IPSDK 0.2
IPSDK : Image Processing Software Development Kit

◆ eBorderExtensionPolicy

eBEP_Disable 

Border extension policy is disable : - it is the fastest method

  • it do not require additional memory
  • particules which will touch border on a closing operation may not be correctly eroded
  • holes which will touch border on an opening operation may not be correctly dilated
eBEP_Enable 

Border extension policy is enable : - it is the slowest method

  • it will require the allocation of a additional enlarged image
  • particules which will touch border on a closing operation will be correctly eroded
  • holes which will touch border on an opening operation will be correctly dilated

◆ eCircleIntensityType

eCIT_BrightCircles 

Only detect bright circles on dark background.

eCIT_DarkCircles 

Only detect dark circles on bright background.

eCIT_Both 

Detect all circles, dark and bright.

◆ eCylinderUnrollingIntegrationType

eCUIT_Max 

compute the maximum along thickness

eCUIT_Mean 

compute the mean along thickness

eCUIT_Min 

compute the minimum along thickness

◆ eDilateLabelMode

eDLM_Basins 

Adjacent label basins.

eDLM_Split 

Label basins with separation line.

◆ eAxis

eA_X 

indicates the x-axis

eA_Y 

indicates the y-axis

eA_Z 

indicates the z-axis

◆ ePropagationAxis

ePA_X 

propagation along the x-axis

ePA_Y 

propagation along the y-axis

ePA_Z 

propagation along the z-axis

◆ ePropagationDirection

ePD_Direct 

direct path propagation

ePD_Reverse 

reverse path propagation

◆ eLocalExtremumType

eLET_Max 

Local maxima extremum.

eLET_Min 

Local minima extremum.

eLET_StrictMax 

Local maxima extremum without plateau.

eLET_StrictMin 

Local minima extremum without plateau.

◆ eExtremumExtractType

eEET_Max 

Extremum type set to maximum.

eEET_Min 

Extremum type set to minimum.

eEET_Both 

Extremum type set to both maximum AND minimum.

◆ eFrequencyBandPassFilterType

eFBPFT_Gaussian 

Gaussian filter.

eFBPFT_LogGabor 

Log-Gabor filter.

◆ eHoughAccumThresholdType

eHATT_Absolute 

absolute threshold

eHATT_MaxRatio 

ratio of the maximum of the image of accumulators

eHATT_HistoQuantile 

quantile of the histogram of the image of accumulators

◆ eHoughCircles2dAlgoMethod

eHCAM_Gradient 

Gradient-based method.

eHCAM_GradientPhaseCoded 

Gradient-based phase-coded method.

◆ eImgComparisonToleranceType

eICTT_None 

no tolerance defined

eICTT_Absolute 

absolute tolerance

eICTT_Relative 

relative tolerance

eICTT_AbsoluteOrRelative 

absolute or relative tolerance

◆ eInterpolationPolicy

eIP_NearestNeighbour 

Nearest neighbour interpolation policy.

eIP_Linear 

Linear interpolation policy (bilinear for 2d case and trilinear for 3d case)

eIP_Cubic 

Cubic interpolation policy (bicubic for 2d case and tricubic for 3d case)

◆ eLabelReindexingPolicy

eLRP_KeepIndexes 

ask for no label reindexation (holes may appear between label indexes)

eLRP_Reindex 

ask for label reindexation (no holes between label indexes)

◆ eLawTexPostProcMethod

eLTPPM_Mean 

compute mean on a square neighbourhood

eLTPPM_MeanAbs 

compute mean of absolute values of elements on a square neighbourhood

eLTPPM_Gaussian 

gaussian smoothing on a square neighbourhood

eLTPPM_GaussianAbs 

gaussian smoothing of absolute values of elements on a square neighbourhood

eLTPPM_Variance 

variance on a square neighbourhood

◆ eLawTexPreProcMethod

eLTPPM_SubtractMeanSmoothed 

Subtract local mean smoothed.

eLTPPM_SubtractGaussianSmoothed 

Subtract local gaussian smoothed.

◆ eLightnessType

eLT_Average 

(channel0 + ... + channelN) / NbChannels

eLT_Lightness 

(max(channel0, ..., channelN) + min(channel0, ..., channelN)) / 2

eLT_Luminosity 

0.21 R + 0.72 G + 0.07 B -> The input image MUST be have 3 channels (assumed to be RGB)

◆ eMaskMode

eMM_Direct 

use directly the mask

eMM_Reverse 

use the reversed mask (process 0 as 1 and 1 as 0)

◆ eMatchStatsCriterion

eMinMax 

adjust image dynamic on min and max

eMeanStdDev 

adjust image dynamic on mean and variance

eHistogram 

adjust image dynamic on histogram

◆ eMinMaxDefinition

eMMD_ImgMinMax 

minimal and maximal intensities of the image

eMMD_UserDefinedMinMax 

minimal and maximal intensities are defined by the user

◆ ePCAReductionCriterion

ePCARC_VarianceThresholding 

variance thresholding

ePCARC_ExpectedNbEltsInSequence 

expected dimensionality of output sequence image

◆ ePredefinedCyclicRgbColorMapId

ePCRCMI_Random1_12Colors 

Random1 (12 colors)

ePCRCMI_Random1_12Colors_FirstBlack 

Random1 (12 colors), first color is black.

ePCRCMI_Random2_12Colors 

Random2 (12 colors)

ePCRCMI_Random2_12Colors_FirstBlack 

Random2 (12 colors), first color is black.

ePCRCMI_Random1_24Colors 

Random1 (24 colors)

ePCRCMI_Random1_24Colors_FirstBlack 

Random1 (24 colors), first color is black.

◆ ePredefinedSmoothRgbColorMapId

ePSRCMI_Jet 

JET color map.

ePSRCMI_InvJet 

inverted JET color map

ePSRCMI_Viridis 

Viridis color map (matplotlib-like)

ePSRCMI_Magma 

Magma color map (matplotlib-like)

ePSRCMI_Inferno 

Inferno color map (matplotlib-like)

ePSRCMI_Plasma 

Plasma color map (matplotlib-like)

◆ eProcessingOptimizationPolicy

ePOP_MaximizeSpeed 

Processing will favor speed execution.

ePOP_MinimizeMemory 

Processing will favor low memory consumption.

◆ eProjStatType

ePST_Max 

compute the maximum

ePST_Mean 

compute the mean

ePST_Median 

compute the median

ePST_Min 

compute the minimum

ePST_StdDev 

compute the standard deviation

ePST_Sum 

compute the sum

ePST_Variance 

compute the variance

◆ eRandomAlgoType

eRAT_NonRandom 

use a non random algorithm

eRAT_Random 

use a random algorithm

◆ eRandomForestFeaturesDimension

eRFFD_2d 

Compute 2D features.

eRFFD_3d 

Compute 3D features.

◆ eRandomForestFeaturesType

eRFFT_GaussianSmoothing 

Random forest feature type is gaussian smoothing.

eRFFT_MeanSmoothing 

Random forest feature type is mean smoothing.

eRFFT_Laplacian_Difference 

Random forest feature type is Laplacian.

eRFFT_High_Pass 

Random forest feature type is high pass.

eRFFT_Variance 

Random forest feature type is variance.

eRFFT_Erosion 

Random forest feature type is erosion.

eRFFT_Dilation 

Random forest feature type is dilation.

eRFFT_Opening 

Random forest feature type is opening.

eRFFT_Closing 

Random forest feature type is closing.

eRFFT_Dilatation 

Random forest feature type is dilation (kept for backward compatibility)

◆ eRgbColorMapCycleMode

eRCMCM_AllCyclic 

All the elements of the LUT take part to the cycle.

eRCMCM_ExcludeFirstFromCycle 

All the elements of the LUT take part to the cycle, except the first one.

◆ eShapeGreyscaleType

eSGT_Dark 

computation considers dark shapes

eSGT_Light 

computation considers light shapes

◆ eSkeletonAlgoType

eSAT_Default 

Default algorithm.

eSAT_Smooth 

Smooth skeleton computation.

◆ eSmoothRgbColorMapRangeMode

eSRCMRM_ImageRange 

Image range.

eSRCMRM_DataTypeRange 

Input image data type range.

◆ eSuperPixelsType

eSPT_Number 

computation considers number of super pixels

eSPT_Size 

computation considers size of super pixels

◆ eWatershedOutputMode

eWOM_Lines 

Watershed lines extraction mode for watershed algorithm.

eWOM_Basins 

Watershed basins extraction mode for watershed algorithm.

◆ eWatershedSeparationMode

eWSM_Split 

Split of input image mode for watershed algorithm.

eWSM_Lines 

Watershed lines extraction mode for watershed algorithm.

eWSM_Basins 

Watershed basins extraction mode for watershed algorithm.

eWSM_SplitLabel 

Split of input image mode for watershed algorithm.

◆ eWatershedProcessingMode

eWPM_OptimizeSpeed 

Parallelised mode for watershed algorithm.

eWPM_Repeatable 

Repeatable mode for watershed algorithm.

◆ eZoomInterpolationMethod

eZIM_NearestNeighbour 

Nearest neighbour interpolation method.

eZIM_Linear 

Linear interpolation method (bilinear for 2d case and trilinear for 3d case)

eZIM_Cubic 

Cubic interpolation method (bicubic for 2d case and tricubic for 3d case)

eZIM_VolumeWeightedMean 

Volume weighted mean interpolation method.

◆ eCornerDetectionMethod

eCDM_ShiTomasi 

Shi-Tomasi corner detection method.

eCDM_Harris 

Harris corner detection method.

◆ eImageValueExtractionMethod

eIVEM_ImageRatio 

Ratio between number of requested samples R and total number of image pixels T defined used method : tiled method if R < ImageRatioThreshold * T, striped method otherwise.

eIVEM_NbSample 

Number of requested samples R defined used method : tiled method if R < NbDataThreshold, striped method otherwise.

eIVEM_ForceTiled 

Force used method to tiled one.

eIVEM_ForceStriped 

Force used method to striped one.

◆ eKDEBandwidthMethod

eKDEBM_ScottsRule 

Scott's rule is used : $h=n^{\frac{-1}{d+4}}$ [D.W. Scott, "Multivariate Density Estimation: Theory, Practice, and Visualization", John Wiley & Sons, New York, Chicester, 1992.].

eKDEBM_SilvermansRule 

Silverman's rule is used : $h=\left(n\times\frac{d + 2}{4}\right )^{\frac{-1}{d + 4}}$ [B.W. Silverman, "Density Estimation for Statistics and Data Analysis", Vol. 26, Monographs on Statistics and Applied Probability, Chapman and Hall, London, 1986.].

eKDEBM_GaussianUnivariate 

Optimal bandwidth parameter under gaussian basis function used to approximate univariate data : $h=\left(\frac{4\hat{\sigma}^5}{3n}\right)^\frac{1}{5}\approx 1.06\hat{\sigma}n^\frac{-1}{5}$ [Silverman, B.W. (1998). Density Estimation for Statistics and Data Analysis. London: Chapman & Hall/CRC. p. 48. ISBN 0-412-24620-1.].

eKDEBM_UserDefined 

Bandwidth parameter is a user defined parameter.

◆ eRadialBasisKernelParamsPolicy

eRBKPP_FromDataDynamic 

Parameters value are defined as a ratio of input data dynamic.

eRBKPP_UserValue 

Parameters are a user provided value.

◆ eRegistrationMotionModel2d

eRMM2d_Rigid 

2d registration motion model associated to a rigid transformation (also known as Euclidean transformation)

eRMM2d_Similarity 

2d registration motion model associated to a similarity transformation

eRMM2d_Homography 

2d registration motion model associated to a homography transformation

eRMM2d_Translation 

2d registration motion model associated to a translation transformation

◆ eRegistrationMotionModel3d

eRMM3d_Rigid 

3d registration motion model associated to a rigid transformation (also known as Euclidean transformation)

eRMM3d_Similarity 

3d registration motion model associated to a similarity transformation

eRMM3d_Homography 

3d registration motion model associated to a homography transformation

eRMM3d_Translation 

3d registration motion model associated to a translation transformation