Research on ESNI algorithm for image recognition of boiler water level gauge

Author

Liu Ziyu ; Zhang Changsheng ; Sun Bin ; Qian Bin ; Tian Haiyong ; Zhang Hanping

Author_Institution

Fac. of Inf. Eng. & Autom., Kunming Univ. of Sci. & Technol., Kunming, China

fYear

2014

fDate

May 31 2014-June 2 2014

Firstpage

4106

Lastpage

4111

Abstract

The value of boiler drum level is usually obtained by operator visual reading to bicolor water level gauge, but it brings the problems such as inefficiency and bigger subjective error. In order to get value corresponding to the meter in the image, the processing algorithms based on Chow-Kaneko threshold processing and ISEF (Infinite Symmetric Exponential Filter) for image of the gauge are studied, and the matrix image is calculated by binarization, segmentation of liquid level area and precisely positioning of pixels, in particular, the recognition algorithm ESNI (Eight-subblocks Number Identification) based on Bayesian classifier is proposed. Experiment shows that the above methods and the system implementation are of higher precision, speed and better robustness.

Keywords

Bayes methods; boilers; filtering theory; image recognition; image segmentation; pattern classification; production engineering computing; Bayesian classifier; Chow-Kaneko threshold processing; ESNI algorithm; ISEF; boiler drum level value; boiler water level gauge; eight-subblocks number identification; image recognition; industrial boiler; infinite symmetric exponential filter; liquid level area binarization; liquid level area segmentation; matrix image; Algorithm design and analysis; Classification algorithms; Image edge detection; Image segmentation; Laplace equations; Liquids; Bicolor level gauge; Chow - Kaneko threshold algorithm; ESNI; ISEF;

fLanguage

English

Publisher

ieee

Conference_Titel

Control and Decision Conference (2014 CCDC), The 26th Chinese

Conference_Location

Changsha

Print_ISBN

978-1-4799-3707-3

Type

conf

DOI

10.1109/CCDC.2014.6852900

Filename

6852900