Images to PDF

How do you combine mixed images into a correctly sized PDF?

Add the images, check their order in the thumbnail grid, and use the conversion button to create one PDF page per image. For example, combine a JPG receipt, a PNG confirmation, and a WebP diagram into a three-page submission. Drag the receipt first, choose Letter with Contain, and leave Page margin at zero for a straightforward layout. Auto (per image) lets wide and tall images use different page orientations.

For a concrete layout calculation, put a 1200 × 800 image on a portrait Letter page. Letter measures 612 × 792 PDF points. Contain uses the smaller of 612 ÷ 1200 and 792 ÷ 800, so the scale is 0.51. The image occupies 612 × 408 points, centered with 192 points above and below. This preserves all edges without changing the source pixel count. Review the preview before converting, especially when switching to Cover.

Which formats work, and what happens to their quality?

The picker accepts JPG, PNG, WebP, BMP, GIF, TIFF, HEIC, HEIF, and SVG. Acceptance and decoding are separate: formats requiring conversion depend on the browser's createImageBitmap support. A file that cannot be decoded is skipped. If a HEIC or TIFF does not appear in the grid, export it as JPG or PNG first rather than assuming the PDF contains it.

Processing with Original selected, no crop, and Max PDF size Off

Input	PDF preparation	Quality control
JPG / JPEG	Direct JPEG embedding	No additional JPEG encoding
PNG	Direct PNG embedding	No canvas conversion
WebP, GIF, SVG	Browser decoding, then PNG	Conversion quality does not change PNG encoding
BMP, TIFF, HEIC / HEIF	Browser decoding, then JPEG	Conversion quality controls JPEG encoding

WebP routes to PNG even when it is opaque; the route uses the file's MIME type rather than testing individual pixels for transparency. SVG becomes a raster image, not editable vector artwork. The tool does not expand animations or multipage source images into separate PDF pages. Conversion quality starts at 92% and ranges from 50% to 100%; it does not recompress untouched native JPG files.

How are images stored inside the finished PDF?

PDF pages place image objects within a coordinate system measured in points, with 72 points per inch. The image's pixel dimensions and its drawn size on the page are separate properties. A large photograph can be drawn small without adding or discarding pixels through layout alone. Enlarging that drawing makes individual source pixels cover more page area; it cannot restore detail missing from the photo.

The embedding library stores JPEG data in an image stream using DCTDecode. PNG embedding decodes the source and builds compressed image data, with a separate soft mask for alpha when needed. Direct PNG embedding therefore does not mean copying the PNG file container unchanged. Transparency describes how an image combines with the page behind it; it is distinct from the page's paper dimensions.

Rotation changes how the image is drawn in the PDF. Cropping and scan filters instead prepare new pixel data, while a size target invokes JPEG optimization. This distinction helps explain why changing orientation alone and enabling a size limit have different effects. The thumbnails are editing aids; PDF creation uses the prepared image data rather than a screenshot of the grid.

Which page size, orientation, and DPI should you choose?

Fit to image is the default. It derives each page from that image's dimensions rather than imposing one paper size. Fixed choices are A4, A3, A5, Letter, Legal, Tabloid, and Custom. The A-series labels follow ISO 216: A4 is 210 × 297 mm, A3 is 297 × 420 mm, and A5 is 148 × 210 mm. Letter is 8.5 × 11 inches, Legal is 8.5 × 14 inches, and Tabloid is 11 × 17 inches.

Custom dimensions use millimeters. Fixed pages offer Portrait, Landscape, and Auto (per image). Auto considers the cropped dimensions and manual rotation. With multiple images on fixed pages, select a thumbnail and enable Customize this image to override its orientation or fit while other pages retain their settings.

Resolution appears for Fit to image, with 72, 150, and 300 DPI choices. The calculation is page length in points = image pixels × 72 ÷ DPI, plus the margins on both sides. Higher DPI makes a smaller physical page from the same pixels; it adds no image detail. Fixed paper dimensions ignore this DPI setting. Page margin ranges from 0 to 50; although the UI labels it px, the engine uses that numeric value as PDF points.

When should you use Contain, Cover, Stretch, or Crop?

Contain preserves proportions and keeps the complete image within the available area at the normal centered position. Cover preserves proportions but fills the area, with excess extending beyond it. Use its zoom and directional controls to choose the visible portion. Stretch fills both dimensions independently and distorts an image whose proportions differ from the page.

Crop is a separate rectangular edit. Choose Free, A4 portrait, Letter, or Square, adjust the box, then Apply. Reset restores the full selection. Arrow keys move by 1%, Shift uses 10%, and Alt resizes. Cropping happens before size optimization at conversion time: cropped PNG stays PNG, while cropped JPEG is encoded at 92% quality. Rotation alone is drawn into the PDF without re-encoding the image.

How do scan filters, page numbers, and document details work?

For document photos, select Scan cleanup filter before adding files, or remove and re-add files already in the grid. Original leaves them unfiltered. Grayscale and Increase contrast produce JPEG; Black and white produces PNG and starts with a threshold of 128. These are pixel filters, not text recognition: the converter does not generate searchable OCR text or correct perspective.

Add page numbers offers Bottom center, Bottom right, and Top right positions, with either a number or Page 1 of N format and a starting number. Document title (optional) and Author (optional) set PDF fields; when either is filled, Creator and Producer become brownie.tools. For a document requiring signatures afterward, continue with Fill and Sign PDF.

How do you turn phone photos of documents into a clean PDF?

Start with readable photographs: flatten the paper, hold the camera parallel to it, include every edge, and use even lighting without a shadow across the text. A rectangular crop can remove the desk around a sheet, but it cannot straighten a trapezoid caused by photographing at an angle. Retake a blurred or badly skewed page before assembling the document.

1. Decide whether the originals need cleanup. Use Grayscale for neutral document pages, or try Black and white for clear dark lettering. Compare faint handwriting, stamps and pale marks before committing to a threshold. Keep Original for material whose colors carry meaning.
2. Select the filter before adding the photos. Confirm that every expected sheet appears, especially when importing phone formats that require browser decoding.
3. Arrange pages in reading order, rotate sideways sheets and crop unnecessary borders. Keep signatures, reference numbers and notes inside the crop.
4. Choose A4 or Letter with Contain for a consistent document package. Use per-image overrides when a wide attachment needs a landscape page. Add page numbers only if they will not obscure existing content.
5. Convert, open the result and inspect small text at a useful zoom. If requesting a smaller PDF, check legibility again after optimization as well as the reported file size.

The pages remain images. Search, text selection and screen-reader reading order are not created by these cleanup steps. When a submission requires searchable or accessible text, use a workflow that supplies OCR and document structure rather than assuming a PDF extension provides them.

Which settings suit common image-to-PDF tasks?

• Receipts and supporting evidence: keep the transaction date, totals and identifying details visible. A consistent paper size and Contain help prevent clipped edges. Confirm the recipient's page and file-size requirements before export.
• Scanned lessons or handwritten notes: sequence sheets by their actual contents, including backs of pages. Test a cleanup filter on faint pencil marks before applying it to the whole set.
• Photo portfolios and lookbooks: Fit to image follows each photograph's proportions. A fixed page size gives a consistent viewing frame; Cover is useful only when losing edge content is acceptable.
• Mixed project attachments: combine screenshots, camera images and exported diagrams in one reading sequence. Check rasterized SVG labels and confirm that each accepted file actually decoded.
• Print handoffs: choose the requested physical page dimensions and inspect the effective detail at that size. A DPI choice does not create resolution or establish press readiness; ask the printer about bleed, profiles and other production requirements.

What if the PDF is too large or an image fails?

Max PDF size is Off initially. Its 1, 2, 5, and 10 MB presets and custom 0.1–100 MB limit trigger JPEG re-encoding for all images, replacing transparency with white. The optimizer searches a shared quality level and tries up to four downscale rounds. It keeps the long side at least 1,000 pixels without enlarging smaller originals.

The target uses decimal megabytes: one MB is 1,000,000 bytes. Planning estimates PDF overhead, so always check the final reported size against your upload requirement. If the limit is missed, you can still download the smallest attempted result. Reduce the batch or prepare smaller source images if needed. Large batches also depend on available browser memory; there is no promise of unlimited capacity. Use Merge PDF when your inputs are already PDFs.

Do local processing and PDF conversion remove metadata?

Your selected file bytes are read locally, converted in the browser, and assembled with pdf-lib. The page's Content Security Policy permits connections only to this site. This describes connection destinations; the conversion path itself does not upload your images. You choose when to download or share the result.

Do not treat conversion as metadata sanitization. An untouched JPG is passed directly to JPEG embedding, so information stored inside those bytes can remain. Canvas operations re-encode image pixels, while PDF title and author are separate document fields. Before submitting, open the PDF and inspect page count, reading order, cropped edges, text legibility, and actual file size. Back to editing lets you revise the batch before downloading again.