Batch resize, convert and compress. 100% in your browser – files never leave your device.
Each image gets the highest quality that fits the limit; dimensions shrink only if needed. For PNG/BMP/TIFF/ICO only dimensions can shrink.
With lock on, the height follows each image, so mixed batches stay undistorted. Pixel dimensions decide how much detail a file holds; DPI is only a print instruction stored in the file.
Uses smooth multi-step enlargement, then sharpening (set to 40% automatically; adjust it under Transform & adjust). Browser upscaling cannot invent real detail, so very large factors look soft. Output is capped at about 16,000 px / 120 megapixels.
AI upscaling predicts plausible detail – it can invent or change fine text, digits and scale marks. Never treat AI-upscaled nameplates, gauges or drawings as evidence of real values. Input is limited to 1.5 MP for 2× and 0.4 MP for 4×. When AI is selected, the factor/width above is ignored.
The crop editor (Advanced level) fills these crop boxes for you. Crop and auto-trim run first, so every size mode works on the trimmed picture. Auto-trim removes borders matching the top-left pixel colour (good for product shots). Padding adds a margin inside the final frame. Rounded corners are transparent in PNG/WebP/AVIF/ICO and use the background colour elsewhere. DPI is written into the file for print workflows; it does not change pixel count.
Tokens: {name} {w} {h} {n} {n3} (001, 002…) {date} {fmt}. ICO is capped at 256 px. Every output is checked after encoding; if your browser produces a different format, nothing is saved. PDF holds one page per image. Formats your browser cannot encode are greyed out.
Brightness/contrast/saturation/grayscale/sepia/blur need a browser with canvas filters (Chrome, Edge, Firefox). Metadata is removed unless you choose to keep it under Output format.
Position, size, opacity and angle apply to both the text and the logo. Tiled text uses the angle (or −30° when the angle is 0).
Pixels vs DPI: pixel dimensions are the real amount of detail. DPI/PPI is a separate label telling printers how many pixels to place per inch – changing it never adds detail. For print, aim for about 300 DPI at the final size (150–200 is acceptable for posters viewed from a distance).
Colour: the browser converts every picture to 8-bit sRGB RGB for processing, and outputs untagged sRGB. CMYK and 16-bit output are not supported; CMYK or wide-gamut (Adobe RGB, Display P3) inputs are converted to sRGB, so colours can shift slightly. Original ICC profiles are reported in the inspector but not re-attached, because the pixels are already converted. Transparency is kept for PNG, WebP, AVIF and ICO; JPG, BMP, TIFF and PDF use the background colour.
Drag on the picture to draw a crop, drag inside it to move, drag a corner to resize. The box fills the crop % fields and applies to every image in the batch at the same relative position. Smart crop is a content-detection heuristic (edges and colour contrast), not AI. Perspective: drag the four dots onto the corners of a nameplate, panel or drawing photographed at an angle, then press APPLY PERSPECTIVE.
Diagnostics are automatic heuristic indicators measured on a reduced copy, not calibrated measurements.