EigenvaluesToStructureness3d
Computes a structure score image from the eigenvalues of a three-dimensional tensor image.
Access to parameter description
This algorithm allows extracting tubular, blob-like, and plate-like structures from dark or bright background.
The computation is based on eigenvalues $\lambda_{1}$, $\lambda_{2}$ and $\lambda_{3}$ where $\mid \lambda_{1} \mid \geq \mid \lambda_{2} \mid \geq \mid \lambda_{3} \mid $.
The structure score for bright tubular structures corresponds to:
$ score = \begin{cases} 0 & \text{ if } \lambda_1 >0 \text{ or } \lambda_2 >0 \\ (1-\exp(-\frac{\mathcal{R}_a^2}{2\alpha^2}))\exp(-\frac{\mathcal{R}_b^2}{2\beta^2})(1-\exp(-\frac{S^2}{2c^2})) & \end{cases} $
The score for bright blob-like structures is computed as follows:
$ score = \begin{cases} 0 & \text{ if } \lambda_1 >0 \text{ or } \lambda_2 >0 \text{ or } \lambda_3 >0\\ (1-\exp(-\frac{\mathcal{R}_a^2}{2\alpha^2}))(1-\exp(-\frac{\mathcal{R}_b^2}{2\beta^2})) (1-\exp(-\frac{S^2}{2c^2})) \end{cases} $
The score for bright plate-like structures is computed as follows:
$ score = \begin{cases} 0 & \text{ if } \lambda_1 >0 \\ \exp(-\frac{\mathcal{R}_a^2}{2\alpha^2})\exp(-\frac{\mathcal{R}_b^2}{2\beta^2})(1-\exp(-\frac{S^2}{2c^2})) & \end{cases} $
Where:
This algorithm outputs a float grayscale image where each pixel intensity represents a structure score between 0 and 1.
The method for tubular structures is referenced by Alejandro Frangi publication:
A.F.Frangi, W.J.Niessen, K.L.Vincken, M.A.Viergever. "Multiscale vessel enhancement filtering". Lecture Notes in Computer Science(MICCAI), vol. 1496, pp. 130-137, 1998.
See also
Access to parameter description
This algorithm allows extracting tubular, blob-like, and plate-like structures from dark or bright background.
The computation is based on eigenvalues $\lambda_{1}$, $\lambda_{2}$ and $\lambda_{3}$ where $\mid \lambda_{1} \mid \geq \mid \lambda_{2} \mid \geq \mid \lambda_{3} \mid $.
The structure score for bright tubular structures corresponds to:
$ score = \begin{cases} 0 & \text{ if } \lambda_1 >0 \text{ or } \lambda_2 >0 \\ (1-\exp(-\frac{\mathcal{R}_a^2}{2\alpha^2}))\exp(-\frac{\mathcal{R}_b^2}{2\beta^2})(1-\exp(-\frac{S^2}{2c^2})) & \end{cases} $
The score for bright blob-like structures is computed as follows:
$ score = \begin{cases} 0 & \text{ if } \lambda_1 >0 \text{ or } \lambda_2 >0 \text{ or } \lambda_3 >0\\ (1-\exp(-\frac{\mathcal{R}_a^2}{2\alpha^2}))(1-\exp(-\frac{\mathcal{R}_b^2}{2\beta^2})) (1-\exp(-\frac{S^2}{2c^2})) \end{cases} $
The score for bright plate-like structures is computed as follows:
$ score = \begin{cases} 0 & \text{ if } \lambda_1 >0 \\ \exp(-\frac{\mathcal{R}_a^2}{2\alpha^2})\exp(-\frac{\mathcal{R}_b^2}{2\beta^2})(1-\exp(-\frac{S^2}{2c^2})) & \end{cases} $
Where:
- $S$ is the Hessian norm $ S = \left \| \mathcal{H} \right \| = \sqrt{\sum \lambda_i^2} $
- $\alpha$ is a threshold which controls the flatness sensitivity,
- $\beta$ is a threshold which controls the blobness sensitivity,
- c is a sensitivity threshold which controls the noise influence. $c = noiseCutoff \cdot S_{max}$ with $S_{max}$ the maximum Hessian norm in the image.
This algorithm outputs a float grayscale image where each pixel intensity represents a structure score between 0 and 1.
The method for tubular structures is referenced by Alejandro Frangi publication:
A.F.Frangi, W.J.Niessen, K.L.Vincken, M.A.Viergever. "Multiscale vessel enhancement filtering". Lecture Notes in Computer Science(MICCAI), vol. 1496, pp. 130-137, 1998.
See also
Function Syntax
This function returns outputImage.
// Function prototype
std::shared_ptr< iolink::ImageView > eigenvaluesToStructureness3d( std::shared_ptr< iolink::ImageView > inputEigenvaluesImage, EigenvaluesToStructureness3d::Lightness lightness, EigenvaluesToStructureness3d::StructureType structureType, double alpha, double beta, double noiseCutoff, std::shared_ptr< iolink::ImageView > outputImage = nullptr );
This function returns outputImage.
// Function prototype.
eigenvalues_to_structureness_3d(input_eigenvalues_image: idt.ImageType,
lightness: EigenvaluesToStructureness3d.Lightness = EigenvaluesToStructureness3d.Lightness.BRIGHT,
structure_type: EigenvaluesToStructureness3d.StructureType = EigenvaluesToStructureness3d.StructureType.ROD,
alpha: float = 0.75,
beta: float = 0.75,
noise_cutoff: float = 0.5,
output_image: idt.ImageType = None) -> idt.ImageType
This function returns outputImage.
// Function prototype.
public static IOLink.ImageView
EigenvaluesToStructureness3d( IOLink.ImageView inputEigenvaluesImage,
EigenvaluesToStructureness3d.Lightness lightness = ImageDev.EigenvaluesToStructureness3d.Lightness.BRIGHT,
EigenvaluesToStructureness3d.StructureType structureType = ImageDev.EigenvaluesToStructureness3d.StructureType.ROD,
double alpha = 0.75,
double beta = 0.75,
double noiseCutoff = 0.5,
IOLink.ImageView outputImage = null );
Class Syntax
Parameters
| Parameter Name | Description | Type | Supported Values | Default Value | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
 |
inputEigenvaluesImage |
The input image containing the eigenvalues field. Type must be float. Spectral series size is 3 (channel 0 = largest eigenvalue, channel 1 = medium eigen value, channel 2 = smallest eigenvalue). | Image | Multispectral | nullptr | ||||||
|
lightness |
The type of structure lightness to extract.
|
Enumeration | BRIGHT | |||||||
|
structureType |
The type of structure shapes to extract.
|
Enumeration | ROD | |||||||
|
alpha |
The flatness sensitivity threshold. It corresponds to the alpha term of the score equation. | Float64 | >0 | 0.75 | ||||||
|
beta |
The blobness sensitivity threshold. It corresponds to the beta term of the score equation. | Float64 | >0 | 0.75 | ||||||
|
noiseCutoff |
The noise scale factor. It is used for computing the c term of the score equation. | Float64 | >0 | 0.5 | ||||||
 |
outputImage |
The score output image. X, Y, and Z dimensions, calibration and interpretation of the output image are forced to the same values as the input. The spectral dimension is forced to 1. Type is forced to float and all values are between 0 and 1. | Image | nullptr | |||||||
| Parameter Name | Description | Type | Supported Values | Default Value | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
|
input_eigenvalues_image |
The input image containing the eigenvalues field. Type must be float. Spectral series size is 3 (channel 0 = largest eigenvalue, channel 1 = medium eigen value, channel 2 = smallest eigenvalue). | image | Multispectral | None | ||||||
|
lightness |
The type of structure lightness to extract.
|
enumeration | BRIGHT | |||||||
|
structure_type |
The type of structure shapes to extract.
|
enumeration | ROD | |||||||
|
alpha |
The flatness sensitivity threshold. It corresponds to the alpha term of the score equation. | float64 | >0 | 0.75 | ||||||
|
beta |
The blobness sensitivity threshold. It corresponds to the beta term of the score equation. | float64 | >0 | 0.75 | ||||||
|
noise_cutoff |
The noise scale factor. It is used for computing the c term of the score equation. | float64 | >0 | 0.5 | ||||||
|
output_image |
The score output image. X, Y, and Z dimensions, calibration and interpretation of the output image are forced to the same values as the input. The spectral dimension is forced to 1. Type is forced to float and all values are between 0 and 1. | image | None | |||||||
| Parameter Name | Description | Type | Supported Values | Default Value | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
|
inputEigenvaluesImage |
The input image containing the eigenvalues field. Type must be float. Spectral series size is 3 (channel 0 = largest eigenvalue, channel 1 = medium eigen value, channel 2 = smallest eigenvalue). | Image | Multispectral | null | ||||||
|
lightness |
The type of structure lightness to extract.
|
Enumeration | BRIGHT | |||||||
|
structureType |
The type of structure shapes to extract.
|
Enumeration | ROD | |||||||
|
alpha |
The flatness sensitivity threshold. It corresponds to the alpha term of the score equation. | Float64 | >0 | 0.75 | ||||||
|
beta |
The blobness sensitivity threshold. It corresponds to the beta term of the score equation. | Float64 | >0 | 0.75 | ||||||
|
noiseCutoff |
The noise scale factor. It is used for computing the c term of the score equation. | Float64 | >0 | 0.5 | ||||||
|
outputImage |
The score output image. X, Y, and Z dimensions, calibration and interpretation of the output image are forced to the same values as the input. The spectral dimension is forced to 1. Type is forced to float and all values are between 0 and 1. | Image | null | |||||||
Object Examples
auto neuron_eigenvalues = readVipImage( std::string( IMAGEDEVDATA_IMAGES_FOLDER ) + "neuron_eigenvalues.vip" ); EigenvaluesToStructureness3d eigenvaluesToStructureness3dAlgo; eigenvaluesToStructureness3dAlgo.setInputEigenvaluesImage( neuron_eigenvalues ); eigenvaluesToStructureness3dAlgo.setLightness( EigenvaluesToStructureness3d::Lightness::BRIGHT ); eigenvaluesToStructureness3dAlgo.setStructureType( EigenvaluesToStructureness3d::StructureType::ROD ); eigenvaluesToStructureness3dAlgo.setAlpha( 0.75 ); eigenvaluesToStructureness3dAlgo.setBeta( 0.75 ); eigenvaluesToStructureness3dAlgo.setNoiseCutoff( 0.5 ); eigenvaluesToStructureness3dAlgo.execute(); std::cout << "outputImage:" << eigenvaluesToStructureness3dAlgo.outputImage()->toString();
neuron_eigenvalues = imagedev.read_vip_image(imagedev_data.get_image_path("neuron_eigenvalues.vip"))
eigenvalues_to_structureness_3d_algo = imagedev.EigenvaluesToStructureness3d()
eigenvalues_to_structureness_3d_algo.input_eigenvalues_image = neuron_eigenvalues
eigenvalues_to_structureness_3d_algo.lightness = imagedev.EigenvaluesToStructureness3d.BRIGHT
eigenvalues_to_structureness_3d_algo.structure_type = imagedev.EigenvaluesToStructureness3d.ROD
eigenvalues_to_structureness_3d_algo.alpha = 0.75
eigenvalues_to_structureness_3d_algo.beta = 0.75
eigenvalues_to_structureness_3d_algo.noise_cutoff = 0.5
eigenvalues_to_structureness_3d_algo.execute()
print("output_image:", str(eigenvalues_to_structureness_3d_algo.output_image))
ImageView neuron_eigenvalues = Data.ReadVipImage( @"Data/images/neuron_eigenvalues.vip" );
EigenvaluesToStructureness3d eigenvaluesToStructureness3dAlgo = new EigenvaluesToStructureness3d
{
inputEigenvaluesImage = neuron_eigenvalues,
lightness = EigenvaluesToStructureness3d.Lightness.BRIGHT,
structureType = EigenvaluesToStructureness3d.StructureType.ROD,
alpha = 0.75,
beta = 0.75,
noiseCutoff = 0.5
};
eigenvaluesToStructureness3dAlgo.Execute();
Console.WriteLine( "outputImage:" + eigenvaluesToStructureness3dAlgo.outputImage.ToString() );
Function Examples
auto neuron_eigenvalues = readVipImage( std::string( IMAGEDEVDATA_IMAGES_FOLDER ) + "neuron_eigenvalues.vip" ); auto result = eigenvaluesToStructureness3d( neuron_eigenvalues, EigenvaluesToStructureness3d::Lightness::BRIGHT, EigenvaluesToStructureness3d::StructureType::ROD, 0.75, 0.75, 0.5 ); std::cout << "outputImage:" << result->toString();
neuron_eigenvalues = imagedev.read_vip_image(imagedev_data.get_image_path("neuron_eigenvalues.vip"))
result = imagedev.eigenvalues_to_structureness_3d(neuron_eigenvalues, imagedev.EigenvaluesToStructureness3d.BRIGHT, imagedev.EigenvaluesToStructureness3d.ROD, 0.75, 0.75, 0.5)
print("output_image:", str(result))
ImageView neuron_eigenvalues = Data.ReadVipImage( @"Data/images/neuron_eigenvalues.vip" ); IOLink.ImageView result = Processing.EigenvaluesToStructureness3d( neuron_eigenvalues, EigenvaluesToStructureness3d.Lightness.BRIGHT, EigenvaluesToStructureness3d.StructureType.ROD, 0.75, 0.75, 0.5 ); Console.WriteLine( "outputImage:" + result.ToString() );