> The astronomical news community would do the public a very big favor if they started stating clearly which pictures are taken by telescopes and which are cgi.

> The astronomical news community would do the public a very big favor if they started stating clearly which pictures are taken by telescopes and which are cgi. I agree that there ought to be clearer indications of which images come from sensor data and which are purely simulated, but all sensor images are heavily processed. I worked on Mars Global Surveyor, Mars Odyssey, MRO, LRO, and other NASA missions, and wrote image processing software for various instruments on those spacecraft (primarily infrared cameras). They're not simply snapshots like you'll get from your DSLR. Even the most DSLR-like cameras tend to image in many spectral bands (e.g. 11 bands ranging from infrared through visual into UV). An image you see in a magazine typically selects three spectral bands and assigns them to red, green, and blue. Almost none of the images you see are true color, in part because almost none of the instruments have 3-band RGB sensors like you might see in your DSLR. There are many reasons for this, but the two biggest ones are that much of the universe is opaque to visual frequencies (you need infrared or other sensors to penetrate clouds of dust), and much of the scientifically interesting data isn't present in the visual spectrum (infrared is used to determine mineral composition, like finding hematite on Mars, which is evidence of water in the past). Images of the surface of planets comes from a complex stitching-together of long, thin strips of data (again in many non-visual frequencies), which are taken from different angles, at different times of day. In order to create the uniform, visually-appealing (and also scientifically-valuable) images of the surface of Mars that you see, the images are heavily processed to balance albedo and adjust other anomalies. One little-known fact is that gravitationally, none of the large, rocky celestial bodies are spherical; the moon in particular is a very lumpy potato shape. This means that an orbit around the moon is nowhere near a perfect ellipse; it wobbles up and down all the time, which means the distance of the camera from the surface changes all the time. All of this has to be compensated for. During processing, everything is stretched to align, and hundreds/thousands of images of the same region from different times (over many years) are averaged together to get the final product. All of this is a long way of saying that literally every astronomical image you've seen is computer-generated and heavily processed. There's value in knowing which images started with sensor data and which are pure simulation, but there's no value in non-computer-generated imagery. It's hard to explain this quickly in an image caption. We used to have to deal with conspiracy quacks all the time, demanding that we release the "real images" rather than our "manipulated images" so that they could find the aliens on Mars, but the simple truth is that there are no "real images". It's all sensor data in a form that is completely useless without extensive processing.