Fluorescent signal detection of chromatographic chip by algorithms of pyramid connection and Gaussian mixture model.
- Author:
Beibei HU
1
;
Xueqing ZHANG
;
Haopeng CHEN
;
Daxiang CUI
Author Information
1. School of Software, Shanghai JiaoTong University, Shanghai 200240. jliubme@tsinghua.edu.cn
- Publication Type:Journal Article
- MeSH:
Algorithms;
Chromatography;
Fluorescent Antibody Technique;
methods;
Image Processing, Computer-Assisted
- From:
Chinese Journal of Medical Instrumentation
2011;35(2):83-86
- CountryChina
- Language:Chinese
-
Abstract:
We proposed a new algorithm for automatic identification of fluorescent signal. Based on the features of chromatographic chips, mathematic morphology in RGB color space was used to filter and enhance the images, pyramid connection was used to segment the areas of fluorescent signal, and then the method of Gaussian Mixture Model was used to detect the fluorescent signal. Finally we calculated the average fluorescent intensity in obtained fluorescent areas. Our results show that the algorithm has a good efficacy to segment the fluorescent areas, can detect the fluorescent signal quickly and accurately, and finally realize the quantitative detection of fluorescent signal in chromatographic chip.
