Differential Interference Contrast (DIC) line integration along a given orientation on 2d images. More...

IPSDKIPLFILTERING_API image::ImagePtr	ipsdk::imaproc::filter::dicIntegration2dImg (const image::ImageConstPtr &pInImg, const ipReal32 inStdDev, const ipReal32 theta)
	wrapper function for Differential Interference Contrast (DIC) line integration along a given orientation on 2d images More...

IPSDKIPLFILTERING_API void	ipsdk::imaproc::filter::dicIntegration2dImg (const image::ImageConstPtr &pInImg, const ipReal32 inStdDev, const ipReal32 theta, const image::ImagePtr &pOutImg)
	wrapper function for Differential Interference Contrast (DIC) line integration along a given orientation on 2d images More...

Detailed Description

Differential Interference Contrast (DIC) line integration along a given orientation on 2d images.

This algorithm enhances objects from Differential Interference Contrast (DIC) like images. The idea is to reconstruct the objects from their gradient by accumulating the intensities along the gradient direction.

The figure below illustrates this principle:

The process is deomposed in to the following steps:

First high-pass filter
Integration along an orientation InTheta (in radians), provided as an input parameter
Second high-pass filter

The highpass filters consist in the subtraction between the image $I(\textbf{x})$ and its blurred version $I_b(\textbf{x})$ :

$I_H(\textbf{x}) = I(\textbf{x}) - I_b(\textbf{x})$

where $I_H(\textbf{x})$ is the result of the high-pass filter at position $\textbf{x}$ and $I_b(\textbf{x})$ is the image $I(\textbf{x})$ blurred by a 1D Gaussian smoothing with a standard deviation InStdDev and an orientation of InTheta.

The following figrue shows a simple example of the kind of result expected the DIC integration filtering:

If the integration direction matches the opposite of the gradient orientation, objects will appear dark instead of bright:

Note: It is recommanded to use a standard deviation higher or equal to the objects' half-size.

Here is a example of result on a real image:

In practice, the blur may alter the result because the gradient do not cancel itself along the integration line. It can be caused by a difference between the integration direction and the actual gradient direction. It can generates linear artefact in the integration direction.

This effect can be removed by oriented morphological operation or an oriented VSNR.

Finally, keep in mind that the result of the DIC filter does not allow to correctly segment the objects, because their shape is degraded, but it allows a good identification.

Attributes description

Attribute description for algorithm :

Name	ToolTip	Default Initializer
ipsdk::imaproc::attr::InImg	[Input] image for processing operation	X
ipsdk::imaproc::attr::InStdDev	[Input] standard deviation used for operation along all axis	X
ipsdk::imaproc::attr::InTheta	[Input] 2d orientation for processing operation	X
ipsdk::imaproc::attr::OutWk1RealImg	[Output] Temporary working image for algorithm	ipsdk::imaproc::duplicateInOut (_pOutWk1RealImg, _pInImg, ipsdk::image::eImageBufferType::eIBT_Real32)
ipsdk::imaproc::attr::OutRealImg	[Output] image for processing operation (data contained in image buffer are reals)	ipsdk::imaproc::duplicateInOut (_pOutRealImg, _pInImg, ipsdk::image::eImageBufferType::eIBT_Real32)

Global Rule description

Global rule description for algorithm :
ipsdk::imaproc::matchSize (_pInImg,_pOutWk1RealImg) &&
ipsdk::imaproc::matchSize (_pInImg,_pOutRealImg)

Example of Python code :

Example imports

import PyIPSDK

import PyIPSDK.IPSDKIPLFiltering as filter

Code Example

    # opening of input images
    inImg = PyIPSDK.loadTiffImageFile(inputImgPath)
    
    # Calculation
    outImg = filter.dicIntegration2dImg(inImg, stdDev, theta)

Example of C++ code :

Example informations

Associated library

IPSDKIPLFiltering

Header file

#include <IPSDKIPL/IPSDKIPLFiltering/Processor/DicIntegration2dImg/DicIntegration2dImg.h>

Code Example

        // opening input image
        ImagePtr pInImg = loadTiffImageFile(inImgPath);
        
        // compute the oriented integration
        ImagePtr pOutImg = dicIntegration2dImg(pInImg, stdDev, theta);

See also: DicIntegration2dImgLvl1; DicIntegration2dImgLvl2

Function Documentation

◆ dicIntegration2dImg() [1/2]

IPSDKIPLFILTERING_API image::ImagePtr ipsdk::imaproc::filter::dicIntegration2dImg	(	const image::ImageConstPtr &	pInImg,
		const ipReal32	inStdDev,
		const ipReal32	theta
	)

wrapper function for Differential Interference Contrast (DIC) line integration along a given orientation on 2d images

Exceptions

ipsdk::processor::IPSDKBaseProcessingException on failure

◆ dicIntegration2dImg() [2/2]

IPSDKIPLFILTERING_API void ipsdk::imaproc::filter::dicIntegration2dImg	(	const image::ImageConstPtr &	pInImg,
		const ipReal32	inStdDev,
		const ipReal32	theta,
		const image::ImagePtr &	pOutImg
	)

wrapper function for Differential Interference Contrast (DIC) line integration along a given orientation on 2d images

Exceptions

ipsdk::processor::IPSDKBaseProcessingException on failure