摘要
液体的物理化学性质直接影响到其液滴生长,提出基于CCD的图像液滴分析方法。该方法通过CCD摄像机实时监测液滴生长过程中的轮廓变化情况,经图像处理系统对液滴图像进行预处理、分割和特征提取。提取了面积、周长、液滴长度、矩形度和圆形度等液滴轮廓特征参数,并用主成分分析(PCA)法对这5项指标进行了分析。结合实验,说明了该方法用于液体分析和鉴别的可行性。
A new method to analyse liquid drop images based on COD is introduced in this paper,which overcomes the limitations of fiber and the capacitive drop analysis method, and enhances the discrimination ability of liquid drop analysis technology. The instant liquid drop profile during its growth is monitored by COD camera real-timely,and the collected images are processed and the feature extraction is realized after image processing. The region area, boundary perimeter, drop length, satiety, circularity, and the profile edge of the liquid drop image are extracted and analyzed with principal component analysis（POA）. The feasibility of image liquid drop analysis technology is proved with experiments.
出处
《光电子．激光》
EI
CAS
CSCD
北大核心
2006年第8期1004-1008,共5页
Journal of Optoelectronics·Laser
基金
科技部"中国新加坡联合研究计划(NSTB-MOSTJointResearchProgram)"资助项目(003/101/04)
关键词
液滴分析
CCD
图像处理
特征提取
主成分分析(PCA)
liquid drop analysis
CCD
image processing
feature extraction
principal componet analysis （PCA）
