ImageWatermarker<T>
Models & Types · AiDotNet.Safety.Watermarking
The safety pipeline’s image watermarker: the keyed spread-spectrum DCT watermark of
InvisibleImageWatermarker with a configurable reporting threshold.
Inheritance: ModelBase<T, Vector<T>, Vector<T>> → SafetyModuleBase<T> → ImageSafetyModuleBase<T> → ImageWatermarkerBase<T> → InvisibleImageWatermarker<T> → ImageWatermarker<T>
For Beginners
image watermarking hides an invisible, password-protected “stamp” inside a picture, so you can later prove the picture came from your system. You cannot see the stamp, but the watermarker that made it can find it again, even after mild noise or re-saving with 8-bit precision.
How It Works
EmbedWatermark marks images the application produces;
EvaluateImage reports a finding when an image carries the mark of this
watermarker’s key with at least the configured confidence. A watermarker only recognises marks made with its own
key, so the pipeline detects the application’s own watermark, not other vendors’ (no single detector can).
Constructors
| Constructor | Summary |
|---|---|
ImageWatermarker(double detectionThreshold, double watermarkStrength, string key) |
Initializes a new image watermarker. |
Properties
| Property | Summary |
|---|---|
string ModuleName { get; } |