Zhenjiang, 212013, China.
School of Electrical and Information Engineering, Jiangsu University, Zhenjiang, 212013, China.
Curr Res Food Sci. 2025 Jun 29;11:101131. doi: 10.1016/j.crfs.2025.101131. eCollection 2025.
This study aims to rapidly and non-destructively identify the geographical origin of black beans () using a portable near-infrared (NIR) spectrometer, addressing the challenge of distinguishing black beans due to significant regional variations in quality. A total of 400 black bean samples were collected from five regions in China. To improve classification accuracy, a novel model combining uncorrelated discriminant transform (UDT) with extreme gradient boosting (XGBoost) was proposed for feature extraction and classification. When evaluated with k-nearest neighbor (KNN), naive Bayes (NB), and support vector machine (SVM) classifiers, UDT achieved accuracies of 96.25 %, 93.75 %, and 96.25 %, respectively, outperforming Foley-Sammon transform (FST) and discriminant principal component analysis (DPCA). The UDT + XGBoost combination achieved the highest classification accuracy of 100 %. For robust validation, a 5-fold cross-validation strategy was applied to the UDT + XGBoost model, achieving an average accuracy of 96.00 %. This study provides a reliable method for black bean origin traceability and authenticity.
本研究旨在使用便携式近红外（NIR）光谱仪快速、无损地识别黑豆的地理来源，以应对因品质存在显著区域差异而导致难以区分黑豆的挑战。共从中国五个地区收集了400份黑豆样本。为提高分类准确率，提出了一种将不相关判别变换（UDT）与极端梯度提升（XGBoost）相结合的新型模型用于特征提取和分类。当使用k近邻（KNN）、朴素贝叶斯（NB）和支持向量机（SVM）分类器进行评估时，UDT分别达到了96.25%、93.75%和96.25%的准确率，优于福利-萨蒙变换（FST）和判别主成分分析（DPCA）。UDT + XGBoost组合实现了最高的100%分类准确率。为进行稳健验证，对UDT + XGBoost模型应用了5折交叉验证策略，平均准确率达到96.00%。本研究为黑豆产地溯源和真伪鉴别提供了一种可靠方法。
