Changing image DPI changes a print-density instruction stored in a JPEG or PNG; it does not change the image's pixel dimensions or add image detail.
DPI tells supporting software how many existing pixels to place in each physical inch, so the same pixel grid can have different intended print sizes.
DPI, pixels, and physical size
DPI means dots per inch. In a digital image file, it acts as a print-density instruction: how many image pixels should be placed in each printed inch. It is separate from pixel dimensions, which are the numeric width and height of the image grid.
The basic relationship is intended size in inches = pixel count ÷ DPI. For the width, use the image's pixel width; for the height, use its pixel height. This is a layout calculation, not a guarantee that every printer or application will interpret the metadata in the same way.
One worked example: for a 3600-pixel-wide image intended to print 12 inches wide, Target DPI = 3600 pixels ÷ 12 inches = 300 DPI. At that density, the file can instruct a compatible layout program to treat those pixels as a 12-inch-wide image.
Raising DPI makes the same pixels imply a smaller physical layout, while lowering it makes them imply a larger one. Neither direction creates pixels, sharpens edges, recovers missing detail, crops, or enlarges the picture. A low-resolution scan can still look soft when printed large.
| Term | What it represents | Effect of changing DPI |
|---|---|---|
| DPI | A saved print-density instruction | Changes the metadata value |
| Pixel dimensions | The image grid's width and height | Stay unchanged |
| Image detail | The visual information captured in the source | Does not increase |
| Intended print size | The physical dimensions suggested by an application | May change in supporting software |
On a website, DPI normally has no visible effect. Browsers lay out images using CSS and the page's dimensions rather than the print-density instruction stored in the file.
JPEG and PNG store density differently
For a JPEG, Image DPI Changer reads or writes JFIF APP0 density fields and uses dots per inch. The JFIF specification defines how those density fields are stored. If the selected JPEG has no JFIF density header, the tool adds a small JFIF header without altering the compressed picture.
For a PNG, the tool writes the standard pHYs physical-pixel chunk using pixels per metre. The PNG specification defines that chunk. The page converts the entered DPI using the exact 0.0254-metre-per-inch relation, but PNG stores an integer number of pixels per metre. A readback can therefore differ from the requested DPI by a very small rounding fraction without any change to the image pixels.
The tool also reads standard EXIF resolution fields after a file is selected, but its output focuses on JFIF or pHYs metadata. It does not attempt to merge or rewrite every possible EXIF, XMP, ICC, IPTC, camera, GPS, or vendor extension. A JPEG can contain several forms of resolution information, and applications may choose a different field from the one that was updated.
| Format | Updated field | Storage unit | Possible readback |
|---|---|---|---|
| JPEG | JFIF APP0 density fields | Dots per inch | The requested whole-number DPI |
| PNG | pHYs physical-pixel chunk | Integer pixels per metre | A tiny rounding difference may appear |
Set a new DPI on one JPEG or PNG
- Open Image DPI Changer and choose one JPEG or PNG file up to 25 MiB. Review any detected print-density metadata, including whether the page found a usable JFIF or pHYs density.
- Enter a whole-number target DPI, such as 72, 150, 300, or 600. This changes the print metadata only: the output retains the original pixel width and height without cropping, sharpening, enlarging, or compressing the image.
- Select Update image DPI, download the new copy, and verify its intended physical size in the destination layout or print application. Keep the original file separately as a backup.
Processing happens locally, and the file is not uploaded to a server. The source pixels and compressed image data are copied as bytes rather than redrawn through a browser canvas or encoded a second time, avoiding a new lossy JPEG encoding. The downloaded file can still differ at the file level because its density fields are updated and a small JFIF header may be added.
If an operation fails, the page clears its previous result link instead of presenting an older file as though it had just been updated.
Check the downloaded file in its destination
Open the downloaded file in the program that will place it in the final layout. Set the frame or placement to the intended physical size, then inspect the dimensions, resolution information, and output preview shown by that program. A metadata viewer can confirm the stored JFIF or pHYs value, but the destination layout application remains the practical test.
Verification matters because a layout application, printer workflow, or export tool may use EXIF or another resolution field instead of the JFIF or pHYs field that was just updated. No metadata tool can force every downstream program to interpret print size identically. For a separate readback checklist, see how to check the result after using an image DPI calculator.
For a print provider, follow the provider's documented requirements for pixel count, bleed, export profile, and file format. A DPI label alone may not satisfy the complete production specification.
Before editing, use the DPI Converter when you need to work out a physical size from pixel dimensions and intended DPI. This helps avoid choosing a density that would force an unsuitable layout.
Choose the operation that matches the goal
DPI metadata is the right target when the existing pixels are suitable for the planned print size but the saved density instruction needs to match the layout.
- You need the same source grid to have a different intended print size.
- A layout program is interpreting the density instruction inconsistently with your intended dimensions.
- A print provider requests a particular DPI value and separately confirms that the pixel dimensions and other production requirements are acceptable.
Use a different operation when the objective goes beyond metadata:
- If the actual pixel dimensions must change, use the Image Resizer. Resampling changes the grid but cannot restore detail that was absent from the source.
- If you need a format conversion, crop, added margin, or a different canvas, use a workflow designed for that operation; the DPI changer does not perform those tasks.
- If small text, edges, or photographs look soft at the planned output size, obtain a higher-resolution source or inspect the final output more closely. DPI metadata cannot compensate for insufficient source pixels.
Changing DPI can therefore correct a print-layout instruction while leaving the underlying image quality exactly as it was.
Limits and metadata caveats
The accepted input is deliberately narrow: one JPEG or PNG, no larger than 25 MiB, with a whole-number target DPI. The tool validates the file signature and bounded header structure. Truncated JPEG marker segments, malformed PNG chunks, duplicate pHYs physical-pixel chunks, and a pHYs chunk placed after image data are rejected rather than patched optimistically.
The tool does not convert image formats, change canvas size, add print margins, estimate printer quality, or guarantee that a social platform will retain the density value. It focuses on portable JFIF or pHYs fields rather than synchronizing every possible metadata family.
Keep the original file, test the downloaded image in the workflow that matters, and inspect small text, edges, and photographs at their final output dimensions. If those elements look soft, the underlying issue is insufficient source pixels rather than a missing DPI label.
Related reading: Image to Base64 Explained: Formats, Limits, and Use.