【目的】建立有效区别毛橘红与光橘红2种药材的识别模型。【方法】收集不同产地的23批化橘红药材样品的指纹图谱,采用主成分分析法提取主成分,利用BP神经网络进行模式识别。【结果】建立了有效识别毛橘红和光橘红的神经网络模型,有效识别率超过91.3%,其中毛橘红均能被正确识别。【结论】神经网络技术可有效识别出道地药材毛橘红。 【Objective】 To establish a discriminative model that distinguishes two medicinal materials, namely, the mandarin orange and the mandarin orange. 【Method】 The fingerprints of 23 batches of Cymbidium officinale collected from different areas were collected, and the principal components were extracted by principal component analysis (PCA) and the patterns were identified by BP neural network. 【Result】 The neural network model was established to effectively identify the tangerine and the tangerine. The effective recognition rate was over 91.3%. The tangerine red could all be correctly identified. 【Conclusion】 Artificial neural network technology can effectively identify the dermatophyte red orange.