Enum OrientationMethod 

#[non_exhaustive]
pub enum OrientationMethod {
    RingFit,
    DiskFit,
}

Method used to fit the two grid axes at a detected corner.

The default Self::RingFit covers the vast majority of use cases. Use Self::DiskFit when you need improved axis accuracy on corners with strong projective skew (axis separation far from 90°).

All variants emit axes[0] and axes[1] under the same canonical convention documented on crate::detect::CornerDescriptor: axes[0].angle ∈ [0, π), axes[1].angle ∈ (axes[0].angle, axes[0].angle + π), and the CCW arc (axes[0], axes[1]) is a dark sector of the corner. Downstream consumers can therefore compare axes[0] slot parity between corners regardless of which method produced them.

Variants (Non-exhaustive)

Non-exhaustive enums could have additional variants added in future. Therefore, when matching against variants of non-exhaustive enums, an extra wildcard arm must be added to account for any future variants.

RingFit

Fit the parametric two-axis chessboard intensity model I(φ) = μ + A·tanh(β·sin(φ−θ₁))·tanh(β·sin(φ−θ₂)) to 16 ring samples via Gauss-Newton seeded from the 2nd-harmonic orientation. Suspicious local minima are retried from a small deterministic seed grid over the same 16 samples. When a radius-10 descriptor ring is requested and still looks suspicious, the canonical radius-5 ring is sampled as a cheap safety check and used when it produces a valid contrast/residual fit. Per-axis 1σ uncertainties are calibrated by a piecewise-linear lookup table keyed on the contrast-relative residual, bringing reported sigmas closer to the empirical RMSE.

This is the default method. Suitable for the full range of standard chessboard images.

DiskFit

Full-disk crossing-line estimator. Samples all image pixels in a disk around the corner center and fits two possibly non-orthogonal axes from the resulting gradient field. When local evidence is weak or the ring fit already indicates a clean orthogonal corner, falls back to Self::RingFit output transparently.

Use this when standard chessboards are imaged under strong projective warp (axis separation far from 90°). Higher per-corner cost than RingFit (~5–10× on typical hardware), but the lazy-gate short-circuits on clean inputs so the average cost is much lower.

Output axes use the same canonical convention as Self::RingFit — see the type-level doc comment above and crate::detect::CornerDescriptor.

Trait Implementations

impl Clone for OrientationMethod

fn clone(&self) -> OrientationMethod

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for OrientationMethod

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for OrientationMethod

fn default() -> OrientationMethod

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for OrientationMethod

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl PartialEq for OrientationMethod

fn eq(&self, other: &OrientationMethod) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Serialize for OrientationMethod

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl Copy for OrientationMethod

impl StructuralPartialEq for OrientationMethod