ImageMoment

void avl::ImageMoment
(
	const avl::Image& inImage,
	atl::Optional<const avl::Region&> inRoi,
	avl::ImageMomentType::Type inMomentType,
	bool inCentral,
	float& outMoment,
	atl::Optional<float&> outNormMoment = atl::NIL
)

void ImageMoment
(
	Image inImage,
	Region? inRoi,
	ImageMomentType inMomentType,
	bool inCentral,
	out float outMoment,
	out float? outNormMoment
)

The operation computes the mathematical features of an image called moments. Those are sums computed as follows:

\[ \begin{aligned} Moment_{0,0}(Image,Roi) &= \sum_{p \in Roi} Image(p_x,p_y) \cdot 1 \\ Moment_{0,1}(Image,Roi) &= \sum_{p \in Roi} Image(p_x,p_y) \cdot p_y \\ Moment_{1,0}(Image,Roi) &= \sum_{p \in Roi} Image(p_x,p_y) \cdot p_x \\ Moment_{0,2}(Image,Roi) &= \sum_{p \in Roi} Image(p_x,p_y) \cdot p_y^2 \\ Moment_{1,1}(Image,Roi) &= \sum_{p \in Roi} Image(p_x,p_y) \cdot p_x p_y \\ Moment_{2,0}(Image,Roi) &= \sum_{p \in Roi} Image(p_x,p_y) \cdot p_x^2 \end{aligned} \]

The summing is conducted over all pixels in inRoi region, while \( p_x \) and \( p_y \) denote, accordingly, x and y coordinate of a pixel.

When inCentral parameter is set, the image is shifted before computations, so that its mass center is at location (0,0).

Error type	Description
DomainError	Empty region on input in ImageMoment.
DomainError	Black image on input in ImageMoment.

RegionMoment, ShapeMoment