嗅觉可视化技术对虾储藏期内新鲜度变化的研究  

Evaluation of Litopenaeus Vannamei Freshness Based on Olfaction Visualization Detection Technique

作　　者：黄星奕[1] 蒋飞燕[1] 周芳[1] 

机构地区：[1]江苏大学食品和生物工程学院,镇江212003

出　　处：《食品工业》2013年第1期10-12,共3页The Food Industry

基　　金：国家自然基金(31071549);公益性行业(农业)科技专项(201003008-04);博士点基金(20093227110007);江苏省青蓝工程(2008)资助项目;江苏省高校优势学科建设工程资助项目

摘　　要：新鲜度是虾类产品品质和加工适性的一个重要指标。研究利用自制的4×4可视化传感器阵列提取储藏在4℃下1～7 d的南美白对虾的挥发性气味信息。将提取的传感器阵列与虾顶空挥发性气体反应前后的颜色变化值作为特征值,对其进行主成分分析。构建以前7个主成分作为输入变量的虾新鲜度等级的BP神经网络判别模型。对训练集样本,模型的判别正确率达100%,对预测集样本,模型的识别正确率达91.30%。研究结果表明:利用嗅觉可视化技术来检测虾的新鲜度是可行的。Freshness is an important factor for quality evaluation and processing eligibility of shrimp products. Litopenaeus vannamei was taken as detection samples of shrimp in this research. A 4×4 colorimetric sensor array was developed to get the information of different volatile gases during storage of Litopenaeus vannamei samples which were stored at 4 ℃ for 1-7 d. The responses of sensor array, RGB （red-green-blue） color change values of the sensor array before and after exposed to the headspace volatile gases of Litopenaeus vannamei were analyzed with principal component analysis （PCA）. The former 7 principal components were used as inputs of the back-propagation artificial neural network （BP-ANN） to distinguish the freshness of Litopenaeus vannamei samples. For 47 samples in training set, the discriminating rate of the model reached 100% and for 23 samples in predicting set, the discriminating rate reached 91.30%. The results show that it is feasible to use colorimetric sensor array to evaluate shrimp freshness.

关 键 词：南美白对虾 嗅觉可视化 新鲜度 BP神经网络 

分 类 号：TS254.4[轻工技术与工程—水产品加工及贮藏工程]