Are you a Lightroom user? Something big has just happened in their latest release 12.3

Are you a Lightroom user?

Something big has just happened in their latest release 12.3

A few days ago I was able to take my camera out at night and shoot a handheld photo at a higher shutter speed then usual, resulting in a less grainy image after processing. Yes, I did have image stabilisation switched on my lens, but the real progress came by being able to increase my ISO to 3200, something I don’t like to do normally.  In fact I rarely go above ISO800 as a rule of thumb, the reason being because of the grainy image I will get. True, many cameras these days have advanced sensors that are capable of increasing ISO to infinite numbers, but with my camera daring to increase ISO usually ends with a rough image. The combination of a crop sensor densely packed pixels aids this result. I am sure many photographers have their own personal cut off point for ISO.

As photographers who shoot in manual will know the trick is to keep ISO as low as possible to retain the best image quality, however there are times when you simply need to let in that artificial light. After all no light = no image!  And ISO is there for a reason within the exposure triangle.

In the past when my image was a little grainy, I’d jump into Lightroom and do is simply Noise Reduction under the Detail tab. That would be okay-ish, but I found as you applied this slider the image detail became less sharper. Sharpening the image further then tended to have the effect of applying clarity. Enter Adobe’s version of Topaz, Denoise. I was excited to have a go with this new feature and it didn’t disappoint.

So what actually happens when you send your image to Denoise?

A new DNG file is created from scratch (this might go some way to explain why the process itself can take up to 12 mins to complete). RAW files are pretty heavy.

The purpose of noise reduction is to decrease the ‘noise’ in natural images while minimizing the loss of original features and improving the signal-to-noise ratio (SNR). DeNoise AI is an intelligent noise reduction software that uses deep learning technology to accurately distinguish between real image detail and noise. According to Adobe: “Denoise uses machine learning to interpolate those patterns and remove noise at the same time. That is, the models are designed and trained to perform both demosaicing and denoising in a single step. Micha?l Gharbi and Bo Sun developed the core technology behind Denoise. As with our previous Enhance features, the idea is to train a computer using a large set of example photos. Specifically, they used millions of pairs of high-noise and low-noise image patches so that the computer can figure out how to get from one to the other. “

The structure is called a “deep convolutional neural network,” a fancy way of saying that what happens to a pixel depends on the pixels immediately around it. In other words, to understand how to up sample a given pixel, the computer needs some context, which it gets by analysing the surrounding pixels. It’s much like how, as humans, seeing how a word is used in a sentence helps us to understand the meaning of that word. Anyone who uses Photoshop regularly will know that Adobe is very good at using algorithms to detect pixel vicinity and map out pixel patterns.

Is it ethical to manipulate images this way?

That’s for you to decide. If you are a digital purist I would imagine you would be strongly against it and any other ‘enhancements’ applied to image after it hits your sensor. However, there is also an argument that slight enhancements can do just that and in any case photo modifications have been occurring since image digitalisation began. The move to ‘improve’ and culture images these days is seismic. Practically every image we see has been treated in some form or another.

Although Topaz? has been doing this for a while, with their own product it is nice to see that Adobe have integrated this into Lightroom?. Personally, I intend to use this feature from time to time. The difference for me can be as significant as leaving the tripod at home every now and then.

Footnote: I write this on the same day that Geoffrey Hinton announces his retirement from Google. Hinton is regarded as the founder of AI and expresses his concerns over its growth and an inevitable control by “Bad Actors”. Dr Hinton's pioneering research on neural networks and deep learning has paved the way for current AI systems like ChatGPT.

Thanks for reading.

EXIF: SS1/125 Handheld F3.2 ISO3200 Canon 90D Sigma 50mm Art Lens

Author: Fred Watson 02 May 2023