AutocorrelationFilter

Computes the cross-correlation of an image by itself.

Access to parameter description

For an introduction: see section Image Correlation.

This algorithm consists in performing a convolution of an image with itself. In 2D, the autocorrelation value, for a pixel at $(n,m)$ coordinates, is given by: $$ O(n,m)=\sum_{i=1}^{N} \sum_{j=1}^{M} I(i,j)\times I(n+i-\frac{N}{2},m+j-\frac{M}{2}) $$ where $N$ is the image height and $M$ its width in pixels.
The output score is real-valued and normalized to have a maximum of 1.
The meanNormalization option activates a mode where the image average intensity is subtracted from the input intensities before computing autocorrelation scores. When this option is enabled, the minimum autocorrelation score is generally negative. Else, if the input image has an unsigned data type, the minimum autocorrelation score is positive.

Contrary to the CrossCorrelation2d algorithm, which is more dedicated to correlate a small pattern on a larger image, this algorithm processes all image pixels and does not set uncalculated values in the output image.

See also

• CrossCorrelation2d
• VarianceFilter2d

// Function prototype.
std::shared_ptr< iolink::ImageView >
autocorrelationFilter( std::shared_ptr< iolink::ImageView > inputImage,
                       bool meanNormalization,
                       std::shared_ptr< iolink::ImageView > outputImage = NULL );

// Function prototype.
autocorrelation_filter( input_image, mean_normalization = True, output_image = None )

// Function prototype.
public static IOLink.ImageView
AutocorrelationFilter( IOLink.ImageView inputImage,
                       bool meanNormalization = true,
                       IOLink.ImageView outputImage = null );

// Command constructor.
AutocorrelationFilter();

// Get method of the inputImage parameter.
std::shared_ptr< iolink::ImageView > inputImage() const;
// Set method of the inputImage parameter.
void setInputImage( std::shared_ptr< iolink::ImageView > inputImage );

// Get method of the meanNormalization parameter.
bool meanNormalization() const;
// Set method of the meanNormalization parameter.
void setMeanNormalization( const bool& meanNormalization );

// Get method of the outputImage parameter.
std::shared_ptr< iolink::ImageView > outputImage() const;
// Set method of the outputImage parameter.
void setOutputImage( std::shared_ptr< iolink::ImageView > outputImage );

// Method to launch the command.
void execute();

// Property of the inputImage parameter.
AutocorrelationFilter.input_image
// Property of the meanNormalization parameter.
AutocorrelationFilter.mean_normalization
// Property of the outputImage parameter.
AutocorrelationFilter.output_image

// Method to launch the command.
execute()

// Command constructor.
AutocorrelationFilter()

// Property of the inputImage parameter.
AutocorrelationFilter.inputImage
// Property of the meanNormalization parameter.
AutocorrelationFilter.meanNormalization
// Property of the outputImage parameter.
AutocorrelationFilter.outputImage

// Method to launch the command.
Execute()

std::shared_ptr< iolink::ImageView > polystyrene = ioformat::readImage( std::string( IMAGEDEVDATA_IMAGES_FOLDER ) + "polystyrene.tif" );

AutocorrelationFilter autocorrelationFilterAlgo;
autocorrelationFilterAlgo.setInputImage( polystyrene );
autocorrelationFilterAlgo.setMeanNormalization( true );
autocorrelationFilterAlgo.execute();

std::cout << "outputImage:" << autocorrelationFilterAlgo.outputImage()->toString();

polystyrene = ioformat.read_image(imagedev_data.get_image_path("polystyrene.tif"))

autocorrelation_filter_algo = imagedev.AutocorrelationFilter()
autocorrelation_filter_algo.input_image = polystyrene
autocorrelation_filter_algo.mean_normalization = True
autocorrelation_filter_algo.execute()

print( "output_image:", str( autocorrelation_filter_algo.output_image ) );

ImageView polystyrene = ViewIO.ReadImage( @"Data/images/polystyrene.tif" );

AutocorrelationFilter autocorrelationFilterAlgo = new AutocorrelationFilter
{
    inputImage = polystyrene,
    meanNormalization = true
};
autocorrelationFilterAlgo.Execute();

Console.WriteLine( "outputImage:" + autocorrelationFilterAlgo.outputImage.ToString() );

std::shared_ptr< iolink::ImageView > polystyrene = ioformat::readImage( std::string( IMAGEDEVDATA_IMAGES_FOLDER ) + "polystyrene.tif" );

auto result = autocorrelationFilter( polystyrene, true );

std::cout << "outputImage:" << result->toString();

polystyrene = ioformat.read_image(imagedev_data.get_image_path("polystyrene.tif"))

result = imagedev.autocorrelation_filter( polystyrene, True )

print( "output_image:", str( result ) );

ImageView polystyrene = ViewIO.ReadImage( @"Data/images/polystyrene.tif" );

IOLink.ImageView result = Processing.AutocorrelationFilter( polystyrene, true );

Console.WriteLine( "outputImage:" + result.ToString() );