By This Hour Science Desk
NASA’s Astronomy Picture of the Day for September 27 presents Andromeda in two states: one that retains the visible disruptions of a long ground-based exposure and another in which those disruptions have been substantially reduced. The pairing makes a narrow but consequential point about astronomical imaging. A photograph of a distant galaxy can carry marks made by Earth’s atmosphere, passing aircraft, satellites and camera hardware alongside light from the object the observer intends to show.
The image feature, titled Andromeda Before and After Photoshop, identifies the target as M31, the Andromeda galaxy, and frames the comparison around the task of separating the galaxy’s appearance from unwanted foreground and instrumental effects. The “before” view is not presented as an error or a failed observation. It is presented as the result of a long exposure through Earth’s skies using a digital camera, with the ordinary imperfections that arrangement can leave in an image.
That distinction matters because the before-and-after format invites a question more precise than whether an image has been edited: what, exactly, has been changed? NASA’s accompanying explanation says the processing was intended to suppress features that do not belong to Andromeda, rather than to manufacture a version of the galaxy detached from the recorded data. The comparison therefore puts the practical work of image cleaning in full view instead of leaving the finished astronomical portrait to stand on its own.
A 2019 observing run behind a 2026 feature
The page says the image was built from 223 individual exposures, each lasting 300 seconds, obtained in 2019 from a garden observatory in Portugal. That chronology is central to reading the image properly. The September 2026 publication date marks the Astronomy Picture of the Day feature, while the observations themselves are described as having been made years earlier. The final presentation is consequently the product of an observing sequence and later image processing, not a single untouched camera frame taken for the daily feature.
NASA describes the original material as a stack of those 223 exposures. The reported number of frames also helps explain why the page focuses so explicitly on cleanup. A collection of exposures can preserve repeated views of the same astronomical subject while also gathering interruptions that occur in particular frames or affect particular pixels. The feature does not claim that every visible mark in the earlier version belongs to one source or arose in one moment. Instead, it identifies several kinds of defects and trails visible in the unprocessed presentation.
Among the conspicuous marks cited are bright, parallel trails attributed to airplanes and long, continuous tracks attributed to satellites. NASA also points to short streaks from cosmic rays and to bad pixels. The descriptions distinguish the artifacts by their appearance as well as their proposed origin: some cross the image as tracks, while others are brief marks or pixel-level faults. In each case, the explanation treats them as additions imposed on the record of Andromeda, not as structures in the galaxy itself.
That framing limits what can fairly be concluded from the comparison. The page supports the conclusion that its featured processing reduced the named disturbances and sought a cleaner representation of M31. It does not, on the supplied account, provide a frame-by-frame audit, settings for the processing steps, or a quantitative measurement of how much each type of artifact was removed. Nor does it offer the paired images as a general test of every approach to astronomical image processing. It is a specific account of this image and this workflow.
The “before” image is part of the evidence
By putting the earlier-looking version beside the cleaner one, the feature makes the unwanted material legible to a viewer. Aircraft trails are described as bright and parallel. Satellite trails are described as long and continuous. Cosmic-ray marks are shorter, and defective pixels are a separate camera-related flaw. These are not presented merely as abstract technical concerns; they are named as visible elements of the source image.
The before image also complicates the common shorthand that a camera simply captures an object as it is. NASA’s description says the exposure passed through Earth’s atmosphere and a busy sky before being recorded with a digital camera. In that account, the raw result contains both the intended astronomical subject and conditions introduced closer to the observer. The comparison is useful precisely because it retains the earlier version rather than treating its irregularities as something that should never be seen.
Andromeda remains the subject in both versions, but the page’s editorial emphasis falls on the boundary between the subject and the contamination around it. The final image is described as an effort to better reproduce the galaxy’s real appearance by minimizing effects that have no connection to the galaxy. That is a more limited claim than saying processing reveals a perfectly complete or final truth about M31. The source does not promise perfection; it says the identified defects were greatly reduced.
Its language is also important because it does not characterize the work as an attempt to mislead the audience. NASA explicitly contrasts the stated aim with the construction of a digital fantasy. Yet that assurance rests on the source’s description of intent and method. The page offers a visual comparison and identifies categories of removed or reduced material, but the available information does not independently establish every decision made during processing or resolve how a viewer should judge the final image beyond the source’s explanation.
The title names Photoshop, while the explanation names other tools
There is a notable terminology problem within the NASA page itself. Its title uses “Photoshop” to describe the before-and-after comparison. But the explanation says the imperfections were not removed with Photoshop specifically. Instead, it names Astro Pixel Processor, DeepSkyStacker and PixInsight among the software packages used to greatly reduce them.
The two formulations need not be read as identical statements. In the title, “Photoshop” appears to function as a familiar shorthand for image alteration or retouching. In the body text, however, NASA draws a specific technical distinction: the work it describes was done with a series of other software packages, not with Photoshop itself. Readers should therefore not take the headline-style wording as confirmation that Adobe Photoshop was the program used in the workflow.
The discrepancy is material rather than cosmetic. The page’s argument is partly about how the finished view was made and why. When it identifies particular tools and says they were used to reduce airplane trails, satellite trails, cosmic-ray streaks and bad pixels, it offers a more exact description than the title’s broad label. The body explanation is the stronger basis for the software claim because it directly addresses the method.
At the same time, the supplied page context does not spell out the division of labor among Astro Pixel Processor, DeepSkyStacker and PixInsight. It says the packages were included in the processing sequence, but does not assign a named operation to any one program. It also does not describe whether additional tools were used, how individual exposures were selected, or whether some traces remained in the final version. Those omissions do not negate the comparison; they define the limits of what the source establishes.
Why the paired image changes how the finished portrait is read
The value of the feature lies less in an assertion that the processed view is untouched than in its acknowledgement that the view has been worked on. NASA places an image containing terrestrial and digital interference beside one designed to reduce that interference. That presentation gives viewers a way to see the stated reason for processing: to reduce visible material that the page says is unrelated to the distant galaxy.
It also separates two ideas that are often collapsed into one. First, the image has been processed. Second, the stated purpose of that processing was to remove or reduce contaminants rather than invent galactic structures. The NASA page argues for the second point by naming the interference visible before processing and by setting the two views together. But it does not provide enough detail in the supplied material to independently evaluate every consequence of the workflow for faint or small-scale image detail.
For readers, the comparison offers a practical way to interpret a polished astronomical image. A clean final view should not automatically be understood as a single direct exposure without intervention. Here, NASA says the underlying picture derives from hundreds of exposures and was refined with several specialized packages. The displayed result remains tied, in the page’s account, to the recorded Andromeda observations, while the altered elements are characterized as aircraft, satellites, cosmic rays and defective pixels rather than features of M31.
The feature credits Kees Scherer for the image. It lists Robert Nemiroff, Jerry Bonnell, Cecilia Chirenti and Keighley Rockcliffe as authors and editors. Those credits identify the people attached to the presentation, but they do not expand the available technical record beyond the description on the page.
This report has not been independently corroborated. It is based on NASA’s single Astronomy Picture of the Day page and its account of the 2019 observations, the 223 exposures, the named artifacts and the software used. The internal mismatch between the title’s reference to Photoshop and the explanation’s statement that Photoshop was not specifically used is unresolved in the material provided; the latter is the page’s more detailed description of the processing.
For further context on this subject, see Andromeda comparison highlights the work of removing earthly artifacts from sky images.
Reporting notes
What is confirmed: NASA names Astro Pixel Processor, DeepSkyStacker and PixInsight as part of the processing sequence.
Why this matters: The comparison distinguishes unwanted Earth- and camera-related artifacts from features of the galaxy while exposing the role of processing.
What remains unclear: The supplied material gives no processing settings or frame-by-frame account, and its title conflicts with its specific software description. This report is based on one source and has not been independently corroborated.
