Investigate rays in Fourier domain for the estimated PRNU #77

New Issue

Benjamin_Loison · 2024-05-16T11:22:55+02:00

Benjamin_Loison commented

2024-05-16 11:22:55 +02:00

A supervisor told me, no matter if I do progress (at least I understand that), we should discuss this aspect.

I am looking for the name of the phenomenon involving rays in the following image:

Source: http://tontonphoto.fr/comment-photographier-rayons-soleil-halo-flare-sunburst/

superposition ondes de toutes les fréquences DuckDuckGo and Google results are not interesting.

nom phénomène rayon du soleil DuckDuckGo and Google results are not interesting.

What about in English?

https://en.wikipedia.org/wiki/Sunbeam is more accurate but still does not seem to interest itself about rays without clouds etc.

The name I had in mind was something like https://en.wikipedia.org/wiki/Hysteresis but it does not seem to match what I am actually thinking about.

Related to #76.

A supervisor told me, no matter if I do progress (at least I understand that), we should discuss this aspect. I am looking for the name of the phenomenon involving rays in the following image: ![image](/attachments/a8892e4f-4777-4f71-aed1-84ce28d3657d) Source: http://tontonphoto.fr/comment-photographier-rayons-soleil-halo-flare-sunburst/ `superposition ondes de toutes les fréquences` DuckDuckGo and Google results are not interesting. `nom phénomène rayon du soleil` DuckDuckGo and Google results are not interesting. What about in English? https://en.wikipedia.org/wiki/Sunbeam is more accurate but still does not seem to interest itself about rays without clouds etc. The name I had in mind was something like https://en.wikipedia.org/wiki/Hysteresis but it does not seem to match what I am actually thinking about. Related to #76.

image.png

20 MiB

Benjamin_Loison added the

enhancement

medium priority

quick

labels 2024-05-16 11:22:55 +02:00

Benjamin_Loison referenced this issue from a commit

2024-05-16 12:22:50 +02:00

#77: Add `articles/Lens flare - Wikipedia/`

Benjamin_Loison commented

2024-05-16 12:59:14 +02:00

Should proceed similarly for rays as for circles, see issues/76#issuecomment-1933.

Should proceed similarly for rays as for circles, see [issues/76#issuecomment-1933](https://gitea.lemnoslife.com/Benjamin_Loison/Robust_image_source_identification_on_modern_smartphones/issues/76#issuecomment-1933).

Benjamin_Loison commented

2024-05-31 16:33:53 +02:00

Automatically translated with LibreTranslate from Google Doc Minutes:

How to generate the FFTs with rays that were previously observed? Once recovered, analyze them by looking at what they give in the real domain. Also analyze the phase. Correcting the radial profile and having a circular profile to count the rays can make sense.

In reality, my calculation method was not as accurate as it is now, since in the case of re-adjustments I used a visualization of the UNPR estimate and not the raw Numpy tables. However, the rays appeared not only on the re-frames but trying to find a seemingly strange result does not seem appropriate.

Automatically translated with [LibreTranslate](https://github.com/LibreTranslate/LibreTranslate) from Google Doc Minutes: > How to generate the FFTs with rays that were previously observed? Once recovered, analyze them by looking at what they give in the real domain. Also analyze the phase. Correcting the radial profile and having a circular profile to count the rays can make sense. > > In reality, my calculation method was not as accurate as it is now, since in the case of re-adjustments I used a visualization of the UNPR estimate and not the raw Numpy tables. However, the rays appeared not only on the re-frames but trying to find a seemingly strange result does not seem appropriate.

Benjamin_Loison closed this issue

2024-05-31 16:33:53 +02:00

Sign in to join this conversation.