Sending a raster logo to a printer or sign maker who asked for vector artwork.
File formats
Raster image
Also called bitmap image, pixel image or raster vs vector. Here is what it means, when it changes what you export, and what people get wrong.
File formatsRuns on your deviceUpdated 10 August 2026
The short answer
A raster image is a picture stored as a grid of coloured pixels, which is how photographs, screenshots and most web images work. Every pixel holds one fixed colour value, so the file has a fixed ceiling on detail, and enlarging it past that means inventing pixels rather than revealing them.
Updated
The short version
Raster image at a glance
| Quick fact | Detail |
|---|---|
| Also called | Bitmap image |
| Stores | One colour value per pixel |
| Resolution dependent | Yes, detail is fixed at capture or export |
| Common formats | JPEG, PNG, WebP, AVIF, GIF, TIFF, BMP |
| Vector counterpart | SVG |
| Displays as | Pixels, since screens are raster devices |
Slide the table sideways to see every column.
How is a raster image different from a vector?
A raster stores the result. A vector stores the recipe. A vector file describes a circle as a centre point, a radius and a fill colour, so the renderer can draw it crisply at any size. A raster file describes the same circle as several thousand individual pixel values, which is exactly right for a photograph and exactly wrong for a logo that has to work on a business card and a lorry.
| Raster | Vector | |
|---|---|---|
| Made of | Pixels on a fixed grid | Points, curves and fills defined by maths |
| Scaling | Detail is capped, degrades when enlarged | Redraws sharp at any size |
| Best for | Photographs, textures, screenshots | Logos, icons, type, diagrams |
| Typical formats | JPEG, PNG, WebP, AVIF, TIFF | SVG, EPS, AI |
| File size driven by | Pixel dimensions and colour depth | Number of shapes and nodes |
Why do raster images fall apart when you enlarge them?
Because there is no extra detail to find. A 500 pixel wide photo contains 500 columns of information and nothing more. Scaling it to 2000 pixels means the software has to produce three new pixels for every original one, and it does that by interpolating: averaging neighbours to guess what should sit between them. Averaging is inherently soft, so the result looks blurred, and if you use a cruder method you get blocky pixelation instead.
AI upscaling changes the guess, not the maths. A trained model predicts plausible detail based on the millions of images it has seen, which is why an upscaler can produce convincing skin texture or sharp lettering where interpolation gives mush. It is still generating detail rather than recovering it, so it works best on ordinary subjects and least well on fine text, logos and anything where the exact original shape matters. Photomop's upscaler runs this in the browser, but the principle holds wherever you do it: start from the largest original you have.
Which formats are raster and which are vector?
Everything that comes out of a camera or a screen is raster: JPEG, PNG, GIF, WebP, AVIF, HEIC, TIFF, BMP and RAW. Vector formats are the ones drawn in illustration software: SVG, EPS and AI. PDF sits across both, because a PDF page can hold vector artwork, live text and embedded raster photos all at once.
Note that SVG can embed a raster image inside itself. A logo that looks like an SVG but contains a pasted PNG will not scale any better than the PNG did, which catches people out when a printer asks for vector artwork.
When is raster the right choice?
Whenever the subject has continuous tone: photographs, textures, gradients, anything shot with a camera. No sensible vector description exists for a face or a piece of fabric, and trying to trace one produces flat, posterised shapes.
Reach for vector when the artwork is made of defined shapes and needs to survive resizing: logos, icons, wordmarks, diagrams and print artwork. In practice most projects need both, with the vector file as the master and raster exports at whatever sizes each platform asks for.
What people get wrong about raster image
Each one is a real failure mode, not a style preference.
Enlarging a 500 pixel screenshot to 3000 pixels and expecting the text inside it to become readable.
Saving a flat-colour logo as JPEG, where compression smears the hard edges that PNG or SVG would keep crisp.
Assuming an SVG is scalable when it just wraps an embedded PNG.
Questions people ask
Raster image, answered
The follow-up questions people search for once they have the definition.
Is a JPEG a raster image?
Yes. JPEG, PNG, GIF, WebP, AVIF, HEIC, TIFF and BMP are all raster formats. They differ in how they compress the pixel grid and what extras they support, such as transparency or animation, but underneath they all store a fixed grid of pixel values.
Can you convert a raster image into a vector?
You can trace it, and for flat artwork such as a black logo on white the result is often excellent. Photographs do not convert usefully: tracing turns continuous tone into flat colour patches that look like a poster print rather than a photo. Redrawing by hand beats automatic tracing for anything important.
Why does my image look blurry after I resize it?
Because enlarging a raster asks software to invent pixels it does not have. Interpolation averages neighbouring pixels, and averaging softens edges. Shrinking rarely causes this problem, so wherever possible export from the largest original rather than scaling a small file up.
Which is better, raster or vector?
Neither, they solve different problems. Raster is the only sensible option for photographs and anything with continuous tone. Vector is the right choice for logos, icons and type that must stay sharp at every size. Most real projects keep vector masters and export raster copies.
Same cluster
More from the glossary
Neighbouring entries, so every term in the set is one hop from every other.
RAW image file
A RAW image file is the unprocessed data a camera sensor recorded, saved before the camera turned it into a finished picture.
Quality setting
A quality setting is the number, usually 0 to 100, that tells a lossy encoder how aggressively to discard detail.
Progressive JPEG
A progressive JPEG stores the image as a series of scans that each add detail, so a rough version of the whole picture appears first and sharpens as more data arrives.
From the studio
Tools that do this
Each one runs on your own device, so the photo stays where it is.
Sources
Where the facts on this page come from. Every link opens the specification, standard or documentation the claim was read at.
Keep reading
Elsewhere on Photomop
Platform image rules
Image size, file type and background rules for 22 selling platforms, each one sourced.
Read itConvert image
JPG, PNG, WebP and AVIF, with an honest answer about HEIC and alpha.
Read itAll the tools
The whole toolbox in three families, with what each one needs before it can run.
Read itOn-device privacy
The pipeline in five steps, what leaves your device, and how to check it yourself.
Read it
Now do it to a photo
Photomop is a photo studio that runs on your own device. Resize, crop, compress, convert, batch edit, remove a background and change one all work in the browser tab you are reading this in, at full resolution and with no watermark. The on-device page shows you how to check that the photo stays put.