Enum imageproc::template_matching::MatchTemplateMethod

pub enum MatchTemplateMethod {
    SumOfSquaredErrors,
    SumOfSquaredErrorsNormalized,
    CrossCorrelation,
    CrossCorrelationNormalized,
}

Scoring functions when comparing a template and an image region.

Variants

SumOfSquaredErrors

Sum of the squares of the difference between image and template pixel intensities. Smaller values indicate a better match.

Without a mask: $$ \text{output}(x, y) = \sum_{x’, y’} \left( \text{template}(x’, y’) - \text{image}(x+x’, y+y’) \right)^2 $$

With a mask: $$ \text{output}(x, y) = \sum_{x’, y’} \left( (\text{template}(x’, y’) - \text{image}(x+x’, y+y’)) \cdot \text{mask}(x’, y’) \right)^2 $$

SumOfSquaredErrorsNormalized

Divides the sum computed using SumOfSquaredErrors by a normalization term. Smaller values indicate a better match.

Without a mask: $$ \text{output}(x, y) = \frac{\sum_{x’, y’} \left( \text{template}(x’, y’) - \text{image}(x+x’, y+y’) \right)^2} {\sqrt{ \sum_{x’, y’} {\text{template}(x’, y’)}^2 \cdot \sum_{x’, y’} {\text{image}(x+x’, y+y’)}^2 }} $$

With a mask: $$ \text{output}(x, y) = \frac{\sum_{x’, y’} \left( (\text{template}(x’, y’) - \text{image}(x+x’, y+y’)) \cdot \text{mask}(x’, y’) \right)^2} {\sqrt{ \sum_{x’, y’}{(\text{template}(x’, y’) \cdot \text{mask}(x’, y’))}^2 \cdot \sum_{x’, y’}{(\text{image}(x+x’, y+y’) \cdot \text{mask}(x’, y’))}^2 }} $$

CrossCorrelation

Cross Correlation. Larger values indicate a better match.

Without a mask: $$ \text{output}(x, y) = \sum_{x’, y’} \left( \text{template}(x’, y’) \cdot \text{image}(x+x’, y+y’) \right) $$

With a mask: $$ \text{output}(x, y) = \sum_{x’, y’} \left( \text{template}(x’, y’) \cdot \text{image}(x+x’, y+y’) \cdot {\text{mask}(x’, y’)}^2 \right) $$

CrossCorrelationNormalized

Divides the sum computed using CrossCorrelation by a normalization term. Larger values indicate a better match.

Without a mask: $$ \text{output}(x, y) = \frac{\sum_{x’, y’} \left( \text{template}(x’, y’) \cdot \text{image}(x+x’, y+y’) \right)} {\sqrt{ \sum_{x’, y’} {\text{template}(x’, y’)}^2 \cdot \sum_{x’, y’} {\text{image}(x+x’, y+y’)}^2 }} $$

With a mask: $$ \text{output}(x, y) = \frac{\sum_{x’, y’} \left( \text{template}(x’, y’) \cdot \text{image}(x+x’, y+y’) \cdot {\text{mask}(x’, y’)}^2 \right)} {\sqrt{ \sum_{x’, y’}{(\text{template}(x’, y’) \cdot \text{mask}(x’, y’))}^2 \cdot \sum_{x’, y’}{(\text{image}(x+x’, y+y’) \cdot \text{mask}(x’, y’))}^2 }} $$

Trait Implementations

impl Clone for MatchTemplateMethod

fn clone(&self) -> MatchTemplateMethod

Returns a copy of the value.

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

Performs copy-assignment from source.

impl Debug for MatchTemplateMethod

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

Formats the value using the given formatter.

impl PartialEq for MatchTemplateMethod

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

This method tests for self and other values to be equal, and is used by ==.

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

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

impl Copy for MatchTemplateMethod

impl Eq for MatchTemplateMethod

impl StructuralPartialEq for MatchTemplateMethod