image-add — Traditional AI Model
Adds two same-size images channel by channel, clipping at 255, so overlapping bright areas get brighter — a lighten blend or double exposure, and a way to union two masks. Sizes must match. CPU, instant.
Details
- CategoryTraditional
- Year-1
- License
Inputs & Outputs
- ImageInputimage
First source image
- ImageInputimage
Second source image to add
- ImageOutputimage
Blended image (pixel addition)
Parameters
- fitstringdefault coveroptions: cover, contain, stretch
How to fit the differently-sized input to the reference resolution.
When the two inputs differ in resolution, how the other input is fit to the reference: 'cover' (default) scales to fill and center-crops without distorting aspect, 'contain' scales to fit and pads, 'stretch' resizes to the exact size (may distort). Both inputs are aligned to one size before the pixel operation.
- referencestringdefault port0options: port0, largest, smallestadvanced
Which input's resolution to match.
When the two images differ in size, whose resolution the result matches: 'port0' (default) the first input (controlled by wiring order), 'largest' the larger by area, 'smallest' the smaller. The other input is fit (see 'fit') to that size before the pixel operation.
Tags
Alternatives in Image Blending
- Image Multiply
Multiply two images pixel-wise (cutout / multiply blend).
- Image Paste by Class
Paste a processed crop back where Image Crop by Class took it.
- Image Paste by Coordinates
Paste a processed crop back at (Left, Top, Right, Bottom) pixel coordinates.
- Image Paste Each by Coordinates
Paste a whole list of processed crops back into the original image.