挥发性盐基氮（TVB-N）含量是评价猪肉新鲜度的重要指标。尝试融合光谱和成像技术检测猪肉中TVB-N含量。实验以不同新鲜度的猪肉样本为研究对象，同时采集近红外光谱数据和图像数据，并对其分别进行特征提取和主成分分析，利用反向传播神经网络构建猪肉TVB-N的定量预测模型。实验结果表明，融合模型要优于单一技术模型，模型交互验证均方根误差（RMSECV）为1.2975，对独立样本预测时相关系数达到0.957。研究表明基于光谱和成像融合技术检测猪肉中TVB-N含量是可行的，检测结果的准确性和稳定性较单一技术有所提高。

Total volatile basic nitrogen (TVB-N) content is an important index in evaluating the pork′s freshness. We attempt to determine TVB-N content in pork by multi-information fusion technique based on spectroscopy and imaging. In experiment, pork samples with different freshness are studied, and the near-infrared spectra and images are collected simultaneously. Principal component analysis (PCA) is implemented on these feature variables from image and spectral information, and a prediction model is developed by the back-propagation artificial neural network (BP-ANN). Experimental results show that the model based on multi-information fusion is superior to the model based on a single technique, the root mean square error of cross-validation in the model is 1.2975, and the correlation coefficients is 0.957 when the model is tested by independent samples in the prediction set. The overall results show that it is feasible to measure TVB-N content in pork by multi-information fusion based on spectra and imaging, and the performance from the model based on multi-infusion fusion is better than that from the model based on a single technique.