extract_scalar_vision()

MAYAFLUX_API double MayaFlux::Yantra::extract_scalar_vision	(	const VisionAnalysis &	analysis,
		const std::string &	qualifier
	)

Extract a named scalar from a VisionAnalysis result.

Supported qualifiers: "box_count" number of bounding boxes "contour_count" number of contours "keypoint_count" number of keypoints "track_count" number of tracked points "component_count" number of connected components "mean_gradient_magnitude" mean of GradientResult::magnitude

Unknown or empty qualifier returns 0.0.

Parameters

analysis	Result produced by VisionAnalyzer.
qualifier	Name of the scalar to extract.

Definition at line 6 of file VisionAnalyzer.cpp.

{
    const auto& f = analysis.frame;
 
    if (qualifier == "box_count")
        return static_cast<double>(f.boxes.size());
    if (qualifier == "contour_count")
        return static_cast<double>(f.contours.size());
    if (qualifier == "keypoint_count")
        return static_cast<double>(f.keypoints.size());
    if (qualifier == "track_count")
        return static_cast<double>(f.tracks.size());
    if (qualifier == "component_count")
        return static_cast<double>(f.components.count);
    if (qualifier == "mean_gradient_magnitude") {
        const auto& mag = f.gradient.magnitude;
        if (mag.empty())
            return 0.0;
        return std::accumulate(mag.begin(), mag.end(), 0.0)
            / static_cast<double>(mag.size());
    }
 
    return 0.0;
}

References MayaFlux::Yantra::VisionAnalysis::frame, MayaFlux::Yantra::DetectionSummary::gradient, and MayaFlux::Kinesis::Vision::GradientResult::magnitude.