@ALDAOperator(genericExecutionMode=ALL, level=STANDARD, allowBatchMode=false) @ALDDerivedClass public class FeatureCalculatorLBP extends FeatureCalculatorLBPJFeatureLib
The operator is based on the JFeatureLib which implements the LBPs according to
M. Heikkilae, M. Pietikaeinen, "A texture-based method for modeling the background and detecting moving objects", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 28, No. 4, pp. 657-662, 2006. and
Ojala et al, "Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 24, No. 7, pp. 971-987, 2002.
Tha basic idea of the extension is to consider not only direct
neighbors, but rather neighbors which are at a specified distance of
the center pixel. To this end a circular neighborhood around each
pixel is defined, and a specific number of sample pixels on the
border of this neighborhood is considered for code generation.
In addition, the size of the region used for generating the code
histogram for each pixel can be configured freely.
Finally, the implementation targets at a larger robustness by
introducing a baseline for the differences between the center and a
pixel along the border which decreases the algorithm's sensitivity
to small intensity changes.
As result the operator either returns a vector of the LBP code histograms of all pixels in the input image, appended row-by-row from top-left to bottom-right. Alternatively a single histogram for the complete image is returned which is extracted by setting the histogram region to the complete image and returning the histogram of the center pixel (with possibly rounded coordinates).
For more information refer to the documentation of the JFeatureLib at Github
FeatureCalculatorLBPJFeatureLib.FeatureType
baselineConstant, histBins, histoRegionSize, numberNeighbors, radius, resultType
inImg, resultObj
Constructor and Description |
---|
FeatureCalculatorLBP()
Default constructor.
|
Modifier and Type | Method and Description |
---|---|
protected JFeatureLibLBPAdapter |
getFeatureOp()
Get instance of actual feature calculator.
|
void |
validateCustom() |
getResultDataObjectInvalid, operate, setBaselineConstant, setFeatureType, setHistBins, setHistoRegionSize, setNumberNeighbors, setRadius
getResultData, setInputImage
readResolve
addOperatorExecutionProgressEventListener, addParameter, addParameter, addParameterUnconditioned, fieldContained, fireOperatorExecutionProgressEvent, getALDPortHashAccessKey, getConstructionMode, getDocumentation, getHidingMode, getInactiveParameterNames, getInInoutNames, getInInoutNames, getInNames, getInOutNames, getMissingRequiredInputs, getName, getNumParameters, getOutInoutNames, getOutNames, getParameter, getParameterDescriptor, getParameterDescriptorUnconditioned, getParameterNames, getParameterUnconditioned, getSupplementalNames, getVerbose, getVersion, handleOperatorExecutionProgressEvent, hasInOutParameters, hasParameter, isAnnotatedParameter, isConfigured, print, print, print, printInterface, printInterface, readHistory, reinitializeParameterDescriptors, removeOperatorExecutionProgressEventListener, removeParameter, runOp, runOp, runOp, setConstructionMode, setConstructionMode, setConstructionMode, setHidingMode, setName, setParameter, setParameterUnconditioned, setVerbose, toStringVerbose, unconfiguredItems, validate, validateGeneric, writeHistory, writeHistory, writeHistory
public FeatureCalculatorLBP() throws de.unihalle.informatik.Alida.exceptions.ALDOperatorException
de.unihalle.informatik.Alida.exceptions.ALDOperatorException
- Thrown in case of failure.public void validateCustom() throws de.unihalle.informatik.Alida.exceptions.ALDOperatorException
validateCustom
in class de.unihalle.informatik.Alida.operator.ALDOperator
de.unihalle.informatik.Alida.exceptions.ALDOperatorException
protected JFeatureLibLBPAdapter getFeatureOp()
getFeatureOp
in class FeatureCalculatorLBPJFeatureLib
Copyright © 2010–2020 Martin Luther University Halle-Wittenberg, Institute of Computer Science, Pattern Recognition and Bioinformatics. All rights reserved.