That's when to use the RAW model, and you have to use supported RAW files. Fortunately, the answer is straightforward: you should use the RAW model at the very beginning of your post-processing workflow. I spent a lot of time reviewing all of the comments left on our DeNoise AI v3.3 update post and our social media channels, and one of the most common themes-and source of confusion-has to do with when to use the RAW model. I recommend reading this article to learn more about how and why we built the RAW model. Both of these file types are classified as compressed RGB files, and while DeNoise AI can do great things with them, there is so much more that we can do when we source the original RAW data. This workflow requires your DAM first to create a duplicate file, often a JPEG or TIFF. We've also loved seeing the noise-reduced photos you've shared with us online! We built the RAW model to deliver the cleanest possible noise reduction by changing the data we analyze.Typically, you'd import your RAW files into your digital asset manager (DAM), go through a curation process, edit your picks, and then send the noisier ones to DeNoise AI. Then I open it back in pixinsight.We've been thrilled with all the positive (and constructive) feedback on the new RAW model recently introduced in DeNoise AI v3.3. I pull that into Topaz and work the image until I think it's removed enough noise/enhanced detail. ![]() I carefully clone over any spots left by Starnet, this is usually just the largest stars then save the file keeping it 16-bit tiff. I save the starless image as a 16-bit TIF file and edit in Photoshop. stars for the star image, and starless for the starless image. I then right click on the gray tab of the image and select Identify, and give it a name. One that is only stars, and one that is only background with nebulosity. On the duplicate image I run StarNet to remove the stars and make sure it has the star mask button is checked. I make a duplicate image here because the StarNet process is destructive, and I sometimes want to go back to the original image before StarNet. I then duplicate the image (whether it's a single channel like HA, or the RGB image). This gets the stars to the perfect value (if you've not overexposed, this makes your brightest stars become white right at the center point. You have to use a small preview window over an area that you know is pure black. After removing any light pollution and cropping out the edges from dithering, I typically do a masked stretch with the default values. My basic processing routine and some examples showing how to use Topaz AI But we can be sure to see more of this stuff in the future. They might not even explicitly state the features are AI based. And pretty soon other software will follow. Photoshop already incorporates AI like features with it’s content aware fill functions. There’s currently an AI web based tool called Star Fixer that can correct your stars for tilt or coma. I expect we’ll see a lot more AI based features added to software processing tools over the coming months to year. It’s here to stay whether we like it or not A good image processor will know and understand how to use masks to control what’s happening to their images, and will apply the effects of the tool in a way that doesn’t lead to misinterpretation of the image data that was captured. I think I fall somewhere in the middle currently along with others that believe the power is in the users hands to wield the software and tools at our disposal in such a way as to not ruin the integrity of the captured image, but to limit the tools use for what it was intended. Others that are new to the hobby, argue that it’s fun to use and effective at getting them good results with little effort. ![]() ![]() ![]() Traditional image processors are very vocal at not using any software that adds new data that didn’t exist in the image you took, and this pertains to Denoise as well as sharpening. The debate is whether or not software like this should be used at all, and if so, should it be a requirement to disclose you used it? Right now the results of this discussion are a mixed bag. When this sharpening is applied, it can have strange adverse effects if overused, or used improperly. Topaz Denoise AI is chiefly at the front of the discussion because it’s extremely good at removing noise from the image, however some of the Denoise settings come with an added feature of AI sharpening. There’s a massive debate happening right now across astrophotography Internet forums over whether or not AI based tools should be used in processing your astro images. The debate over AI (artificial intelligence) based tools in astrophotography
0 Comments
Leave a Reply. |