Image Processing

Open a GLTF, GLB, OBJ, STL or FBX model and read what is really in it: distinct vertices and position entries as two different numbers, triangles, whether the surface is closed, and a screenshot with the model on it.

Read a linear or 2D barcode from an image or camera. Verifies the GS1 check digit, corrects a UPC-A that a browser reports as a thirteen-digit EAN-13, and names which decoder answered.

Chord diagrams whose fingerings are found by searching the fretboard, so the notes under the dots really are the chord's. Ten tunings, sixteen qualities, and the notes named on the page.

See an image through the two published models of colour vision deficiency, Machado 2009 and Vienot 1999, computed in linear light where those models are defined, with the colour pairs that stop being distinguishable listed by name.

A layered raster canvas where a stroke lands under the pointer at every zoom, undo really steps back, and the export is the canvas or a named enlargement of it.

Build an animated GIF89a written from the specification and read back before download, so the frame count, the delays, the loop count and the palette size are reported from the finished file rather than from the request.

Mark a picture up with arrows, shapes, text and freehand, in the picture's own pixels: an annotation is the same size in the file whatever the view is zoomed to.

Remove a background with a model or with a colour key that has a real alpha ramp, scored against a corpus whose right answer was chosen rather than found.

Process many images at once with one operation, with every output decoded again and checked against what was asked for, and the whole batch downloaded as one ZIP.

Arrange photos in one of twelve templates whose every cell rectangle, gap, margin and radius is stated in the output's own pixels before anything is drawn.

A tone curve drawn with its level count and clipping reported, per-range gains in stops of light whose three weights sum to exactly one, and a white balance that is a Bradford adaptation on a published locus. The neutral setting is proved to change nothing.

Extract a palette whose every colour is one the image actually contains, with the share of the picture each one covers. Median cut, k-means in CIE Lab or plain frequency, each named.

Compress to a quality or to a file size, with the size shown being the encoded file to the byte and the cost measured on the pixels: SSIM, PSNR, the worst channel and the share of pixels changed.

Whole-pixel crops with the rectangle typed or dragged, checked by decoding the output and comparing it with the pixels that went in. Quarter turns and mirrors are exact permutations, so nothing is clipped.

Compare two images with a perceptual engine (pixelmatch) that ignores anti-aliasing, or an exact byte engine. Groups changes into regions with bounding boxes, reports SSIM, and handles size mismatches with an explicit policy.

Six published edge operators (Sobel, Prewitt, Scharr, Roberts and both Laplacians) plus the real four-stage Canny, each checked against the browser's own feConvolveMatrix and each naming the paper it comes from.

The seven transforms Filter Effects Module 1 defines, computed from its published matrices and checked against the browser's own implementation, plus exposure in real stops and seeded grain.

Convert between the formats a browser can really write, then open the output and name the format its first bytes say it is. AVIF, BMP and GIF are disabled because a canvas returns a PNG for all three.

Count every pixel into 256 bins per channel, with the luminance definition named rather than assumed: Rec. 709, Rec. 601 or CIE L*. Exports the whole table as RFC 4180 CSV.

Remove EXIF, GPS, XMP and IPTC by rewriting the container rather than re-encoding the picture, so the pixels are byte-for-byte identical and the file gets smaller. Selective removal, batch mode and a ZIP.

Read every metadata block a file carries: EXIF from its TIFF structure, IPTC from its IIM records, XMP from its packet and the ICC colour profile from its header, with GPS converted to decimal degrees with the reference letters applied.

A block average or a centre sample, each named and measurably different, with the colour limit producing exactly that many colours and the count read back off the output.

Resize with the filter named (Lanczos, Mitchell, Catmull-Rom, bilinear, box or nearest) and the option to average LIGHT rather than stored bytes, which is worth 1.24 stops on a detailed picture.

Quarter turns and mirrors as exact permutations, checked by decoding the output and comparing it with what went in. A free angle is marked as the resample it is, and the EXIF Orientation tag is read and named.

Read text off a picture with Tesseract, after preparation that measures the type and enlarges it: 8px text went from a 53 percent character error rate to 8. Every word carries a confidence and a box drawn on the picture.

Convert images to ASCII text art with the ramp measured rather than assumed: each glyph is drawn and its ink counted, the character cell is measured in the font used, and brightness is a named definition.

Turn an image into a data URL whose payload is decoded again and compared byte for byte with the file, with the media type taken from the file's own signature rather than its extension and the size cost of embedding stated in full.

Turn one image into a full icon set, including a real multi-size favicon.ico read back from its own bytes before it is offered. Names the resampling filter, states what a square crop discarded, and decodes every PNG again to confirm its size.

Trace a PNG or JPG into real SVG paths, with holes that stay holes and transparency that is not read as black. The page counts the pixels its own vector differs from the picture, and at the exact setting that count is zero.

Enlarge with the filter named and cited, and the alignment measured: six published kernels matching Pillow to within half a level, premultiplied alpha, and an unsharp mask whose colour space is a choice.

Text, image and tiled watermarks whose padding is measured to the ink, whose rotation keeps the mark on the picture, and whose opacity is source-over: the number is the share of the way from the picture to the mark's own colour.

Caption a picture with text that is measured, wrapped and kept inside the frame. The preview canvas is the exported file, pixel for pixel.

Placeholder images described once and drawn two ways, so the preview, the raster and the SVG are the same picture. The label is measured and made to fit, and the quality slider only appears where quality does something.

Scan QR codes from an image or your camera, in this tab. Reads every code in a picture, rescues a faded or transparent one and says which step worked, and parses WiFi, vCard, calendar, geo and payment payloads by their published grammars.

Wrap a screenshot in a real CSS gradient, a real corner radius and a real device frame. The preview is the export drawn smaller by the same function, so what you see is what the file gets.

Pack frames into a sheet with an atlas that names each sprite's own rectangle, or slice a sheet back apart with margin and spacing read off the pixels. The page checks every frame comes back out of the sheet byte for byte.

Optimize and minify SVG files over a real markup tokenizer, then verify the result by drawing both versions and comparing every pixel.

Render SVG to PNG, JPG or WebP at a chosen size, with the size worked out from the file by the rules that define it and checked against the browser's own rendering.

Twelve procedural patterns that tile exactly, verified live with a measured wrap error and a 3 by 3 repeat view. Perlin, fractal, ridged, Voronoi, marble, wood and geometric grids.