Thread: what's causing this problem?

I just looked more closely at an image I posted on another forum, and I see defects I don't understand. Perhaps someone has a suggestion.

The image was stacked, but the flaws appear in individual images. Here is a screen shot of a portion of the image at 200%, with two of the problematic areas circled:



Here's the image from which this was clipped, with no editing other than a little noise reduction:



There are two problems. First, there is obvious banding. More puzzling is the material in the red circle on the left, which looks like a faint image of the brickwork to the left.

This was shot with a Canon R6 Mark II and an RF 24-105 L at 72mm, ISO 640, f/11, 1/20. That body and lens combination has a total of 8 stops of image stabilization, so camera motion isn't a likely culprit. Moreover, enough motion to cause that would have blurred everything. ISO 640 is not high for that body. Exposure was pretty good, with the histogram extending the full range and the bulk of the pixels near the middle. Lighting was natural light from windows on the left, as you can see from the second image. Given the lighting and my position, I don't think there was a possibility of flair.

I haven't noticed anything like this in other images I've captured with this body, but I will look more closely now.

Any thoughts?

Thanks

I wonder if what you are seeing are the actual bricks that are part of the left hand wall There are quite a number of bricks visible in the image above the fire door and suspect the things that look like bricks are actually bricks.

I can't tell for sure, but the wall to the left appears to have some form of parging covering it, the right angle face has not been treated this way. I suspect that the fire door is a relatively recent addition to this old building, so we see the old materials that have been cut away to install a modern door and the modern steel frame that is part of the fire-resistant door.

The other streaks look like noise / artifacts, but the repetitive patterns are a bit strange looking.

I have seen similar artifacts in images that was traced back to file corruption with a failing memory card. As you stacked using multiple images displaying the same problem I assume corruption is unlikely to be the cause.

There was another image (I think on this site?) that also had strange noise in it and the cause was possibly RFI interference from close by military radar/radio communications facilities.

Thanks for the ideas.

Yes, there is parging on the brickwork. I don't recall if the short perpendicular surface where the door jamb is placed is also parged. However, the streaks are on the metal surfaces of the door jamb and door, which are a smooth matte gray. The lines appear to be a continuation of the mortar lines.

There was nothing covering the door or door jamb.

I first thought it was a stacking error, but I ruled that out by looking at the individual images that went into the stack. It shows up clearly in 2 of the 3 individual images.

Another puzzle: this morning I compared the three images that went into the stack, and this is least apparent in the first image, where the door is most out of focus. That makes sense. However, comparing the last two, it actually seems slightly more apparent in the image in which the door is now quite in focus.

The camera has two SD cards, one new and one old. Unfortunately, I can't easily find out which was used for this image. However, they are dirt cheap, and I'll go ahead and order a replacement for the older one, just in case.

I'm not aware of any potential sources of powerful RF interference in that area. The museum is located in a complex of previously abandoned 19th-century mill buildings, and I'm not aware of contemporary manufacturing on any large scale in the area or any military facilities.

The noise in the rest of the image is also troubling; this is a sensor that can readily handle considerably higher ISOs if images are properly exposed.

I haven't notice these problems in other images taken with this camera. I just checked another taken in the same complex at roughly the same time, and it's clean, but I will have to check more.