Enfuse: focus stack blurring

[Mike Smith]

Hi

I shot two images to focus stack them in Enfuse using the command below which aligns the images before applying a 1.0 weight for contrast using a hard mask.

enfuse.exe" -o DSC_0091_HDR2.jpg --no-ciecam --exposure-weight=0 --saturation-weight=0 --contrast-weight=1 --contrast-window-size=5 --depth=16 --compression=95 --hard-mask aligned_0000.tif aligned_0001.tif

The two images were taken using a 24mm lens on a full frame camera (f8) inside a museum studio and focused at ~0.8m and 2.0m. This gives a DoF that overlaps providing focused areas across the full image. So far the above command seems to give the best results but I get a faint "halo" around the far distance objects in the room which is a result of the out-of-focus areas from the 0.8m focus image (see attached).

Is there a way to get a better stack output or should I have shot more images across the focal range?

thanks

mike

[cspiel]

Mike -

    Your problem is known.  Actually, it is the very issue why I have
entered Enfuse development.

Please consult the Enfuse manual's chapter "Focus Stacks" and
in particular section "Advanced Focus Stacking".  Although there
is no guarantee that you'll be able to shoo away the halos in your
particular stack of images, the Laplacian-of-Gaussian algorithm
can improve the quality of the final image with respect to halo
formation.

HTH,
    Chris

[Mike Smith]

Hi Chris

Thanks very much for this - helped nudge me in the right direction as I had been looking through the Enfuse chapter. I did try some of the LoG stuff and experimented with settings around this

--contrast-edge-scale=31:80%:80%

and

--contrast-edge-scale=31 --contrast-min-curvature=11

What I had found in my case was that increasing the kernel size of the contrast window to 31 significantly improved the stack, along with applying the hardmask. What subsequently made the biggest impact was changing the gray projector to luminance. The relatively large "halo" from the two images need both a large kernel and then adjustment based on luminance to remove it. I was very impressed with the end result

Thanks again

mike