ImageDev

HysteresisThresholding

Performs a thresholding with a conditional propagation.

Access to parameter description

For an introduction: This algorithm uses a hysteresis loop to provide a more connected threshold result.

Two gray level values $\lambda_1$ and and $\lambda_2$ are specified.
The output $O$ is given by: The Figure 1 represents an example of 1-D function. The Figure 2 shows the result of a lower bound threshold with value $\lambda_1$ and $\lambda_2$. Figure 3 is the result of a hysteresis thresholding, where the points in the fuzzy area not connected with points in the retained area are rejected.

<b> Figure 1.</b> 1-D function with low and high thresholds
Figure 1. 1-D function with low and high thresholds

<b> Figure 2.</b> Retained area (red), rejected area (blue) and fuzzy area (green)
Figure 2. Retained area (red), rejected area (blue) and fuzzy area (green)

<b> Figure 3.</b> Hysteresis thresholding, unconnected fuzzy area rejected
Figure 3. Hysteresis thresholding, unconnected fuzzy area rejected

This algorithm can be used after an edge detection, which generates, as well as edges, a lot of noise. True edges have a higher chance to be connected to a retained area than pixels corresponding to noise.
By the way, hysteresis thresholding is the last step of the Canny edge detector algorithm, following the non-maximum suppression step.

See also

Function Syntax

This function returns outputBinaryImage. 
// Function prototype
std::shared_ptr< iolink::ImageView >
hysteresisThresholding( std::shared_ptr< iolink::ImageView > inputImage,
                        iolink::Vector2d thresholdRange,
                        int32_t length,
                        std::shared_ptr< iolink::ImageView > outputBinaryImage = NULL );
This function returns outputBinaryImage. 
// Function prototype.
hysteresis_thresholding( input_image,
                         threshold_range = [255, 128],
                         length = 1,
                         output_binary_image = None )
This function returns outputBinaryImage. 
// Function prototype.
public static IOLink.ImageView
HysteresisThresholding( IOLink.ImageView inputImage,
                        double[] thresholdRange = null,
                        Int32 length = 1,
                        IOLink.ImageView outputBinaryImage = null );

Class Syntax

Parameters

Parameter Name Description Type Supported Values Default Value
input
inputImage
The input image. Image Grayscale nullptr
input
thresholdRange
The low and high threshold levels. Vector2d Any value {255.f, 128.f}
input
length
The maximum length in pixels allowed for considering points in the fuzzy zone (0:until convergence). Int32 >=0 1
output
outputBinaryImage
The output binary image. Its dimensions are forced to the same values as the input. Image nullptr
Parameter Name Description Type Supported Values Default Value
input
input_image
The input image. image Grayscale None
input
threshold_range
The low and high threshold levels. vector2d Any value [255, 128]
input
length
The maximum length in pixels allowed for considering points in the fuzzy zone (0:until convergence). int32 >=0 1
output
output_binary_image
The output binary image. Its dimensions are forced to the same values as the input. image None
Parameter Name Description Type Supported Values Default Value
input
inputImage
The input image. Image Grayscale null
input
thresholdRange
The low and high threshold levels. Vector2d Any value {255f, 128f}
input
length
The maximum length in pixels allowed for considering points in the fuzzy zone (0:until convergence). Int32 >=0 1
output
outputBinaryImage
The output binary image. Its dimensions are forced to the same values as the input. Image null

Object Examples

auto foam = readVipImage( std::string( IMAGEDEVDATA_IMAGES_FOLDER ) + "foam.vip" );

HysteresisThresholding hysteresisThresholdingAlgo;
hysteresisThresholdingAlgo.setInputImage( foam );
hysteresisThresholdingAlgo.setThresholdRange( {128.0, 255.0} );
hysteresisThresholdingAlgo.setLength( 1 );
hysteresisThresholdingAlgo.execute();

std::cout << "outputBinaryImage:" << hysteresisThresholdingAlgo.outputBinaryImage()->toString();

foam = imagedev.read_vip_image(imagedev_data.get_image_path("foam.vip"))

hysteresis_thresholding_algo = imagedev.HysteresisThresholding()
hysteresis_thresholding_algo.input_image = foam
hysteresis_thresholding_algo.threshold_range = [128.0, 255.0]
hysteresis_thresholding_algo.length = 1
hysteresis_thresholding_algo.execute()

print( "output_binary_image:", str( hysteresis_thresholding_algo.output_binary_image ) )

ImageView foam = Data.ReadVipImage( @"Data/images/foam.vip" );

HysteresisThresholding hysteresisThresholdingAlgo = new HysteresisThresholding
{
    inputImage = foam,
    thresholdRange = new double[]{128.0, 255.0},
    length = 1
};
hysteresisThresholdingAlgo.Execute();

Console.WriteLine( "outputBinaryImage:" + hysteresisThresholdingAlgo.outputBinaryImage.ToString() );

Function Examples

auto foam = readVipImage( std::string( IMAGEDEVDATA_IMAGES_FOLDER ) + "foam.vip" );

auto result = hysteresisThresholding( foam, {128.0, 255.0}, 1 );

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

foam = imagedev.read_vip_image(imagedev_data.get_image_path("foam.vip"))

result = imagedev.hysteresis_thresholding( foam, [128.0, 255.0], 1 )

print( "output_binary_image:", str( result ) )

ImageView foam = Data.ReadVipImage( @"Data/images/foam.vip" );

IOLink.ImageView result = Processing.HysteresisThresholding( foam, new double[]{128.0, 255.0}, 1 );

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