Assuming that the background is just solid maroon, I wonder how/why these staticky black dots can be seen in the first place. Is it a result of the camera/lens or how the image was taken? Is it a shadow/dust?
smsunarto
@bernardmc8: If I remember correctly, this is due to the camera noise. Since this is a very low light environment, the camera might have been set to use higher ISO which are more susceptible to noise.
adityaramkumar
I wonder that if there are a lot of extra black dots and we rely on techniques like averaging to denoise, how much will the black dots affect the image? Like perhaps the background wasn't as dark as the denoised image makes it out to be?
There may be ways to correct for this software wise I can think about, like average only the non-black pixels.
micahtyong
I was curious about this as well ^. I did a brief search of some other denoising techniques that could combat this issue. In particular, the order-statistic filter establishes some ranking of the neighboring pixels before replacing the center pixel with, say, the median-ranked pixel. I wonder if we could omit a small portion of super dark or super bright pixels (e.g., consider them outliers), and what that would do to the resulting image.
Assuming that the background is just solid maroon, I wonder how/why these staticky black dots can be seen in the first place. Is it a result of the camera/lens or how the image was taken? Is it a shadow/dust?
@bernardmc8: If I remember correctly, this is due to the camera noise. Since this is a very low light environment, the camera might have been set to use higher ISO which are more susceptible to noise.
I wonder that if there are a lot of extra black dots and we rely on techniques like averaging to denoise, how much will the black dots affect the image? Like perhaps the background wasn't as dark as the denoised image makes it out to be?
There may be ways to correct for this software wise I can think about, like average only the non-black pixels.
I was curious about this as well ^. I did a brief search of some other denoising techniques that could combat this issue. In particular, the order-statistic filter establishes some ranking of the neighboring pixels before replacing the center pixel with, say, the median-ranked pixel. I wonder if we could omit a small portion of super dark or super bright pixels (e.g., consider them outliers), and what that would do to the resulting image.