You can handle this directly in your browser with our free pdf to images tool — nothing gets uploaded.
Walk into any office and you will find three scan modes on every device: color, grayscale, and monochrome (1-bit black and white). Most people tap color because it is the default, then wonder why their PDFs balloon to 8 MB each, why their phone dies halfway through a stack of receipts, and why their printer is out of cyan or magenta ink again. The good news: for the vast majority of document work, 1-bit black and white is not just "good enough" — it is genuinely the right choice. This guide explains the physics behind why, walks you through concrete settings on HP, Canon, and Epson hardware, and tells you exactly when you should *not* reach for monochrome.
Before picking a setting, it helps to understand what the sensor is doing to each pixel.
Color mode samples the red, green, and blue channels separately, typically at 24 bits per pixel (8 bits per channel). A 1080 x 1350 page at 150 dpi in color can easily produce 3–5 MB of raw data before compression.
Grayscale collapses RGB into a single intensity value, usually 8 or 12 bits per pixel. A 200 dpi grayscale page of mixed text and photos lands around 1–2 MB. It looks smooth to the human eye and is fine for X-rays, slides, and scanned artwork.
Monochrome (1-bit B&W) is the most aggressive. The scanner applies a threshold: every pixel becomes either pure black or pure white — no gray in between. One page of 300 dpi text in 1-bit monochrome routinely lands between 50 and 200 KB, sometimes even less when the page is mostly empty.
| Scan Mode | Bits per Pixel | Typical 300 dpi Page Size | Best For | Worst For |
|---|---|---|---|---|
| Color (24-bit) | 24 | 3–6 MB | Photos, colored tabbed index cards | Plain text, battery, ink |
| Grayscale (8-bit) | 8 | 1–2 MB | Medical scans, mixed media | Pure black-on-white text |
| Monochrome (1-bit) | 1 | 50–200 KB | Contracts, invoices, forms | Any document with real color |
The math is brutal. A 20-page invoice stack that color-scan renders at 80 MB can come in at 2–4 MB in monochrome — a 20x reduction. That is not a typo. It is why every serious OCR pipeline, every law-firm e-discovery workflow, and every government records archive defaults to 1-bit B&W.
OCR (optical character recognition) engines do not want to see the color of the ink. They want high-contrast edges. A 1-bit threshold strips every anti-aliased gray pixel and forces the boundary between letter and background into a crisp, unambiguous line. Modern engines like ABBYY, Tesseract, and the built-in readers in Adobe and Apple have been trained on thresholded input, so feeding them 1-bit B&W often produces higher character accuracy than feeding them the "smoother" grayscale image.
There is also a practical benefit: smaller files mean the OCR job starts faster and uses less RAM. On a phone, that translates into more battery life during long scan sessions — no heavy color deconvolution, no big buffers, no network upload of megabytes of redundant channel data.
A color laser or inkjet printer that prints a 24-bit scan has to lay down cyan, magenta, yellow, and black layers even for what was originally a black pen on white paper. A true 1-bit file only ever commands black. For text-heavy documents, that can cut color-ink consumption on the reprint leg by 60–90 percent, and it also halves the toner wear on the fuser. If your workflow is "scan, review, print a clean copy", staying monochrome end-to-end is the cheapest path.
Monochrome is the right answer roughly 80 percent of the time. The other 20 percent matters, so here is where to stop and switch modes:
If in doubt, scan in grayscale first, eyeball the file, and only commit to monochrome once you know the source is clean black-on-white.
Below are the exact paths and values that get the job done on the most common hardware. Adjust the threshold value per document batch rather than trusting the default.
1. Is 1-bit black and white really better for OCR than grayscale?
In most text-heavy document cases, yes. Thresholded pixels give the OCR engine clean, high-contrast letter edges, and the file is 10 to 50 times smaller, so batch jobs finish faster. If your source is printed at 72 dpi or lower and the letters have gray anti-aliasing, grayscale at 300 dpi may actually outscore 1-bit — test both on a sample page before committing to a batch.
2. What resolution should I pair with monochrome scanning?
300 dpi is the universal floor for OCR-grade text. Drop to 150 dpi only for storage archives where text is already digitized and you just need a visual record. Go up to 400 to 600 dpi when the source has small captions or fine rule lines, but expect file sizes to climb.
3. Can I convert a color scan to 1-bit later without quality loss?
You can, but you lose the chance to tune the threshold per region. A color to grayscale to 1-bit pipeline is strictly less information than thresholding the original sensor output, and it doubles the intermediate file size. If you know you need B&W, do it at the scanner.
4. Does monochrome scanning save battery on the phone app?
Yes. The color pipeline on a phone involves RGB demosaicing, white balance, and three-channel compression; monochrome skips all of that. On a mid-range Android phone, a 50-page monochrome batch has been observed to drain 4 to 6 percent less battery than the same batch in color, mostly because the Wi-Fi upload step transfers a smaller file.
5. What if my document has both clean text and a colored stamp?
Two passes: first, monochrome at 300 dpi for the text layer (fast, small, OCR-ready). Second, color at 300 dpi for the same page, then in your PDF editor replace the text layer with the monochrome version and keep the color stamp region. Most law-firm and accounting workflows already do this — it gives you the best of both worlds without inflating every single page.
Pick monochrome (1-bit B&W) by default for contracts, invoices, receipts, forms, and any document where the source is black ink on white paper. You will get a smaller file, sharper OCR, less ink on the next print, and more battery per scan session. Reach for grayscale the moment a photo, faded signature, or low-contrast print enters the frame, and save color for the small minority of pages where hue itself carries meaning. Master the three-mode choice and your scan workflow stops being a coin flip — it becomes a decision you can make in ten seconds per document.
Related reading: annual review scan reconcile paper records, batch ocr scanned pdfs, car repair records scan before selling.