Research on the method of feature extraction based on ARM platform

DOI

10.2991/icadme-15.2015.281How to use a DOI?

Keywords

Colloidal gold reagent; Quantitative detection; Feature extraction; Least square method; Weight function;

Abstract

Detection method of colloidal gold immunochromatography assay concentration using colorimetric principle, when tested in liquid reagent on antigen and antibody response, reagent detection area change color, and compared to standard color grade then the concentration can be qualitative analysis. At present, the concentration of colloidal gold test strip is mostly semi quantitative, so the application scope is limited. This paper designs an image shooting based on embedded ARM platform application which processor using ARM v7 architecture, equipped with WINCE system, the OV9650 CMOS (Complementary Metal Oxide Semi-conductor) image sensor to obtain quantitative colloidal gold test strip image information, while setting manual white balance, exposure time, manual gain, then extracting ruler window setting image cut. And we can judge and recognize the characteristics of the detection regions, through using the least square method to further reduce the error, and for the effective feature region of the most value and area value, design weight function, to improve the repeatability of measurement results. The paper makes a quantitative analysis of the image of the reagent, and the repeatability error can reach ±1%. Through the analysis of the research the results are in accord with the design requirements. The design and algorithm of this paper have some practical value.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Dezhi Zheng
AU  - Kunming Wang
AU  - Jialin Cheng
AU  - Shangchun Fan
PY  - 2015/10
DA  - 2015/10
TI  - Research on the method of feature extraction based on ARM platform
BT  - Proceedings of the 5th International Conference on Advanced Design and Manufacturing Engineering
PB  - Atlantis Press
SP  - 1515
EP  - 1521
SN  - 2352-5401
UR  - https://doi.org/10.2991/icadme-15.2015.281
DO  - 10.2991/icadme-15.2015.281
ID  - Zheng2015/10
ER  -