apply_contour_mask()

MAYAFLUX_API void MayaFlux::Kinesis::Vision::apply_contour_mask	(	std::span< float >	pixels,
		uint32_t	w,
		uint32_t	h,
		uint32_t	channels,
		const Contour &	contour,
		float	origin_x,
		float	origin_y,
		float	scale_x,
		float	scale_y
	)

Zero all pixels in pixels that fall outside contour.

For each pixel in the buffer, reconstructs its normalised image-space position from its row/column index, the crop origin (origin_x, origin_y), and the pixel dimensions. Tests containment using the winding number algorithm against the contour's point polygon. Pixels outside the polygon have all channel values set to 0.

The buffer is modified in place. Pixels inside the polygon are untouched.

Parameters

pixels	Interleaved float buffer, size must be w * h * channels.
w	Buffer width in pixels.
h	Buffer height in pixels.
channels	Channels per pixel.
contour	Contour whose polygon defines the keep region.
origin_x	Normalised x coordinate of the buffer's left edge.
origin_y	Normalised y coordinate of the buffer's top edge.
scale_x	Normalised width covered by one pixel column.
scale_y	Normalised height covered by one pixel row.

Definition at line 211 of file Contours.cpp.

{
    if (pixels.empty() || contour.points.empty() || channels == 0)
        return;
 
    for (uint32_t row = 0; row < h; ++row) {
        const float py = origin_y + (static_cast<float>(row) + 0.5F) * scale_y;
        for (uint32_t col = 0; col < w; ++col) {
            const float px = origin_x + (static_cast<float>(col) + 0.5F) * scale_x;
            if (!point_in_contour(contour.points, px, py)) {
                const size_t base = (static_cast<size_t>(row) * w + col) * channels;
                for (uint32_t c = 0; c < channels; ++c)
                    pixels[base + c] = 0.0F;
            }
        }
    }
}

References h, pixels, and MayaFlux::Kinesis::Vision::Contour::points.

Referenced by MayaFlux::Yantra::VisionExtractor< InputType, OutputType >::extract_implementation().

Here is the caller graph for this function: