Night Photography Revisited: De-Noising vs. Losing Detail

Results of shooting photos at night using an iPhone XS Max and Topaz Labs' "Denoise AI" tool and color/lighting filters to compensate for the noise.

Night Photography Revisited: De-Noising vs. Losing Detail

First, I'll let you check out the photos and then, at the end of this post, I'll share my observations.


To begin, here's a few I took around the above Buddhist temple, in Koshigaya City, Saitama, Japan.

And, an artsy version of the same, care of BeFunky's "Oil Painting DLX" filter.


Next, here are some of aging buildings.


Below is the main residential tower near Koshigaya Station, as seen from various angles.

And, the last one above is a nice segue into a couple of this temple.

And, an artsy version of the same, care of BeFunky's "Underpainting DLX" filter.

Lessons Learned

Shooting

The photos above were taken on my iPhone XS Max.

  1. Noise: To avoid noise, it's best to shoot without the flash, otherwise, that will light up any dust in the air and create unwanted noise. Furthermore, the flash is too weak to light up anything far away.
  2. Limitations: This phone doesn't work so well for night photography. Apparently, the Google Pixel 3 is much better. However, we can expect future software updates, and the next flagship iPhone to be better.
  3. Solutions: Clip-on lenses might capture more light. Shooting via a camera app that uses RAW format might also capture more information that can be brought out in post production. For the above shots, I did neither.

Post Production

  1. Decluttering: Once I choose a photo I'd like to keep, I remove items that create clutter or distraction using TouchRetouch. However, since the images are so rough, it's harder to work with lines.
  2. De-noising: All the above photos turned out noisy. I was able to resolve this to a degree using Topaz Labs' Denoise AI tool. However, I am not satisfied with the results. It could be that if I played around with the parameters such as De-noising, Sharpening, and Adding Detail, that improvements could be achieved. Furthermore, perhaps using their Sharpen AI tool might bring out more detail. However, these tools run slow on my aging MacBook Pro, so I didn't have the patience to try and this factor makes it hard to compare different settings.
  3. Optimizing Colors: Photolemur, which I use on almost all of my daytime photos with significant benefits was not useful for these night shots, so I didn't use it.
  4. Filters: Photos for macOS has a few basic filters. Sometimes these work great, but their intensity can't be adjusted, so I sometimes found it was better to hunt through the filter library of Polarr to find one. If you get the right filter to fit the lighting, it can make a huge difference.
  5. Effects: One last resort if a photo turns out poorly, but you are invested in making use of it somehow is to try running some artistic effect filters on it. This might result in something interesting. BeFunky and Prisma are my go-to apps for this.

Conclusion

I need a lot more practice and should start watching some tutorial videos!