Can Camera-Based Image Processing Accurately Measure Furnace Flame Intensity?

Click For Summary
SUMMARY

The discussion centers on measuring the intensity of a furnace flame using camera-based image processing techniques. Participants suggest utilizing C# or MATLAB to analyze the RGB array of captured images for intensity calculations. Recommendations include employing a reflecting filter to minimize light interference and heat absorption. Additionally, insights are shared about traditional flame detection methods using UV sensors, highlighting advancements in technology over the past forty years.

PREREQUISITES
  • Understanding of camera image processing techniques
  • Familiarity with C# or MATLAB programming
  • Knowledge of optical filters and their applications
  • Basic principles of flame detection technology
NEXT STEPS
  • Research the use of reflecting filters in optical measurements
  • Explore advanced image processing algorithms in MATLAB
  • Investigate modern flame detection technologies and their applications
  • Learn about the calibration of CCD sensors for accurate light measurement
USEFUL FOR

Engineers, researchers, and hobbyists interested in optical measurement techniques, particularly in industrial applications such as steel refining and flame detection.

Nicole04
Messages
2
Reaction score
0
Hi Guys,

A basic oxygen furnace is very large 400 ton vessel that steel is refined in. during the refining process oxygen is blown into the steel. during this process there is a flame at the mouth of the furnace, this is what I am interesting in. during the process the flame changes in size and intensity. I want to be measure the intensity of the flame by using a camera to capture an image then use some c#/matlab code to calculate the intensity based on the RGB array of the image. will this work to give an indication of the relative intensity of the flames ? or have a missed the boat? can you clever people please help?
 
Last edited by a moderator:
Engineering news on Phys.org
You probably want to use a fairly thick filter. CCDs are fairly linear, and in a camera, the lens is usually sized so that other than aiming directly at the sun, the data can be handled. Technically, the light you are trying to measure is not quite as bright as the sun's surface. However the atmosphere knocks the solar intensity down a significant amount.

The best filter for your application is going to reduce the light by several orders of magnitude. You also need to consider the heat absorbed by the filter all up and down the spectrum. This is why I recommend a reflecting filter. The "extra" light from the mirror surface won't contaminate your reading all that much. (And if you care more for cheap than precision you can take almost any pocket mirror and remove the black paint on the back. ;-)
 
this is what I am interesting in. during the process the flame changes in size and intensity.

Interesting project.

the flame detectors in our oil/gas fired boilers used a UV sensor and measured the AC component or 'flicker' of the resulting signal.

But that was forty years ago. Surely they've become more sophisticated.

Same outfit is still in business. This instruction manual from them gives an intro to optical flame detection, see around page 14-15. Probably you're aware of them, though...
http://66.18.210.80/manuals/Fireye/SERVICE MANUAL.pdf
 
@ eachus - Thank you :) , Thats just the sort of thing I had no knowledge of. I am definitely going to give it a try, will let you know how it goes.

@jim- Thanks Jim, that would work, however cost is a factor, so I'm trying to get a cheap 'home-made' solution ;-)
 
Last edited: