1

Given a colour image whose light intensity is dim at the edges and bright in the middle, such as [dim bright dim], which computer vision technique would be recommended to correct this imbalance?

While I believe Histogram Equalisation could help, I've read that it also introduces an unrealistic look to an image by exaggerating the blue channel.

I consider implementing an image processor for this task.

FrustratedWithFormsDesigner
  • 46,105
  • 7
  • 126
  • 176
BlackBox
  • 145
  • 5

1 Answers1

2

One classic approach is using top hat filtering for leveling uneven illumination. You should perform morphological opening with a large structuring element, then remove the result from the original image.

If you have dark objects on light background you should do bottom hat filtering (dual of top hat filtering): subtract the closing with a large structuring element from the original image.

Result of top hat filtering

Robert Harvey
  • 198,589
  • 55
  • 464
  • 673
WebMonster
  • 156
  • 3
  • Thanks WebMonster. I'll take a look at this as I've never heard the name before. Does this approach work equally as well with coloured images as it does greyscale? – BlackBox Apr 30 '14 at 19:30
  • I don't know if there is a generalization of morphological operators for colored images. You could convert the image to HSV color space, process the V channel as a grayscale image, then display the image. There will be some colour bleeding, but it could be good enough depending on your goals. – WebMonster Apr 30 '14 at 20:37