Trait PixelMapper

Source

pub trait PixelMapper {
    // Required methods
    fn image_to_working_pixel(&self, image_xy: [f64; 2]) -> Option<[f64; 2]>;
    fn working_to_image_pixel(&self, working_xy: [f64; 2]) -> Option<[f64; 2]>;
}

Expand description

Mapping between raw image pixels and detector working-frame pixels.

The working frame is the coordinate system used by sampling/fitting stages. For distortion-aware processing this is typically an undistorted pixel frame.

Implement this trait to plug in a custom distortion model. Both methods must be approximate inverses of each other. Return None if a point cannot be mapped (e.g. it falls outside the valid distortion domain).

Built-in implementations: CameraModel (Brown-Conrady radial-tangential) and DivisionModel (single-parameter division model).

§Example

use ringgrid::PixelMapper;

struct Identity;

impl PixelMapper for Identity {
    fn image_to_working_pixel(&self, p: [f64; 2]) -> Option<[f64; 2]> {
        Some(p)
    }
    fn working_to_image_pixel(&self, p: [f64; 2]) -> Option<[f64; 2]> {
        Some(p)
    }
}