Pixel Classification Random Forest Algorithm

image =	pixelClassificationRFImg (inImg,model)
image =	pixelClassificationRFImg (inImg,model,inOptMemoryRatio)
outImg,outRealImg =	pixelClassificationRFWithProbabilitiesImg (inImg,model)
outImg,outRealImg =	pixelClassificationRFWithProbabilitiesImg (inImg,model,inOptMemoryRatio)

Detailed Description

Compute Random Forest pixel classification.

This algorithm allows to classify pixels on an image, by using the random forest algorithm on a trained model

It the model contains two, the output image will be binary, otherwise it will be labeled. The function will first compute all the features, then apply the random forest algorithm. If the features are already compute, they can be given in a sequence image, instead of the input image.

See Smart segmentation module for more details about the model creation.

See also

http://en.wikipedia.org/wiki/Random_forest

Example of Python code :

Example imports

import PyIPSDK
import PyIPSDK.IPSDKIPLMachineLearning as ml

Code Example

    #Opening image and model
    inImg = PyIPSDK.loadTiffImageFile(utImagePath)
    model, boolResult = PyIPSDK.readFromXmlFile(utModelPath)

    outImg = ml.pixelClassificationRFImg(inImg, model)

Example of C++ code :

Example informations

Header file

#include <IPSDKIPL/IPSDKIPLMachineLearning/Processor/PixelClassificationRFImg/PixelClassificationRFImg.h>

Code Example

    ImagePtr pInImg = loadTiffImageFile(inputImgPath);
    FileReadResult res = readFromXmlFile(inputModelPath);
    RandomForestModelPtr pModel = boost::static_pointer_cast<RandomForestModel>(res._pObject);

    ImagePtr pOutBinLabImg = pixelClassificationRFImg(pInImg, pModel);