2014 28 5 0861 ~ 0868 Journal of Nuclear Agricultural Sciences 861 1000-8551 2014 05-0861-08 1 2 1 1 1 1 1 1 315211 2 315100 1 为建立一种快速有效的识别贝类的科学方法, 运用电子鼻区分不同种类和不同加热温度的贝类, 建立一个包含所有种类和温度的贝类模型, 通过判别函数法对该模型进行验证结果表明, 电子鼻能够 区分不同种类以及不同加热温度的贝类 ; 建立的贝类气味指纹模型的成功率可达到 95% 以上可知利用电子鼻建立贝类的气味指纹模型是可行的该模型准确灵敏, 可为食品特别是水产品的快速检测提 供了依据 贝类 ; 电子鼻 ; 气味指纹 ; 模型识别 DOI 10. 11869 /j. issn. 100-8551. 2014. 05. 10 1 1. 1 1-2 1. 2 3-5 1. 2. 1 样品处理 6 7 0. 2 g 8 9 70 80 90 100 110 120 150 30 min 0. 2g 30 min 5 1. 2. 2 电子鼻检测 AIRSENSE 2012-10-09 2014-01-10 03771665 E-mail sxtydingyuan@ 163. com E-mail suxiurong@ nbu. edu. cn
862 28 PEN3 10 2 200s 2. 1 199s 200s 2. 1. 1 不同温度下 6 种贝类挥发性气味指纹 1 1. 3 6 LDA WinMuster LDA LDA 91. 179% DFA LDA 70% ~ 80. 103% 90. 682% 81. 063% 90. 538% 88. 459% 85% 11 LDA 6 Fig. 1 1 LDA Linear discriminant analysis for different shellfish at different temperature conditions
5 863 6 2. 1. 2 同一温度下 6 种贝类挥发性气味指纹 2 90 150 6 LDA 90 150 LDA LDA 86. 741% 89. 984% 81. 943% 90 5 2. 2 LDA 3 3 3 LDA Fig. 3 Linear discriminant analysis for different shellfish Fig. 2 2 6 LDA Linear discriminant analysis for different shellfish at the same temperature conditions
864 28 Table 1 1 Actual kind Euclidean Distance Mahalanobis distance Identification of shellfish by Electronic nose Identification method Correlation Discriminant function method 99. 67% Moerella iridescens 99. 94% 95. 37% 5 cases were all 5 cases were all 5 cases were all 100. 00% 100. 00% identified as Moerella identified as Moerella identified as Moerella 5 cases were all identified as Moerella iridescens. Definiteness iridescens iridescens iridescens was 99. 67% 99. 94% 95. 37% 100. 00% 100. 00%. Scapharca subcrenata Scallop Tegillarca granosa 96. 22% 99. 89% 5 cases were all 5 cases were all 5 cases were all 99. 59% 99. 28% 99. 18% identified as Scapharca identified as Scapharca identified as Scapharca 5 cases were all identified as Scapharca subcrenata. subcrenata subcrenata subcrenata Definiteness was 96. 22% 99. 89% 99. 59% 99. 28% 99. 18%. 99. 94% 99. 99% 5 cases were all 5 cases were all 5 cases were all 100. 00% 100. 00% 100. 00% identified as Scallop identified as Scallop identified as Scallop 5 cases were all identified as Scallop. Definiteness was 99. 94% 99. 99% 100. 00% 100. 00% 100. 00% 91. 18% 93. 33% 5 cases were all 5 cases were all 5 cases were all 99. 04% 96. 78% 98. 20% identified as Tegillarca granosa identified as Tegillarca granosa identified as Tegillarca granosa 5 cases were all identified as Tegillarca granosa Definiteness was 91. 18% 93. 33% 99. 04% 96. 78% 98. 20% 99. 98% Ruditapes philippina 99. 99% 99. 97% 95. 86% 99. 97% 5 cases were all 5 cases were all 5 cases were all 5 cases were all identified as Ruditapes philippina. identified as Ruditapes philippina identified as Ruditapes philippina identified as Ruditapes philippina Definiteness was 99. 98% 99. 99% 99. 97% 95. 86% 99. 97% Cyclina sinensis 99. 98% 98. 57% 5 cases were all 5 cases were all 5 cases were all 88. 01% 90. 53% 90. 53% identified as Cyclina identified as Cyclina identified as Cyclina 5 cases were all identified as Cyclina sinensis. Definiteness sinensis sinensis sinensis was 99. 98% 98. 57% 88. 01% 90. 53% 90. 53% Fig. 4 4 Validation results of Moerella iridescens 2. 3 DFA 4 4 ~ 9
5 865 Fig. 5 5 Validation results of Scapharca subcrenata Fig. 6 6 Validation results of Scallop Fig. 7 7 Validation results of Tegillarca granosa DFA 95% DFA 1 3
866 28 Fig. 8 8 Validation results of Ruditapes philippina Fig. 9 9 Validation results of Cyclina sinensis Fig 10 10 Validation results of Sinonovacula constricta
5 867 10 3 LDA DFA 6 95% 1. GC-MS J. 2013 27 1 81-85 2. LDA 3 2011 array system J 6 12 13 5-13 14 J 2011 945 7 J. 2010 26 21 75 80 8 GC-MS 10 4 LDA J. 2013 27 1 75-80. D. 4 John - Erik H Frank L Jens P W Annette V. Detection of rancidity in freeze stored turkey meat using acommercial gas-sensor. Sensors and Actuators 2006 116 1-2 78-84 5 Tiina R Hanna-Leena A Tiina R Eija S Maria S Raija. Application of an electronic nose for quality assessment of modified atmosphere packaged shrimp meat J. Food Control 2006 17 1 6.. 24 7 941 -.. J. 2011 27 5 358-363 9. J. 2010 35 2 246-249. J. 2011 23 5 1029-1033. 11. J. 2012 26 2 311-316 12. J. 2009 8 9 58-60 13. J. 2011 32 4 273-280 14. J. 2007 31 S1 88-91
868 Journal of Nuclear Agricultural Sciences 2014 28 5 0861 ~ 0868 Establish an Identification Model for Shellfish with the Electronic Nose DING Yuan 1 XU Mao-qin 2 MIAO Fang-fang 1 CAI Jiang-jia 1 ZHOU Jin 1 ZHANG Chun-dan 1 LI Ye 1 SU Xiu-rong 1 1 School of Marine Sciences Ningbo University Ningbo Zhejiang 315211 2 Ningbo City College of Vocational Technology Ningbo Zhejiang 315100 Abstract Developing a scientific processing method to distinguish different shellfish effectively. Use the electronic nose to detect the different kinds and heating temperature of shellfish and establish an model for shellfish. In addition testing the Correctness of the model by discriminant function. The electronic nose is able to distinguish the different shellfish and different heating temperature and the flavor fingerprint model is confirmed accurately with higher than 95% success rate. With the establishment of the flavor fingerprint model electronic nose is able to quickly detect the species of different shellfish providing a basis testing for food especially aquatic products. Key words Shellfish Electronic nose Flavor fingerprint Model identification