Struct SegmentationResult 

pub struct SegmentationResult {
    pub class_map: Array2<usize>,
    pub confidence: Array3<f32>,
    pub classes: Vec<SegmentationClass>,
}

Semantic segmentation result

Fields

class_map: Array2<usize>

Class predictions per pixel (height × width)

confidence: Array3<f32>

Confidence scores per pixel per class (height × width × num_classes)

classes: Vec<SegmentationClass>

Class definitions

Implementations

impl SegmentationResult

pub fn to_color_image(&self) -> Result<RgbImage>

Convert segmentation to color visualization

pub fn get_class_confidence( &self, y: usize, x: usize, class_id: usize, ) -> Option<f32>

Get confidence for a specific class at a pixel

pub fn refine_with_crf( &mut self, original_img: &DynamicImage, iterations: usize, ) -> Result<()>

Apply conditional random field (CRF) post-processing for refinement

Trait Implementations

impl Clone for SegmentationResult

fn clone(&self) -> SegmentationResult

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for SegmentationResult

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

Formats the value using the given formatter.