Image DPI Checker for Publication Figures

How this image DPI checker works

The number that matters for print is effective resolution: the pixels you have divided by the width the figure will be printed at. The checker reads the pixel width from your file, converts the target width to inches, and divides. A 2,000-pixel-wide image printed in an 89 mm (3.5 in) column has about 571 DPI; the same image stretched across a 183 mm (7.2 in) page has about 278 DPI and would miss a 300 DPI minimum.

• Effective DPI = pixel width ÷ print width in inches (1 inch = 25.4 mm).
• Pixels needed = target DPI × print width in inches, rounded to the nearest pixel.
• Maximum print width = pixel width ÷ target DPI.
• Journal presets use the widths and resolution minimums published in each official guide. Where a guide gives no value, the checker falls back to the common 300 / 600 / 1,000 DPI figure resolution guideline and labels it as general.

DPI vs PPI: what the number in your file means

PPI (pixels per inch) describes a digital image; DPI (dots per inch) originally described printer output. Journals and most software use DPI for both, and so does this page. The DPI stored in a file is only a label: a PNG pHYs chunk, a JPEG JFIF density, or an EXIF or TIFF XResolution tag telling software how large to print the pixels. Changing that label from 72 to 300 makes the default print size smaller but adds no detail. A 1,000-pixel image tagged 300 DPI and the same image tagged 72 DPI contain identical information. That is why this checker shows the embedded value but grades your figure on pixel dimensions and target width.

Why upscaling does not fix low resolution

Resampling an image to more pixels interpolates new pixels from the old ones. The file then passes a simple 300 DPI checker, but edges stay soft, small text stays blurry, and fine structure in micrographs or blots is not recovered. Science's instructions for revised manuscripts state that upsampling of images is not permitted, and image-integrity policies at many journals scrutinize processing. The fix is to go back to the source: re-export the chart from your plotting software at the final size, or use the original camera or microscope file.

Use vector formats for line art

Graphs, diagrams, and schematics made of lines and text are best kept as vector files (SVG for editing, PDF or EPS for submission), which scale to any size without pixels. Publishers ask for very high raster resolution for line art, such as 1,000 DPI at Elsevier or 1,200 DPI at Springer and ACS, because raster lines look jagged at lower resolution. If your journal accepts PDF or EPS, a vector export usually removes the problem. Photographs and micrographs stay raster, and combination figures that mix images with text often need 500 to 600 DPI.

Export at the right size from common tools

Set the physical size first and the resolution second, so fonts print at the size you intended.

• matplotlib: create the figure at print size, for example plt.figure(figsize=(3.5, 2.5)) for an 89 mm column, then fig.savefig("fig1.tiff", dpi=600). For line art, fig.savefig("fig1.pdf") keeps text and lines as vectors.
• R and ggplot2: ggsave("fig1.tiff", width = 89, height = 60, units = "mm", dpi = 600, compression = "lzw"), or ggsave("fig1.pdf", width = 89, height = 60, units = "mm") for a vector file.
• PowerPoint and Keynote: slides are sized for screens, and exporting a slide as PNG or JPEG often gives fewer pixels than a full-width figure needs. Build the figure at its final width, export as PDF where the journal accepts it, and check any raster export here.
• Image editors: when changing print size, turn resampling off so only the DPI label changes. If the effective DPI is still too low, the source needs more pixels.