Java Plot.DEFAULT_FLAGS属性代码示例

public static Plot createPlot(double [] yValues, String title, String xLabel, String yLabel){
	double [] xValues = new double [yValues.length];
	double min = Double.MAX_VALUE;
	double max = -Double.MAX_VALUE;
	for (int i = 0; i < xValues.length; i ++){
		min = (yValues[i] < min) ? yValues[i] : min;
		max = (yValues[i] > max) ? yValues[i] : max;
		xValues[i] = i + 1;
	}
	if (min == max){
		max++;
	}
	Plot plot = new Plot(title, xLabel, yLabel, xValues, yValues, Plot.DEFAULT_FLAGS);
	plot.setLimits(1, xValues.length, min, max);
	return plot;
}
 

public static Plot createScatterPlot(String title, double [] xCoords, double [] yCoords, Function func){
	double margin = 0.15;
	int scale = 200;
	func.fitToPoints(xCoords, yCoords);
	float [] xVals = new float[scale];
	float [] yVals = new float[scale];
	double [] stats = DoubleArrayUtil.minAndMaxOfArray(xCoords);
	double range = (stats[1] - stats[0]) * (1+(2*margin));
	for (int i=0;i<scale;i++){
		double x = stats[0] - (range * margin) + ((i + 0.0) / scale) * range;
		xVals[i] = (float) x;
		yVals[i] = (float) func.evaluate(x);
	}
	double avg = 0;
	for (int i=0;i<xCoords.length;i++){
		double other = func.evaluate(xCoords[i]);
		avg += Math.pow(other - yCoords[i],2);
	}
	avg /= xCoords.length;
	avg = Math.sqrt(avg);
	
	Plot plot = new Plot(title + " (r = "+DoubleArrayUtil.correlateDoubleArrays(xCoords, yCoords)+" SSIM: "  + DoubleArrayUtil.computeSSIMDoubleArrays(xCoords, yCoords)  + ") Model: " + func.toString() + " standard deviation along model: " + avg, "X", "Y", xVals, yVals, Plot.DEFAULT_FLAGS);		
	plot.addPoints(xCoords, yCoords, Plot.CIRCLE);
	return plot;
}
 

public static Plot createPlot(double [] xValues, double [] yValues, String title, String xLabel, String yLabel){
	double miny = Double.MAX_VALUE;
	double maxy = -Double.MAX_VALUE;
	double minx = Double.MAX_VALUE;
	double maxx = -Double.MAX_VALUE;
	for (int i = 0; i < xValues.length; i ++){
		miny = (yValues[i] < miny) ? yValues[i] : miny;
		maxy = (yValues[i] > maxy) ? yValues[i] : maxy;
		minx = (xValues[i] < minx) ? xValues[i] : minx;
		maxx = (xValues[i] > maxx) ? xValues[i] : maxx;

	}
	if (miny == maxy){
		maxy++;
	}
	if (minx == maxx){
		maxx++;
	}
	Plot plot = new Plot(title, xLabel, yLabel, xValues, yValues, Plot.DEFAULT_FLAGS);
	plot.setLimits(minx, maxx, miny, maxy);
	return plot;
}
 

public static void main(String[] args){
	
	// Define Point Cloud 
	SimpleMatrix p_k = new SimpleMatrix(4,2);
	SimpleMatrix q_k = new SimpleMatrix(4,2);
	
	p_k.setRowValue(0, new SimpleVector(1,2));
	p_k.setRowValue(1, new SimpleVector(3,6));
	p_k.setRowValue(2, new SimpleVector(4,6));
	p_k.setRowValue(3, new SimpleVector(3,4));
	
	q_k.setRowValue(0, new SimpleVector(-0.7, 2.1));
	q_k.setRowValue(1, new SimpleVector(-2.1, 6.4));
	q_k.setRowValue(2, new SimpleVector(-1.4, 7.1));
	q_k.setRowValue(3, new SimpleVector(-0.7, 4.9));
	
	// Plot Point Cloud
	Plot plot = new Plot("Regression Line", "X", "Y", Plot.DEFAULT_FLAGS);
	plot.setLimits(-10, 10, -10, 10);
	plot.addPoints(p_k.getCol(0).copyAsDoubleArray(), p_k.getCol(1).copyAsDoubleArray(), Plot.BOX);
	plot.addPoints(q_k.getCol(0).copyAsDoubleArray(), q_k.getCol(1).copyAsDoubleArray(), Plot.CIRCLE);
	
	// Calculate registration parameter
	Registration1 reg = new Registration1();
	SimpleVector parameter = reg.registrationUsingPointCorrespondences(p_k, q_k);
	
	// Rotation and translation
	double phi = parameter.getElement(0);
	SimpleVector translation = new SimpleVector(parameter.getElement(1), parameter.getElement(2));
	
	// Transform points
	SimpleMatrix transformedPoints = reg.applyTransform(q_k, phi, translation);
	
	// Add transformed point cloud to the plot
	plot.addPoints(transformedPoints.getCol(0).copyAsDoubleArray(), transformedPoints.getCol(1).copyAsDoubleArray(), Plot.CROSS);
	plot.show();
	
}
 

public static void main(String[] args) {
	
	//
	RANSAC ransac = new RANSAC();
	
	//
	// The point cloud is defined
	//
	
	SimpleMatrix pts = new SimpleMatrix(7,2);
	pts.setRowValue(0, new SimpleVector(0,0));
	pts.setRowValue(1, new SimpleVector(1,1));
	pts.setRowValue(2, new SimpleVector(2,2));
	pts.setRowValue(3, new SimpleVector(3,3));
	pts.setRowValue(4, new SimpleVector(3.2,1.9));
	pts.setRowValue(5, new SimpleVector(4,4));
	pts.setRowValue(6, new SimpleVector(10,1.8));


	//
	// Regression Line 
	//
	
	// Create a scatter plot of the point cloud and fit a regression line
	Plot scatterPlot = new Plot("Regression Line", "X", "Y", Plot.DEFAULT_FLAGS);
	scatterPlot.setLimits(0, 11, 0, 5);
	scatterPlot.addPoints(pts.getCol(0).copyAsDoubleArray(), pts.getCol(1).copyAsDoubleArray(), Plot.BOX);
	scatterPlot.show();
	
	// Calculate the regression line through the given point cloud
	SimpleVector regressionLine = ransac.fitline(pts);
	
	// Add the regression line
	double y11 = regressionLine.getElement(0) * 11 + regressionLine.getElement(1);
	double y0 = regressionLine.getElement(0) * 0 + regressionLine.getElement(1);
	scatterPlot.drawLine(0, y0, 11, y11);
	
	
	//
	// RANSAC 
	//
	
	// Parameters for RANSAC
	double p_opt = 0.9999; 		// probability how likely it is to pick the right mn points
	double p_out = 0.2;			// probability of an outlier
	int min_number = 2;			// minimum number of datapoints required to build the model

	// Create a scatter plot of the point cloud and fit a RANSAC line
	Plot ransac_plot = new Plot("Ransac Line", "X", "Y", Plot.DEFAULT_FLAGS);
	ransac_plot.setLimits(0, 11, 0, 5);
	ransac_plot.addPoints(pts.getCol(0).copyAsDoubleArray(), pts.getCol(1).copyAsDoubleArray(), Plot.BOX);
	
	
	// Compute a line using the RANSAC algorithm and plot it
	SimpleVector ransacLine = ransac.commonRansac(pts, min_number, p_opt, p_out);
	double y1 = ransacLine.getElement(0) * 0 + ransacLine.getElement(1);
	double y2 = ransacLine.getElement(0) * 11 + ransacLine.getElement(1);
	ransac_plot.drawLine(0, y1, 11, y2);
	ransac_plot.show();
		
}