基于KPCA和SOM网络的电子鼻大闸蟹新鲜度评价  

Appraising Method for Freshness of Crabs With Electronic Nose Based on KPCA and SOM Network

作　　者：朱培逸[1] 顾晓云[1] 杜洁[1] 徐本连[1] 鲁明丽[1] 

机构地区：[1]常熟理工学院电气与自动化工程学院,苏州215500

出　　处：《食品工业》2017年第2期200-203,共4页The Food Industry

基　　金：江苏省"青蓝工程"资助项目;江苏省"六大高峰人才"项目(2014-NY-021);常熟理工学院前瞻性项目(QZ1502)

摘　　要：针对大闸蟹新鲜度无损检测试验中,对于活体大闸蟹体内存在复杂、多变的生理指标,导致难以获得准确的辨识结果。通过自制的电子鼻系统采集大闸蟹活体算法的气味信息,采用KPCA算法获取样本的获取大闸蟹样本的二维特征信息,再利用自组织特征映射网络(SOM)实现对大闸蟹新鲜度的评价,并与理化指标挥发性盐基氮进行比较。试验结果表明,基于SOM网络的大闸蟹新鲜度判别的准确度可达到92%,且电子鼻各传感器的变化规律与理化指标判断结果基本一致,因此采用电子鼻技术的大闸蟹新鲜度无损检测是可行的。Freshness of Chinese mitten crab needs to establish an effective and special quality evaluation method, however, the complicated and variable physiological parameters of live crabs make the quality difficult to identify. An electronic nose is designed to sample the complex odors of live crabs using a sensor array which was consists of 7 commercial tin oxide gas sensors made in Japan （Figaro Engineering Inc.）. To get a better feature vector for identifying different crabs, KPCA is employed as a selection index for different features, then Self-organizing Feature Maps （SOM） network is used for modeling quality changes of crabs during storage. At the same time, the crab＇s TVB-N is measured as the evaluation standard. The results show that a high degree of accuracy is achieved with electronic nose based on KPCA-SOM algorithm. Compared the change rule of output of each sensor of electronic nose with the detected results of volatile components, the two test results are better consistent, so electronic nose can be used to identify the freshness of the live crabs.

关 键 词：大闸蟹 新鲜度 电子鼻 自组织特征映射网络 

分 类 号：TS254.7[轻工技术与工程—水产品加工及贮藏工程]