GraXpert's AI background extraction, run from DSPX
GraXpert Background Extraction removes the gradient from an image with the AI model of GraXpert, a separate program that we run for us. We choose subtraction or division and the smoothing, DSPX sends the image to GraXpert and brings the result back, and GraXpert can give us the background it found as a new image. GraXpert is not part of DSPX: we install it ourselves.
GraXpert is free software for astrophotography, made by its own authors and available from its own website. Its background extraction finds the gradient of an image with a neural network, with no samples to place. This process is the way to use it without leaving DSPX: it takes the image of the panel, hands it to the GraXpert installed on this computer, waits for it, and replaces the image with what GraXpert made. The result goes through the image's mask and selection and makes one history step, like the result of any other process.
DSPX ships nothing of GraXpert and downloads nothing of it. GraXpert itself is free software under the GPL-3.0, and the AI models it uses have their own license, Creative Commons BY-NC-SA 4.0, which is for non-commercial use. They are separate software, and their licenses are theirs. We find GraXpert in , where we say which program to run. Our own background extraction, with samples we place and models we choose, is Background Extraction.
The panel has two sections, Background and Model, then the image the process applies to, the buttons and the system bar along the bottom. There is no live preview: GraXpert takes seconds to minutes, so the process runs when we click Apply.
Correction: How the background GraXpert found is removed:
Smoothing: How smooth the background model is, from 0 to 1. 0 by default.
Create background model: Also makes the background GraXpert found as a new image, named after the image with the suffix background. Off by default.
Model version: The version of GraXpert's AI model, written as three numbers such as 1.0.0. Empty, the default, uses the latest one GraXpert has. A model that GraXpert does not have on this computer yet is downloaded by GraXpert itself.
Apply writes the image to a temporary file, runs GraXpert on it, reads the result and replaces the image with it, in one history step that Undo takes back, through the image's mask if it has one. The panel shows the progress GraXpert reports, and the Cancel button stops GraXpert and everything it started. The image keeps its size, its channels and its sample type; integer images get the result clipped to their range, so nothing wraps around. Alpha channels and the image's metadata, astrometric solution included, are not touched.
We want to be exact about what DSPX does on this computer.
-cli -cmd background-extraction,
the file, -output result, then -correction, -smoothing, -bg when we
ask for the background model, -ai_version when we name a model and -gpu when
GPU in Preferences is On or Off. The first line DSPX writes to the Log is this command.Look at the background model. Turn on Create background model once: a model with the structure of the object in it means the gradient is not what GraXpert took it for. GraXpert's own documentation says more about the stage of processing it suits.
Several attempts. Smoothing and the model version change the result. Undo and try another setting; the History panel keeps each step.
Other images. Once the settings are right, we save the process () to apply it to other images from the Process Shelf. GraXpert runs once for each image.
Background Extraction: our own background extraction, with samples we place and polynomial or spline models.
GraXpert Denoise: GraXpert's noise reduction.
GraXpert Deconvolution: GraXpert's deconvolution, in the releases that have it.