摘要
通过CCD监测白酒和掺入50%水的白酒两种不同液体液滴的生长过程,针对最大的液滴提取特征参数,利用主成分分析技术对特征参数进行优化,确定了5项有效的特征参数。最后采用BP神经网络识别不同的液滴。结果表明:不同液体液滴的轮廓特征参数是有区别的。因此,结合模式识别的图像液滴分析技术可以应用于分析和鉴别液体。
Monitor the growth process of two different liquid droplets that are wine and wine adding 50% of water through CCD, get out characteristic parameters of the biggest droplets, use the principal component analysis （PCA） to optimize the characteristic parameter, and determine 5 effective characteristic parameters. Finally the liquid droplets are recognized by using the BP nerve network. The result indicated： the characteristic parameters of liquid droplet are different. Therefore, image droplets parsing technique are used in the analysis and the distinction liquids.
出处
《光学仪器》
2009年第2期38-41,共4页
Optical Instruments
关键词
液滴分析
主成分分析
BP神经网络
droplets analysis
principal component analysis
BP neural network
