PNG Chunk Viewer & Remover

Upload your PNG file to the PNG Chunk Viewer & Remover. The tool automatically parses every chunk and displays them in an interactive table. Select Remove Metadata mode to strip all tEXt, zTXt, iTXt, eXIf, and custom metadata chunks, or use Selective Removal to choose specific chunks. Click 'Sanitize & Download' to get your cleaned PNG — all processed locally in your browser.

No. All processing — PNG parsing, chunk inspection, metadata removal, CRC recalculation, and PNG rebuilding — happens entirely in your browser using raw binary parsing via Web Workers. Your PNG files never leave your device. No uploads, no servers, no third-party access.

PNG tEXt chunks are metadata containers that store human-readable key-value text pairs within PNG files. Common tEXt chunks include Software (editor name), Author, Comment, Copyright, Description, and custom application data. Unlike JPEG EXIF which has a fixed structure, PNG tEXt chunks are flexible and can store any text metadata the generating application chooses to embed.

Yes. Transparency is preserved because the PNG Chunk Viewer & Remover only modifies metadata chunks (tEXt, zTXt, iTXt, eXIf, and custom chunks). The IHDR, PLTE, IDAT, and IEND chunks — which contain the actual image data, palette, and pixel transparency — are never touched. Alpha channel data remains completely intact.

The tool processes one PNG at a time with full chunk inspection and metadata removal. The architecture is designed for future batch processing support with a Web Worker pool. For now, each file receives thorough forensic analysis before any metadata is removed.

Yes. Many applications embed tracking information in PNG metadata chunks. Design tools like Photoshop, Figma, and Canva add Software and Author fields. AI image generators like Stable Diffusion and Midjourney embed full generation parameters including prompts, negative prompts, and workflow data in tEXt chunks. Game engines and custom applications may embed proprietary metadata in custom chunk types.

PNG uses a chunk-based architecture where each piece of data — image header, color profile, pixel data, text metadata — is stored in independent chunks with their own length, type, CRC checksum, and data. JPEG uses marker-based segments within a continuous stream. PNG chunks are self-validating with CRC32 checksums, making them more robust but also requiring proper CRC recalculation after modification.

Yes. AI image generators commonly embed generation parameters in PNG tEXt chunks. Stable Diffusion stores prompts in a 'parameters' or 'Prompt' field. ComfyUI stores the full workflow graph. Midjourney may embed generation seeds and parameters. These chunks can reveal the exact creative process behind an image, making them a privacy concern for artists sharing their work.

Yes. The tool supports Edit mode for text metadata chunks (tEXt and iTXt). You can modify the values of software name, author, comment, prompt, and other text metadata fields instead of removing them entirely. This is useful when preserving the metadata structure is preferred over complete removal.

The tool automatically recalculates CRC32 checksums for every modified chunk. CRC32 is calculated over the chunk type and data fields combined. After removal, anonymization, or editing, the tool computes the correct CRC and writes it to the output file. The resulting PNG has fully valid CRC checksums and passes standard PNG validation.

Upload the PNG to the Chunk Viewer. The tool immediately shows the total chunk count, metadata chunk count, and custom chunk count. Metadata chunks are highlighted with a visual badge. You can click any chunk to view its full contents — including decoded text for tEXt and iTXt chunks. Any chunk that is not IHDR, PLTE, IDAT, or IEND is potentially privacy-sensitive metadata.

tEXt stores uncompressed key-value text pairs using Latin-1 encoding. zTXt is the same as tEXt but with the text value compressed using zlib/deflate compression. iTXt supports international text with UTF-8 encoding, language tags, and optional compression. All three serve the same purpose — storing text metadata — but with different encoding and compression characteristics.