Image to PDF Converter

The Image to PDF converter combines JPG, PNG, WebP, and other image files into a PDF document. Common uses include scanned paperwork, receipts, homework, photo collections, presentation material, evidence files, and form systems that accept PDFs more reliably than loose images. Creating a good PDF involves more than putting pictures in order: page size, portrait or landscape orientation, margins, scaling mode, original resolution, and compression all affect readability and file size. This tool runs in the browser, so images do not need to be uploaded to a server. For official submission or printing, the final PDF should still be opened and checked for rotation, cropping, missing pages, and blurry text.

PDF pages are measured in points, where 1 point equals 1/72 inch. ISO 216 A4 therefore expands to 595.28 × 841.89 pt (210 × 297 mm), A3 to 841.89 × 1190.55 pt, and US Letter (8.5 × 11 in) to 612 × 792 pt. Generation happens in-browser via pdf-lib, which lets the page allocate a PDFDocument, add a page at the chosen size, and place each image into a Form XObject anchored to the page coordinate space. The embedding strategy depends on the source format. JPEG bytes are passed through unchanged using the DCTDecode filter, which avoids re-encoding entirely and keeps the file as compact as the original. PNG goes through embedPng, which decodes the raster, drops it into a FlateDecode stream, and adds an SMask object when an alpha channel is present, so transparency survives the round trip. Image position is then computed under the chosen fit mode: 'contain' scales the image down so both axes fit within the page minus margins, 'cover' fills the page and crops the overflow, and 'actual' treats each image pixel as one 1/72-inch unit and centers the result on the page. The sharp edges of this pipeline are where surprises hide. Actual-size on a 2000 px wide photo produces a 2000/72 ≈ 27.8 in page demand that overflows A4 and silently crops at print time. Embedded ICC profiles are flattened to sRGB, so wide-gamut photos look mildly desaturated on calibrated print. And because each page stores its image as a raster XObject with no text layer, downstream OCR is a separate step, which matters for any agency that requires a searchable PDF for filings.

• Coordinate system: 1 PDF point = 1/72 inch; A4 = 595.28×841.89 pt, A3 = 841.89×1190.55 pt, US Letter = 612×792 pt.
• JPEG inserts via DCTDecode without re-encoding, so the embedded bytes equal the source file size; PNG goes through FlateDecode with an SMask object when alpha is present.
• Fit modes: contain preserves aspect with margin, cover fills and crops, actual treats source pixels as 1 pt = 1/72 inch units and centers the image.
• Page-size math: a 1500 px wide photo at actual size becomes 1500/72 ≈ 20.8 in, which is wider than A4 and will silently crop on print.
• Colour space: embedded ICC profiles are flattened to sRGB on export, so wide-gamut sources look subtly desaturated on calibrated print proofs.
• No text layer: each image is a raster XObject, so the output is image-only and downstream OCR is a separate pass when an agency demands searchable PDF.