Apply Image Morphology (OpenCV)

@opencv1 VersionTransform

Applies an OpenCV morphological operation to an Image with a chosen structuring element, emitting a downstream-resolved oneof t[Image, Mask] — the filtered frame in its input color space, or a re-binarised foreground Mask.

How it fits

Typical backends

Clean up a binarised image with an opening before detection, publishing the result over HTTP.

Convert a file-loaded image to a consistent format, then apply a morphological gradient to highlight edges before publishing it.

Caveats

I/O contractThe output arm is fixed by the downstream connection: the Image arm runs the operation in the input color space, while the Mask arm first converts to grayscale, runs the operation, and re-thresholds to a boolean foreground mask.
Parameter interactionop semantics depend on the input being foreground-bright: op.tophat, op.blackhat and op.gradient are difference operators whose meaning flips if the polarity of the input is inverted.
Limitationkernel_size dimensions and iterations are clamped to a minimum of one, so a zero is silently treated as a single-pixel element or a single pass rather than a no-op.
State lifetimeThe structuring element is built once at startup from kernel_shape and kernel_size and reused for every frame, so changing them requires a redeploy.

Versions

4a721936latestdefault
Automated release
6/25/2026