Hacker News Reader

Comment by danbruc | Hacker News Reader

You can remove blur from images with [blind] deconvolution. [1]

[1] https://en.wikipedia.org/wiki/Blind_deconvolution

Blind deconvolution is useful in astronomy and space domain awareness where you can make reasonable assumptions that the support for the signal pixels is limited against a blank background. This isn't the case in everyday terrestrial photography, and the Lena image on that wiki shows the sorts of results you get from applying a blind deconvolution algorithm naively to this sort of imagery.

> the Lena image on that wiki shows the sorts of results you get from applying a blind deconvolution algorithm naively to this sort of imagery.

the edges seems to be restored fine on that image - this is exactly what you'd need for fingerprints.

Also, the image have clear partial palm prints which police may have had in their db on this guy too.

It wouldn’t even be blind deconvolution since they can use the cheese label as ground truth.

I wouldn't think so, since the level of blur is less or non-existent on the label. In infinitely long focal length, everything is the same level of focus.

Maybe it’s just me, but I tend to see a lot of motion blur on that label, an amount that could easily account for the blur on the fingertip, too. Given the ultra-short focal length of mobile cameras and the large focal depth that comes with it I would expect very little out-of-plane blur on the fingertips.

The label is very blurry.

Plus, the point is that the picture is not a blurred picture of a finger. There is a lot of information in this photo which could be used to help to guide the de-blurring process.

Reply on news.ycombinator.com