More updates to get DelphinID working

src/rawDeepLearningClassifier/dlClassification/animalSpot/StandardModelPane.java

@@ -32,6 +32,7 @@ import pamViewFX.fxNodes.PamSpinner;
 import pamViewFX.fxNodes.PamVBox;
 import pamViewFX.validator.PamValidator;
 import rawDeepLearningClassifier.dlClassification.DLClassiferModel;
+import rawDeepLearningClassifier.dlClassification.StandardClassifierModel;
 
 /**
  * Settings pane for SoundSpot
@@ -163,7 +164,7 @@ public abstract class StandardModelPane extends SettingsPane<StandardModelParams
 			defaultSegmentLenChanged(); 
 			//only set the hop if the user physically changes the toggle switch. This is not included in defaultSegmentLenChanged
 			//becuase defaultSegmentLenChanged can be called from elsewhere
-			int defaultsamples =  getDefaultSamples();
+			int defaultsamples =  getDefaultSamples(dlClassifierModel, paramsClone);
 			dlClassifierModel.getDLControl().getSettingsPane().getHopLenSpinner().getValueFactory().setValue((int) defaultsamples/2);
 		});
 		usedefaultSeg.setPadding(new Insets(0,0,0,0));
@@ -269,7 +270,7 @@ public abstract class StandardModelPane extends SettingsPane<StandardModelParams
 
 //			float sR = dlClassifierModel.getDLControl().getSettingsPane().getSelectedParentDataBlock().getSampleRate(); 
 
-			int defaultsamples =  getDefaultSamples();
+			int defaultsamples =  getDefaultSamples(dlClassifierModel, paramsClone);
 
 			//work out the window length in samples
 			dlClassifierModel.getDLControl().getSettingsPane().getSegmentLenSpinner().getValueFactory().setValue(defaultsamples);
@@ -282,7 +283,7 @@ public abstract class StandardModelPane extends SettingsPane<StandardModelParams
 		}
 	}
 	
-	private int getDefaultSamples() {
+	public static int getDefaultSamples(DLClassiferModel dlClassifierModel, StandardModelParams paramsClone) {
 		float sR = dlClassifierModel.getDLControl().getSettingsPane().getSelectedParentDataBlock().getSampleRate(); 
 		int defaultsamples =  (int) (paramsClone.defaultSegmentLen.doubleValue()*sR/1000.0);
 		return defaultsamples;

src/rawDeepLearningClassifier/dlClassification/archiveModel/ArchiveModelWorker.java

@@ -58,10 +58,11 @@ public class ArchiveModelWorker extends GenericModelWorker {
 	}
 
 	/**
-	 * Prepare the model 
+	 * Prepare the model.
+	 * Note it is important to put a synchonized here or the model loading can fail. 
 	 */
 	@Override
-	public void prepModel(StandardModelParams dlParams, DLControl dlControl) {
+	public synchronized void prepModel(StandardModelParams dlParams, DLControl dlControl) {
 		//ClassLoader origCL = Thread.currentThread().getContextClassLoader();
 		try {
 
@@ -198,6 +199,8 @@ public class ArchiveModelWorker extends GenericModelWorker {
 	 * @throws IOException
 	 */
 	public ArchiveModel loadModel(String currentPath2) throws MalformedModelException, IOException {
+		
+		System.out.println("HELLO MODEL: " +currentPath2 );
 		return new SimpleArchiveModel(new File(currentPath2)); 
 	}
 

src/rawDeepLearningClassifier/dlClassification/delphinID/DelphinIDPane.java

@@ -1,13 +1,26 @@
 package rawDeepLearningClassifier.dlClassification.delphinID;
 
+
+import java.io.File;
+
 import PamController.SettingsPane;
+import javafx.geometry.Pos;
 import javafx.scene.Node;
+import javafx.scene.control.ContentDisplay;
 import javafx.scene.control.Label;
+import javafx.scene.control.Slider;
+import javafx.scene.control.Spinner;
 import javafx.scene.control.Tooltip;
 import javafx.scene.layout.Pane;
+import javafx.scene.paint.Color;
 import javafx.scene.text.Font;
 import javafx.scene.text.FontWeight;
+import pamViewFX.PamGuiManagerFX;
+import pamViewFX.fxGlyphs.PamGlyphDude;
+import pamViewFX.fxNodes.PamHBox;
+import pamViewFX.fxNodes.PamSpinner;
 import pamViewFX.fxNodes.PamVBox;
+import rawDeepLearningClassifier.dlClassification.animalSpot.StandardModelPane;
 
 /**
  * Settings pane for delphin ID. 
@@ -15,7 +28,7 @@ import pamViewFX.fxNodes.PamVBox;
  * @author Jamie Macaulay
  *
  */
-public class DelphinIDPane  extends SettingsPane<DelphinIDParams> {
+public class DelphinIDPane extends SettingsPane<DelphinIDParams> {
 
 	/**
 	 * The main pane. 
@@ -27,6 +40,14 @@ public class DelphinIDPane  extends SettingsPane<DelphinIDParams> {
 	 */
 	private DelphinIDClassifier delphinUIClassifier;
 
+	private PamSpinner<Double> detectionDensitySpinner;
+
+	private Slider decisionSlider;
+
+	private DelphinIDParams currentParams;
+
+	private File currentSelectedFile;
+
 	public DelphinIDPane(DelphinIDClassifier delphinUIClassifier) {
 		super(null);
 		this.delphinUIClassifier = delphinUIClassifier; 
@@ -35,40 +56,113 @@ public class DelphinIDPane  extends SettingsPane<DelphinIDParams> {
 
 	private Pane createPane() {
 
-
 		//font to use for title labels. 
 		Font font= Font.font(null, FontWeight.BOLD, 11);
 
-		Node classifierIcon;
-		classifierIcon = delphinUIClassifier.getModelUI().getIcon();
+		Label classifierIcon;
+		classifierIcon = new Label("DelphinID");
+		PamGuiManagerFX.titleFont2style(classifierIcon);
+		//todo - will need to figure out colour of icon using CSS. 
+		Node icon = PamGlyphDude.createPamIcon("mdi2r-rss", Color.BLACK, PamGuiManagerFX.iconSize);
+		icon.getStyleClass().add(getName()); 
+		icon.setRotate(45);
+		classifierIcon.setGraphic(icon);
+		classifierIcon.setContentDisplay(ContentDisplay.RIGHT);
 
 
+		//		String settings = currentParams.toString();
+		//		classifierIcon.setTooltip(new Tooltip(settings));
+
 		PamVBox vBox = new PamVBox(); 
 		vBox.setSpacing(5.);
 
+		/**Classification thresholds etc to set.**/
+		Label detectionDensity = new Label("Detection Density"); 
+		detectionDensity.setFont(font);
+		String tooltip = "Set the minimum detection density to attempt to classify.";
+		detectionDensity.setTooltip(new Tooltip(tooltip));
+		detectionDensitySpinner = new PamSpinner<Double>(0.0, 1.0, 0.3, 0.1);
+		detectionDensitySpinner.setPrefWidth(70);
+		detectionDensitySpinner.setEditable(true);
+		detectionDensitySpinner.getStyleClass().add(Spinner.STYLE_CLASS_SPLIT_ARROWS_HORIZONTAL);
+
+		PamHBox minDensityHolder = new PamHBox();
+		minDensityHolder.setAlignment(Pos.CENTER_RIGHT);
+		minDensityHolder.setSpacing(5);
+		Label minDensity = new Label("Min. density");
+		minDensityHolder.getChildren().addAll(minDensity, detectionDensitySpinner);
+
 		/**Classification thresholds etc to set.**/
 		Label classiferInfoLabel2 = new Label("Decision Threshold"); 
 		classiferInfoLabel2.setTooltip(new Tooltip("Set the minimum prediciton value for selected classes. If a prediction exceeds this value "
 				+ "a detection will be saved."));
 		classiferInfoLabel2.setFont(font);
 
+		decisionSlider = new Slider();
+		decisionSlider.setMin(0);
+		decisionSlider.setMax(1);
+		decisionSlider.setMajorTickUnit(0.2);
+		decisionSlider.setMinorTickCount(10);
+		decisionSlider.valueProperty().addListener((obsVal, oldVal, newVal)->{
+			classiferInfoLabel2.setText(String.format("Decision Threshold %.2f", newVal));
+		});
+		decisionSlider.setShowTickMarks(true);
+		decisionSlider.setShowTickLabels(true);
 
-		vBox.getChildren().addAll(classifierIcon, classiferInfoLabel2);
+		vBox.getChildren().addAll(classifierIcon, detectionDensity, minDensityHolder, classiferInfoLabel2, decisionSlider);
 
 		return vBox;
 	}
 
 	@Override
 	public DelphinIDParams getParams(DelphinIDParams currParams) {
-		// TODO Auto-generated method stub
-		return null;
+		currParams.threshold = decisionSlider.getValue();
+		currParams.minDetectionDensity = detectionDensitySpinner.getValue();
+		return currParams;
 	}
 
 	@Override
 	public void setParams(DelphinIDParams input) {
-		// TODO Auto-generated method stub
+		this.currentParams = input;
+		decisionSlider.setValue(input.threshold);
+		detectionDensitySpinner.getValueFactory().setValue(input.minDetectionDensity);
+
+		if (input.modelPath!=null) {
+			//this might 
+			currentSelectedFile = new File(currentParams.modelPath);
+
+			//this might change the paramsClone values if the model contains pamguard compatible metadata
+			newModelSelected(currentSelectedFile); 
+		}
 	}
 
+	private void newModelSelected(File currentSelectedFile2) {
+		if (currentParams!=null && currentParams.defaultSegmentLen != null) {
+
+			//System.out.println("Defualt segment length: " + paramsClone.defaultSegmentLen); 
+
+			//cannot use because, if the parent datablock has changed, samplerate will be out of date. 
+			//			int defaultsamples = (int) this.soundSpotClassifier.millis2Samples(paramsClone.defaultSegmentLen); 
+
+
+			//			float sR = dlClassifierModel.getDLControl().getSettingsPane().getSelectedParentDataBlock().getSampleRate(); 
+
+			int defaultsamples =  StandardModelPane.getDefaultSamples(delphinUIClassifier, currentParams);
+
+			//work out the window length in samples
+			delphinUIClassifier.getDLControl().getSettingsPane().getSegmentLenSpinner().getValueFactory().setValue(defaultsamples);
+			//			dlClassifierModel.getDLControl().getSettingsPane().getHopLenSpinner().getValueFactory().setValue((int) defaultsamples/2);
+
+			delphinUIClassifier.getDLControl().getSettingsPane().getSegmentLenSpinner().setDisable(true); 
+		}
+		else {
+			delphinUIClassifier.getDLControl().getSettingsPane().getSegmentLenSpinner().setDisable(false); 
+		}
+
+	}
+
+
+
 	@Override
 	public String getName() {
 		return "delphinIDParams";

src/rawDeepLearningClassifier/dlClassification/delphinID/DelphinIDParams.java

@@ -9,4 +9,10 @@ public class DelphinIDParams extends StandardModelParams {
 	 */
 	private static final long serialVersionUID = 1L;
 	
+	/**
+	 * The minimum detection density. 
+	 */
+	public double minDetectionDensity = 0.3;
+	
+
 }

src/rawDeepLearningClassifier/dlClassification/delphinID/DelphinIDWorker.java

@@ -48,6 +48,11 @@ public class DelphinIDWorker extends  ArchiveModelWorker {
 			System.err.println("Error: could not find whistle2image transform in DelphinID JSON file. Model will not work.");
 			this.setModel(null); // set model to null to make sure nothing works and errors are thrown
 		}
+		
+		dlParams.binaryClassification = new boolean[dlParams.classNames.length];
+		for (int i=0; i<dlParams.classNames.length; i++) {
+			dlParams.binaryClassification[i]=true;
+		}
 	}
 
 

src/rawDeepLearningClassifier/dlClassification/delphinID/Whistles2Image.java

@@ -1,6 +1,8 @@
 package rawDeepLearningClassifier.dlClassification.delphinID;
 
 import java.awt.Color;
+import java.awt.Graphics2D;
+import java.awt.RenderingHints;
 import java.awt.image.BufferedImage;
 import java.awt.image.Raster;
 import java.util.ArrayList;
@@ -72,7 +74,7 @@ public class Whistles2Image extends FreqTransform {
 		for (int i=0; i<imaged.length; i++) {
 			for (int j=0; j<imaged[0].length; j++) {
 				color = raster.getPixel(i, j, color);
-				imaged[i][j] = color[0]/255.; //normalize
+				imaged[i][j] = (255-color[0])/255.; //normalize
 			}
 		}
 //		
@@ -173,7 +175,7 @@ public class Whistles2Image extends FreqTransform {
 	 * @param markerSize - the marker size in pixels
 	 * @return an image with y axis as frequency and x axis as time. 
 	 */
-	private BufferedImage makeScatterImage(ArrayList<double[][]> points, double[] size, double[] xlims, double[] ylims, double markerSize) {
+	public static BufferedImage makeScatterImage(ArrayList<double[][]> points, double[] size, double[] xlims, double[] ylims, double markerSize) {
 
 		BufferedImage canvas = new BufferedImage((int) size[0], (int) size[1], BufferedImage.TYPE_INT_RGB);
 
@@ -187,9 +189,13 @@ public class Whistles2Image extends FreqTransform {
 				x = ((points.get(j)[i][0]-xlims[0])/(xlims[1]-xlims[0]))*size[0];
 				y = ((points.get(j)[i][1]-ylims[0])/(ylims[1]-ylims[0]))*size[1];
 				
-//				System.out.println("Fill oval: x" + x + " y: " + y + " time: " + points.get(j)[i][0]);
+				//System.out.println("Fill oval: x" + x + " y: " + y + " time: " + points.get(j)[i][0]);
 				
-				canvas.getGraphics().fillOval((int) (x+markerSize/2),(int) (y-markerSize/2), (int) markerSize,(int) markerSize);
+				Graphics2D g2 = (Graphics2D) canvas.getGraphics();
+
+		        g2.setRenderingHint(RenderingHints.KEY_ANTIALIASING, RenderingHints.VALUE_ANTIALIAS_ON);
+
+		        g2.fillOval((int) (x+markerSize/2),(int) (y-markerSize/2), (int) markerSize,(int) markerSize);
 			}
 		}
 
@@ -209,4 +215,5 @@ public class Whistles2Image extends FreqTransform {
 	
 
 
+
 }

src/rawDeepLearningClassifier/segmenter/SegmenterProcess.java

@@ -62,6 +62,16 @@ public class SegmenterProcess extends PamProcess {
 	 */
 	private SegmenterGroupDataBlock segmenterGroupDataBlock;
 
+	/**
+	 * The first clock update - segments for detection groups (not raw sound data) are referenced from this. 
+	 */
+	private long firstClockUpdate; 
+	
+	/**
+	 * The current segmenter detection group.
+	 */
+	private SegmenterDetectionGroup segmenterDetectionGroup = null; 
+
 
 	public SegmenterProcess(DLControl pamControlledUnit, PamDataBlock parentDataBlock) {
 		super(pamControlledUnit, parentDataBlock);
@@ -111,7 +121,7 @@ public class SegmenterProcess extends PamProcess {
 	 */
 	@Override
 	public ArrayList getCompatibleDataUnits(){
-		return new ArrayList<Class<? extends PamDataUnit>>(Arrays.asList(RawDataUnit.class, ClickDetection.class, ClipDataUnit.class));
+		return new ArrayList<Class<? extends PamDataUnit>>(Arrays.asList(RawDataUnit.class, ClickDetection.class, ClipDataUnit.class, ConnectedRegionDataUnit.class));
 	}
 
 
@@ -178,6 +188,8 @@ public class SegmenterProcess extends PamProcess {
 
 		setParentDataBlock(rawDataBlock);
 		
+		this.firstClockUpdate = -1;
+
 	}
 
 	/*
@@ -232,12 +244,29 @@ public class SegmenterProcess extends PamProcess {
 
 		//TODO
 		//this contains no raw data so we are branching off on a completely different processing path here.
-		//Whislte data units are saved to a buffer and then fed to the deep learning algorohtm
+		//Whislte data units are saved to a buffer and then fed to the deep learning algorithms
+	
+		if (segmenterDetectionGroup==null) {
+			//iterate until we find the correct time 
+			long segmentStart = firstClockUpdate;
+			while() {
+				
+			}
+		}
 		
 		
 	}
 	
 	
+	public void masterClockUpdate(long milliSeconds, long sampleNumber) {
+		super.masterClockUpdate(milliSeconds, sampleNumber);
+		if (firstClockUpdate<0) {
+			firstClockUpdate = milliSeconds;
+		}
+	}
+
+
+
 	/**
 	 * A new raw data unit. 
 	 * @param obs - the PAM observable 

src/test/rawDeepLearningClassifier/DelphinIDTest.java

@@ -0,0 +1,91 @@
+package test.rawDeepLearningClassifier;
+
+import static org.junit.jupiter.api.Assertions.assertEquals;
+
+import java.awt.image.BufferedImage;
+import java.awt.image.Raster;
+import java.io.IOException;
+import java.nio.file.Path;
+import java.nio.file.Paths;
+import java.util.ArrayList;
+
+import org.jamdev.jdl4pam.utils.DLMatFile;
+import org.junit.jupiter.api.Test;
+
+import rawDeepLearningClassifier.dlClassification.delphinID.Whistles2Image;
+import us.hebi.matlab.mat.format.Mat5;
+import us.hebi.matlab.mat.types.MatFile;
+import us.hebi.matlab.mat.types.Matrix;
+
+public class DelphinIDTest { 
+	
+	
+	@Test
+	public void whistle2ImageTest() {
+	
+		System.out.println("Whislte2Image test started");
+
+		/**
+		 * Test whether the Whistles2Image transform works properly
+		 */
+		String relMatPath  =	"./src/test/resources/rawDeepLearningClassifier/DelphinID/whistle_image_example.mat";
+
+		Path path = Paths.get(relMatPath);
+	
+		// Create MAT file with a scalar in a nested struct
+		try {
+			MatFile matFile = Mat5.readFromFile(path.toString());
+			Matrix array = matFile.getArray("tfvalues");
+			
+			//the values for the whistle detector.
+			double[][] whistleValues = DLMatFile.matrix2array(array);
+			
+			//the image after compression 
+			array = matFile.getArray("image1compressedgrayscale");
+			double[][] compressedWhistleImage = DLMatFile.matrix2array(array);
+
+			//the whistle2Image transform image
+			array = matFile.getArray("image1originalgrayscalenorm");
+			double[][] whislteImage = DLMatFile.matrix2array(array);
+			
+			//now perform the image transform in Java 
+			double[] freqLimits = new double[] {0., 20000.};
+			double[] size = new double[] {680., 480.};
+			
+			ArrayList<double[][]> whistleImageArr = new ArrayList<double[][]>();
+			whistleImageArr.add(whistleValues);
+			
+			BufferedImage canvas = Whistles2Image.makeScatterImage(whistleImageArr, size, new double[]{48, 48. + 4.}, freqLimits,  5.);
+
+			double[][] imaged = new double[(int) size[0]][(int) size[1]];
+
+			float[] color = new float[3];
+			Raster raster = canvas.getData();
+			for (int i=0; i<imaged.length; i++) {
+				for (int j=0; j<imaged[0].length; j++) {
+					color = raster.getPixel(i, j, color);
+					imaged[i][j] = (255-color[0])/255.; //normalize
+				}
+			}
+			
+			//now save this image to a MATFILE
+			// Create MAT file with a scalar in a nested struct
+			MatFile matFileWrite = Mat5.newMatFile()
+			    .addArray("image1originalgrayscalenorm",DLMatFile.array2Matrix(imaged));
+			// Serialize to disk using default configurations
+			Mat5.writeToFile(matFileWrite, "C:\\Users\\Jamie Macaulay\\MATLAB Drive\\MATLAB\\PAMGUARD\\deep_learning\\delphinID\\whistle_image_example_java.mat");
+			
+			System.out.println("Whislte2Image test finished");
+			
+		} catch (IOException e) {
+			// TODO Auto-generated catch block
+			e.printStackTrace();
+			assertEquals(false, false);
+		}
+		
+		
+		
+		
+	}
+
+}