Teaching a computer to read

Image analysis of electrical meters

Terrance Lovell, Dale Henry Litwhiler

Research output: Contribution to journalArticle

Abstract

There exists a vast infrastructure of heritage analog and digital meters installed in commercial and industrial applications. These devices typically have no built-in means of automated reading. Modifying the equipment is not a viable option in many applications. With the low cost of USB digital cameras and the availability of Lab VIEW™ VISION, a cost-effective method of reading multiple meters of assorted types can be created. Duplicating the process that a human performs while reading a meter display is daunting. However, this process is simplified by using virtual instruments (VIs), which perform essential functions such as edge, pattern and rotation detection. As part of an undergraduate research project, a computer, using Lab VIEW™ Vision, together with a USB digital camera is used to read a digital multimeter (DMM) and an analog watt-hour meter. Circular edge detection, pattern searches, and rotation detection are used to locate dials and segments and to determine their values. Horizontal and vertical edge detection and region of interests (ROI) are used to identify and determine the values of a DMM's display. The ability to read meters with only minimal human interaction increases accuracy and speed. This feature and the ability to create visual data logging have many uses in educational and industrial applications. This paper presents techniques for identifying and reading meter data. The basics of reference images and their use in image analysis are explored in reading legacy DMM and analog watt-hour meters.

Original languageEnglish (US)
JournalASEE Annual Conference and Exposition, Conference Proceedings
StatePublished - 2006

Fingerprint

Image analysis
Teaching
Watt hour meters
Digital cameras
Edge detection
Industrial applications
Display devices
Costs
Availability

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Cite this

@article{fae0ac22d6204d57a69215237e4543b5,
title = "Teaching a computer to read: Image analysis of electrical meters",
abstract = "There exists a vast infrastructure of heritage analog and digital meters installed in commercial and industrial applications. These devices typically have no built-in means of automated reading. Modifying the equipment is not a viable option in many applications. With the low cost of USB digital cameras and the availability of Lab VIEW™ VISION, a cost-effective method of reading multiple meters of assorted types can be created. Duplicating the process that a human performs while reading a meter display is daunting. However, this process is simplified by using virtual instruments (VIs), which perform essential functions such as edge, pattern and rotation detection. As part of an undergraduate research project, a computer, using Lab VIEW™ Vision, together with a USB digital camera is used to read a digital multimeter (DMM) and an analog watt-hour meter. Circular edge detection, pattern searches, and rotation detection are used to locate dials and segments and to determine their values. Horizontal and vertical edge detection and region of interests (ROI) are used to identify and determine the values of a DMM's display. The ability to read meters with only minimal human interaction increases accuracy and speed. This feature and the ability to create visual data logging have many uses in educational and industrial applications. This paper presents techniques for identifying and reading meter data. The basics of reference images and their use in image analysis are explored in reading legacy DMM and analog watt-hour meters.",
author = "Terrance Lovell and Litwhiler, {Dale Henry}",
year = "2006",
language = "English (US)",
journal = "ASEE Annual Conference and Exposition, Conference Proceedings",
issn = "2153-5965",

}

TY - JOUR

T1 - Teaching a computer to read

T2 - Image analysis of electrical meters

AU - Lovell, Terrance

AU - Litwhiler, Dale Henry

PY - 2006

Y1 - 2006

N2 - There exists a vast infrastructure of heritage analog and digital meters installed in commercial and industrial applications. These devices typically have no built-in means of automated reading. Modifying the equipment is not a viable option in many applications. With the low cost of USB digital cameras and the availability of Lab VIEW™ VISION, a cost-effective method of reading multiple meters of assorted types can be created. Duplicating the process that a human performs while reading a meter display is daunting. However, this process is simplified by using virtual instruments (VIs), which perform essential functions such as edge, pattern and rotation detection. As part of an undergraduate research project, a computer, using Lab VIEW™ Vision, together with a USB digital camera is used to read a digital multimeter (DMM) and an analog watt-hour meter. Circular edge detection, pattern searches, and rotation detection are used to locate dials and segments and to determine their values. Horizontal and vertical edge detection and region of interests (ROI) are used to identify and determine the values of a DMM's display. The ability to read meters with only minimal human interaction increases accuracy and speed. This feature and the ability to create visual data logging have many uses in educational and industrial applications. This paper presents techniques for identifying and reading meter data. The basics of reference images and their use in image analysis are explored in reading legacy DMM and analog watt-hour meters.

AB - There exists a vast infrastructure of heritage analog and digital meters installed in commercial and industrial applications. These devices typically have no built-in means of automated reading. Modifying the equipment is not a viable option in many applications. With the low cost of USB digital cameras and the availability of Lab VIEW™ VISION, a cost-effective method of reading multiple meters of assorted types can be created. Duplicating the process that a human performs while reading a meter display is daunting. However, this process is simplified by using virtual instruments (VIs), which perform essential functions such as edge, pattern and rotation detection. As part of an undergraduate research project, a computer, using Lab VIEW™ Vision, together with a USB digital camera is used to read a digital multimeter (DMM) and an analog watt-hour meter. Circular edge detection, pattern searches, and rotation detection are used to locate dials and segments and to determine their values. Horizontal and vertical edge detection and region of interests (ROI) are used to identify and determine the values of a DMM's display. The ability to read meters with only minimal human interaction increases accuracy and speed. This feature and the ability to create visual data logging have many uses in educational and industrial applications. This paper presents techniques for identifying and reading meter data. The basics of reference images and their use in image analysis are explored in reading legacy DMM and analog watt-hour meters.

UR - http://www.scopus.com/inward/record.url?scp=85029130528&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85029130528&partnerID=8YFLogxK

M3 - Article

JO - ASEE Annual Conference and Exposition, Conference Proceedings

JF - ASEE Annual Conference and Exposition, Conference Proceedings

SN - 2153-5965

ER -