基于多指标融合的猪肉新鲜度智能检测方法  

Intelligent Detection of Pork Freshness Based on Fusion of Multiple Indexes

作　　者：李学相[1] 

机构地区：[1]郑州大学软件技术学院,河南郑州450002

出　　处：《计算机仿真》2012年第9期224-227,共4页Computer Simulation

基　　金：国家高技术研究发展计划(2007AA010408)

摘　　要：研究猪肉新鲜度检测优化问题,传统人工感官评定和化学方法测量均检测不准确,无法对猪肉新鲜度进行准确评价,检测误差大。为提高猪肉新鲜度检测准确率,将多种评价指标(特征)相融合,提出一种利用智能猪肉新鲜度检测方法。首先分别采用图像和气味采集装置收集各种评价参数值,然后采用主成分分析对评价指标进行融合,最后采用支持向量机建立猪肉新鲜度检测模型。仿真结果表明,相对于传统检测算法,智能检测算法提高了猪肉新鲜度检测准确率,实现了猪肉新鲜度智能检测。The paper proposed a pork freshness detection method based on support vector machine. Firstly, the collection devices of image and the odor were used to collect all kinds of evaluation parameters, and then the principal component was analysed to the evaluation of fusion. Finally, support vector machine was used to build pork freshness detection model. The simulation results show that, compared with the traditional detection algorithm, this algorithm improves the pork freshness detection accuracy rate and realizes the pork freshness detection.

关 键 词：支持向量机 猪肉新鲜度 检测 主成分分析 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]