0

I have the following cell reconstructed images. 12 images of the same cell

I would like to implement a automatic detection of the most sharpest image, which in our case is the 4th , 5th or 6th image. If we pick one of them is equally correct.

What it would be a good approach to detect the sharpest image? Is gradient methods helpful?

Elena Stamatelou
  • 81
  • 3
  • 13

1 Answers1

1

In other words, you are looking for a blur metric.

Kanjar De, V. Masilamani, Image Sharpness Measure for Blurred Images in Frequency Domain, Procedia Engineering, Volume 64, 2013, Pages 149-158

propose a simple approach based on Fourrier Transform, author citation:

when the blur in an image increases the number of high frequency component in the images decreases

Another approach is to say the more my image is sharp, the more edges I have, as in

P. Marziliano, F. Dufaux, S. Winkler and T. Ebrahimi, A no-reference perceptual blur metric, Proceedings. International Conference on Image Processing, 2002, pp. III-57-III-60 vol.3.

Vincent Vidal
  • 323
  • 2
  • 11