中山大学学报自然科学版 ›› 2010, Vol. 49 ›› Issue (2): 28-30.

• 研究论文 • 上一篇    下一篇

基于电子鼻的鱼类新鲜度估计研究

刘红秀1,李洪波1,李卫东2,骆德汉3   

  1. (1.广东药学院医药信息工程学院,广东 广州 510006;2.广东药学院基础学院,广东 广州 510006;3.广东工业大学信息工程学院, 广东 广州510006)
  • 收稿日期:2009-04-09 修回日期:1900-01-01 出版日期:2010-03-25 发布日期:2010-03-25

Research on the Fish Freshness Assessment Based on Electronic Nose

LIU Hongxiu1, LI Hongbo1,LI Weidong2, LUO Dehan 3   

  1. (1. College of Medical Information Engineering, Guangdong Pharmaceutical University, Guangzhou 510006, China;2. School of Basic Courses, Guangdong Pharmaceutical University, Guangzhou 510006, China;3. Department of Information Engineering, Guangdong University of Technology, Guangzhou 510006, China )
  • Received:2009-04-09 Revised:1900-01-01 Online:2010-03-25 Published:2010-03-25

摘要: 以新西兰市场上最受欢迎的四类鱼(红甲鱼、鲂鱼、唇指鲈和(澳洲)鲹)为对象研究鱼的新鲜度。在同一实验室环境下,运用便携式电子鼻Cyranose 320测量这四类鱼被储藏第1,2,5,6,7,8,9,10(第3,4天的未测量)天后对应的同一样品,每个样品测量一次对应每个传感器平均采样2000个左右数据,获得大约2048×106[4(鱼)×8(天)×32(传感器)×2000(采样)=2048000]个数据。将实验数据进行特征提取及人工神经网络(ANN)分析处理,得到传感器对每类鱼每天的响应模式,进而估计鱼的新鲜度,获得了91%以上的正确识别率。研究结果表明该方法是实用可行的。

关键词: 电子鼻, 信息处理, 神经网络, 鱼的新鲜度估计

Abstract: The freshness on four selected types of fish (Red Snapper, Gurnard, Tarakihi and Trevally) which are the most common fish in the New Zealand market was investigated. A portable Cyranose 320 Enose was used in our experiments under the same laboratory condition. It converted the odour of four selected types of fish to smell prints over days 1, 2, 5, 6, 7, 8, 9, and 10 after catching the fish (no data was collected on days 3 and 4). Approximately 2 000 samples were collected by each sensor during each process. About 2 048 000 data samples [4 (fish) × 8 (days) ×32 (sensors) ×2 000 (samples) =2 048 000] were obtained. Extracted features from the Enose sensors and artificial neural network (ANN)were used to assess the freshness of the fish by classifying the smell print data according to the day of data collection. The proposed system has been successful in identifying the number of days after catching the fish with an accuracy of up to 91%. The result showed that the proposed network architecture proved very suitable for fish freshness assessment.

Key words: electronic nose, information processing, artificial neural network, fish freshness assessment

中图分类号: