/// <summary>
/// This implementation is based on the following articles:
/// 1. Gómez-Lopera, Juan Francisco, et al. (2000)
/// An analysis of edge detection by using the Jensen-Shannon divergence.
/// Journal of Mathematical Imaging and Vision 13.1: 35-56.
///
/// 2. Katatbeh, Qutaibeh D., et al. (2015)
/// An Optimal Segmentation Method Using Jensen–Shannon Divergence via a Multi-Size Sliding Window Technique.
/// Entropy 17.12: 7996-8006.
/// </summary>
static void Main(string[] args)
{
IndexTransformator transformator = new IndexTransformator();
InputParser inputParser = new InputParser(transformator);
HeftArrayCreator heftArrayCreator = new HeftArrayCreator(transformator);
int serverNO;
int pointNO;
int spaceDimension;
int histogramResolution;
int scaleNumber;
int cellMaxValue;
int slidingWindowSize;
Array array;
bool together = inputParser.determineTogetherOrSeparately();
if (together)
{
array = inputParser.parseInputFile(out spaceDimension, out histogramResolution,
out serverNO, out pointNO, out scaleNumber, out cellMaxValue, out slidingWindowSize);
}
else
{
parseInputSeparately(inputParser, out serverNO, out pointNO, out spaceDimension,
out histogramResolution, out scaleNumber, out cellMaxValue, out array, out slidingWindowSize);
}
Console.WriteLine("Point no.: {0}", pointNO);
ShannonEntropyComputer entropyComputer = new ShannonEntropyComputer();
JenShaDivComputer jenShaDivComputer = new JenShaDivComputer(entropyComputer);
FrequencyComputer frequencyComputer = new FrequencyComputer(array, transformator, cellMaxValue,
scaleNumber);
Array heftArray = heftArrayCreator.createHeftArray(spaceDimension, histogramResolution, array);
Array frequencyArray = frequencyComputer.createFrequencyArray(spaceDimension, histogramResolution);
Divider divider = new Divider(array, heftArray, frequencyArray, transformator, jenShaDivComputer,
spaceDimension, histogramResolution, serverNO, slidingWindowSize);
Coords[] partition = divider.determinePartition();
writeOutTiles(serverNO, spaceDimension, partition);
}
