@Override
public void apply(Mat src, Mat dst) {
Core.split(src, mChannels);
final Mat r = mChannels.get(0);
final Mat g = mChannels.get(1);
final Mat b = mChannels.get(2);
Core.min(b, r, b);
Core.min(b, g, b);
Core.merge(mChannels, dst);
}
/**
* Generates a mask of all resistors present in the given Mat. Also displays this mask in the
* Bottom Left frame of the GUI.
*
* @param imgCap The Mat image to generate the mask for
* @param type The threshold operation type
* @return The mask as a Mat
*/
private Mat generateResistorMask(Mat imgCap, int type) {
Mat imgHSV = new Mat();
Mat satImg = new Mat();
// convert the input image from BGR to HSV
Imgproc.cvtColor(imgCap, imgHSV, Imgproc.COLOR_BGR2HSV);
ArrayList<Mat> channels = new ArrayList<Mat>();
Core.split(imgHSV, channels);
// extract the saturation channel
satImg = channels.get(1);
// remove the background and the resistor leads (combined with previous blurring)
// thresh ~86
Imgproc.threshold(satImg, satImg, RESISTOR_MASK_THRESHOLD, 255, type);
paintBL(satImg);
return satImg;
}
public static Mat Histogram(Mat im) {
Mat img = im;
Mat equ = new Mat();
img.copyTo(equ);
// Imgproc.blur(equ, equ, new Size(3, 3));
Imgproc.cvtColor(equ, equ, Imgproc.COLOR_BGR2YCrCb);
List<Mat> channels = new ArrayList<Mat>();
Core.split(equ, channels);
Imgproc.equalizeHist(channels.get(0), channels.get(0));
Core.merge(channels, equ);
Imgproc.cvtColor(equ, equ, Imgproc.COLOR_YCrCb2BGR);
Mat gray = new Mat();
Imgproc.cvtColor(equ, gray, Imgproc.COLOR_BGR2GRAY);
Mat grayOrig = new Mat();
Imgproc.cvtColor(img, grayOrig, Imgproc.COLOR_BGR2GRAY);
System.out.println("Histogram work ///");
return grayOrig;
}
/**
* @param inputImg
* @param minValue
* @param maxValue
* @return Mat
*/
public static Mat thresholding(Mat inputImg, Integer minValue, Integer maxValue) {
Mat frame = inputImg;
// яскравість
// frame.convertTo(frame , -1, 10d * 33 / 100, 0);
// Imgproc.medianBlur(frame,frame, 17);
// Core.bitwise_not(frame,frame );
// Mat frame = new Mat(image.rows(), image.cols(), image.type());
// frame.convertTo(frame, -1, 10d * 20 / 100, 0);
Mat hsvImg = new Mat();
List<Mat> hsvPlanes = new ArrayList<>();
Mat thresholdImg = new Mat();
int thresh_type = Imgproc.THRESH_BINARY_INV;
// if (this.inverse.isSelected())
// thresh_type = Imgproc.THRESH_BINARY;
// threshold the image with the average hue value
// System.out.println("size " +frame.size());
hsvImg.create(frame.size(), CvType.CV_8U);
// Imgproc.cvtColor(frame, hsvImg, Imgproc.COLOR_BGR2HSV);
Core.split(hsvImg, hsvPlanes);
// get the average hue value of the image
// double threshValue = PreProcessingOperation.getHistAverage(hsvImg, hsvPlanes.get(0));
// System.out.println(threshValue);
/*
if(threshValue > 40){
maxValue = 160;
}else{
maxValue = 40;
}*/
// Imgproc.threshold(hsvPlanes.get(1), thresholdImg, minValue , maxValue , thresh_type);
Imgproc.blur(thresholdImg, thresholdImg, new Size(27, 27));
// dilate to fill gaps, erode to smooth edges
Imgproc.dilate(thresholdImg, thresholdImg, new Mat(), new Point(-1, -1), 1);
Imgproc.erode(thresholdImg, thresholdImg, new Mat(), new Point(-1, -1), 1);
Imgproc.threshold(thresholdImg, thresholdImg, minValue, maxValue, Imgproc.THRESH_BINARY);
// create the new image
Mat foreground = new Mat(frame.size(), CvType.CV_8UC3, new Scalar(255, 255, 255));
Core.bitwise_not(thresholdImg, foreground);
frame.copyTo(foreground, thresholdImg);
///////////////////////////////////////////////////////////////////////////////////////
///
////
return foreground;
/*Mat hsvImg = new Mat();
List<Mat> hsvPlanes = new ArrayList<>();
Mat thresholdImg = new Mat();
int thresh_type = Imgproc.THRESH_BINARY_INV;
// threshold the image with the average hue value
hsvImg.create(inputImg.size(), CvType.CV_8U);
Imgproc.cvtColor(inputImg, hsvImg, Imgproc.COLOR_BGR2HSV);
Core.split(hsvImg, hsvPlanes);
// get the average hue value of the image
double threshValue = PreProcessingOperation.getHistAverage(hsvImg, hsvPlanes.get(0));
Imgproc.threshold(hsvPlanes.get(0), thresholdImg, minValue,
maxValue, thresh_type);
Imgproc.blur(thresholdImg, thresholdImg, new Size(3, 3));
// dilate to fill gaps, erode to smooth edges
Imgproc.dilate(thresholdImg, thresholdImg, new Mat(), new Point(-1, -1), 3);
Imgproc.erode(thresholdImg, thresholdImg, new Mat(), new Point(-1, -1), 1);
Imgproc.threshold(thresholdImg, thresholdImg, minValue,
maxValue, Imgproc.THRESH_BINARY);
// create the new image
Mat foreground = new Mat(inputImg.size(), CvType.CV_8UC3, new Scalar(255, 255, 255));
inputImg.copyTo(foreground, thresholdImg);
Core.bitwise_not(foreground,foreground);
return foreground;*/
}
/**
* @param inputImg
* @return Mat
*/
public static Mat kmeans(Mat inputImg) {
Mat rgba = inputImg;
Mat tempMat = inputImg;
rgba = new Mat(inputImg.cols(), inputImg.rows(), CvType.CV_8UC3);
inputImg.copyTo(rgba);
List<Mat> hsv_planes_temp = new ArrayList<Mat>(3);
Core.split(tempMat, hsv_planes_temp);
double threshValue1 = PreProcessingOperation.getHistAverage(inputImg, hsv_planes_temp.get(0));
sample.util.Estimate.setFirstHistAverageValue(threshValue1);
System.out.println("Defore eqau " + threshValue1);
System.out.println(
Estimate.getBlueAverage() + " ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;");
if (threshValue1 > 140) {
if (Estimate.getBlueAverage() > 110) {
rgba.convertTo(rgba, -1, 10d * 31 / 100, 0);
System.out.println("11");
} else {
rgba.convertTo(rgba, -1, 10d * 40 / 100, 0);
System.out.println("12");
}
} else if (threshValue1 > 135) {
rgba.convertTo(rgba, -1, 10d * 32 / 100, 0);
System.out.println("21");
} else if (threshValue1 > 125) {
if (Estimate.getBlueAverage() > 110) {
rgba.convertTo(rgba, -1, 10d * 30 / 100, 0);
rgba = PreProcessing.Dilate(rgba, 5);
System.out.println("31");
} else {
rgba.convertTo(rgba, -1, 10d * 37 / 100, 0);
System.out.println("32");
}
} else if (threshValue1 > 120) {
rgba.convertTo(rgba, -1, 10d * 35 / 100, 0);
System.out.println("41");
} else if (threshValue1 > 110) {
if (Estimate.getBlueAverage() > 110) {
rgba.convertTo(rgba, -1, 10d * 35 / 100, 0);
rgba = PreProcessing.Dilate(rgba, 5);
System.out.println("51");
}
} else if (threshValue1 > 100) {
if (Estimate.getBlueAverage() > 107) {
rgba.convertTo(rgba, -1, 10d * 24 / 100, 0);
rgba = PreProcessing.Dilate(rgba, 5);
System.out.println("61");
} else if (Estimate.getBlueAverage() > 90) {
rgba.convertTo(rgba, -1, 10d * 30 / 100, 0);
rgba = PreProcessing.Dilate(rgba, 5);
System.out.println("62");
}
} else if (threshValue1 > 50) {
if (Estimate.getBlueAverage() > 160) {
rgba.convertTo(rgba, -1, 10d * 30 / 100, 0);
rgba = PreProcessing.Dilate(rgba, 3);
System.out.println("81");
} else if (Estimate.getBlueAverage() > 160) {
rgba.convertTo(rgba, -1, 10d * 27 / 100, 0);
rgba = PreProcessing.Dilate(rgba, 9);
System.out.println("82");
} else if (Estimate.getBlueAverage() > 130) {
rgba.convertTo(rgba, -1, 10d * 30 / 100, 0);
rgba = PreProcessing.Dilate(rgba, 9);
System.out.println("83");
} else if (Estimate.getBlueAverage() > 70) {
rgba.convertTo(rgba, -1, 10d * 29 / 100, 0);
rgba = PreProcessing.Dilate(rgba, 9);
System.out.println("84");
}
} else if (threshValue1 > 30) {
if (Estimate.getBlueAverage() > 190) {
rgba.convertTo(rgba, -1, 10d * 25 / 100, 0);
System.out.println("91");
} else if (Estimate.getBlueAverage() > 160) {
rgba.convertTo(rgba, -1, 10d * 35 / 100, 0);
System.out.println("92");
}
} else {
if (Estimate.getBlueAverage() > 240) {
rgba.convertTo(rgba, -1, 10d * 24 / 100, 0);
System.out.println("7");
} else {
rgba.convertTo(rgba, -1, 10d * 17 / 100, 0);
System.out.println("7");
}
}
tempMat.release();
Mat mHSV = new Mat();
Imgproc.cvtColor(rgba, mHSV, Imgproc.COLOR_RGBA2RGB, 3);
Imgproc.cvtColor(rgba, mHSV, Imgproc.COLOR_RGB2HSV, 3);
List<Mat> hsv_planes = new ArrayList<Mat>(3);
Core.split(mHSV, hsv_planes);
Mat channel = hsv_planes.get(0);
channel = Mat.zeros(mHSV.rows(), mHSV.cols(), CvType.CV_8UC1);
hsv_planes.set(2, channel);
Core.merge(hsv_planes, mHSV);
mHSV.convertTo(mHSV, CvType.CV_8UC1);
mHSV = Histogram(mHSV);
/*
Mat clusteredHSV = new Mat();
mHSV.convertTo(mHSV, CvType.CV_32FC3);
TermCriteria criteria = new TermCriteria(TermCriteria.EPS + TermCriteria.MAX_ITER,100,0.1);
Core.kmeans(mHSV, 1, clusteredHSV, criteria, 20, Core.KMEANS_PP_CENTERS);
Mat hsvImg = new Mat();
List<Mat> hsvPlanes = new ArrayList<>();
Mat thresholdImg = new Mat();
int thresh_type = Imgproc.THRESH_BINARY_INV;
hsvImg.create(mHSV.size(), CvType.CV_8U);
Imgproc.cvtColor(mHSV, hsvImg, Imgproc.COLOR_BGR2HSV);
Core.split(hsvImg, hsvPlanes);
Imgproc.threshold(hsvPlanes.get(1), thresholdImg, 0 , 200 , thresh_type);
double threshValue = PreProcessingOperation.getHistAverage(hsvImg, hsvPlanes.get(0));
Estimate.setSecondHistAverageValue(threshValue);
System.out.println("After equa " + Estimate.getSecondHistAverageValue());*/
Imgproc.threshold(mHSV, mHSV, 0, 150, Imgproc.THRESH_BINARY_INV);
// mHSV.convertTo(mHSV, CvType.CV_8UC1);
return mHSV;
}
/**
* Extracts and classifies colour bands for each Resistor. Each ColourBand object is instantiated
* and linked to their parent Resistor object.
*
* @param resistorList A list of Resistor objects from which to extract the colour bands
* @param paintDebugInfo If ture, the extracted colour band ROIs are displayed on the GUI
*/
private void extractColourBandsAndClassify(List<Resistor> resistorList, boolean paintDebugInfo) {
if (resistorList.size() > 0) {
for (int r = 0; r < resistorList.size(); r++) {
Mat resImg = resistorList.get(r).resistorMat;
Mat imgHSV = new Mat();
Mat satImg = new Mat();
Mat hueImg = new Mat();
// convert to HSV
Imgproc.cvtColor(resImg, imgHSV, Imgproc.COLOR_BGR2HSV);
ArrayList<Mat> channels = new ArrayList<Mat>();
Core.split(imgHSV, channels);
// extract channels
satImg = channels.get(1); // saturation
hueImg = channels.get(0); // hue
// threshold saturation channel
Mat threshedROISatBands = new Mat(); // ~130 sat thresh val
Imgproc.threshold(satImg, threshedROISatBands, SAT_BAND_THRESH, 255, Imgproc.THRESH_BINARY);
// threshold hue channel
Mat threshedROIHueBands = new Mat(); // ~50 hue thresh val
Imgproc.threshold(hueImg, threshedROIHueBands, HUE_BAND_THRESH, 255, Imgproc.THRESH_BINARY);
// combine the thresholded binary images
Mat bandROI = new Mat();
Core.bitwise_or(threshedROIHueBands, threshedROISatBands, bandROI);
// find contours in binary ROI image
ArrayList<MatOfPoint> contours = new ArrayList<MatOfPoint>();
Mat hierarchy = new Mat();
Imgproc.findContours(
bandROI,
contours,
hierarchy,
Imgproc.RETR_EXTERNAL,
Imgproc.CHAIN_APPROX_SIMPLE,
new Point(0, 0));
// remove any remaining noise by only keeping contours which area > threshold
for (int i = 0; i < contours.size(); i++) {
double area = Imgproc.contourArea(contours.get(i));
if (area < MIN_BAND_AREA) {
contours.remove(i);
i--;
}
}
// create a ColourBand object for each detected band
// storing its center, the contour and the bandROI
for (int i = 0; i < contours.size(); i++) {
MatOfPoint contour = contours.get(i);
// extract this colour band and store in a Mat
Rect boundingRect = Imgproc.boundingRect(contour);
Mat mask = Mat.zeros(bandROI.size(), CvType.CV_8U);
Imgproc.drawContours(mask, contours, i, new Scalar(255), Core.FILLED);
Mat imageROI = new Mat();
resImg.copyTo(imageROI, mask);
Mat colourBandROI = new Mat(imageROI, boundingRect);
// instantiate new ColourBand object
ColourBand cb = new ColourBand(findCenter(contour), contour, colourBandROI);
// cluster the band colour
cb.clusterBandColour(BAND_COLOUR_K_MEANS);
// classify using the Lab colourspace as feature vector
Mat sampleMat =
new Mat(1, 3, CvType.CV_32FC1); // create a Mat contacting the clustered band colour
sampleMat.put(0, 0, cb.clusteredColourLAB[0]);
sampleMat.put(0, 1, cb.clusteredColourLAB[1]);
sampleMat.put(0, 2, cb.clusteredColourLAB[2]);
Mat classifiedValue = new Mat(1, 1, CvType.CV_32FC1);
Mat neighborResponses = new Mat(); // dont actually use this
Mat dists = new Mat(); // dont actually use this
// classify
knn.find_nearest(sampleMat, 3, classifiedValue, neighborResponses, dists);
// cast classified value into Colour enum and store
cb.classifiedColour = ColourEnumVals[(int) classifiedValue.get(0, 0)[0]];
// add the band to the parent resistor
resistorList.get(r).bands.add(cb);
}
// paint the extracted band ROIs
if (paintDebugInfo) {
Mat finalBandROIMask = Mat.zeros(bandROI.size(), CvType.CV_8U);
for (int i = 0; i < contours.size(); i++) {
Scalar color = new Scalar(255, 255, 255);
Imgproc.drawContours(
finalBandROIMask, contours, i, color, -1, 4, hierarchy, 0, new Point());
}
Mat colourROI = new Mat();
resImg.copyTo(colourROI, finalBandROIMask);
paintResistorSubRegion(colourROI, r);
}
}
}
}
/**
* Identifies the color in the frame
*
* @param in the Mat image in the region of interest
* @return the color
*/
public char identifyColor(Mat in) {
// Mat blue = new Mat(in.rows(), in.cols(), CvType.CV_8UC1);
// Mat green = new Mat(in.rows(), in.cols(), CvType.CV_8UC1);
// Mat red = new Mat(in.rows(), in.cols(), CvType.CV_8UC1);
// split the channels of the image
Mat blue = new Mat(); // default is CV_8UC3
Mat green = new Mat();
Mat red = new Mat();
List<Mat> channels = new ArrayList<Mat>(3);
Core.split(in, channels);
blue = channels.get(0); // makes all 3 CV_8UC1
green = channels.get(1);
red = channels.get(2);
// System.out.println(blue.toString());
// add the intensities
Mat intensity = new Mat(in.rows(), in.cols(), CvType.CV_32F);
// Mat mask = new Mat();
Core.add(blue, green, intensity); // , mask, CvType.CV_32F);
Core.add(intensity, red, intensity); // , mask, CvType.CV_32F);
// not sure if correct from here to ...
Mat inten = new Mat();
Core.divide(intensity, Scalar.all(3.0), inten);
// System.out.println(intensity.toString());
// Core.divide(3.0, intensity, inten);
// if intensity = intensity / 3.0; means element-wise division
// use intensity.muls(Mat m)
// so make new Mat m of same size that has each element of 1/3
/*
* or
* About per-element division you can use Core.divide()
Core.divide(A,Scalar.all(d), B);
It's equivalent to B=A/d
*/
// find normalized values
Mat bnorm = new Mat();
Mat gnorm = new Mat();
Mat rnorm = new Mat();
// blue.convertTo(blue, CvType.CV_32F);
// green.convertTo(green, CvType.CV_32F);
// red.convertTo(red, CvType.CV_32F);
Core.divide(blue, inten, bnorm);
Core.divide(green, inten, gnorm);
Core.divide(red, inten, rnorm);
// find average norm values
Scalar val = new Scalar(0);
val = Core.mean(bnorm);
String value[] = val.toString().split(",");
String s = value[0].substring(1);
double bavg = Double.parseDouble(s);
val = Core.mean(gnorm);
String value1[] = val.toString().split(",");
String s1 = value1[0].substring(1);
double gavg = Double.parseDouble(s1);
val = Core.mean(rnorm);
String value2[] = val.toString().split(",");
String s2 = value2[0].substring(1);
double ravg = Double.parseDouble(s2);
// ... here
// original values
/*
// define the reference color values
//double RED[] = {0.4, 0.5, 1.8};
//double GREEN[] = {1.0, 1.2, 1.0};
double BLUE[] = {1.75, 1.0, 0.5};
//double YELLOW[] = {0.82, 1.7, 1.7};
double ORANGE[] = {0.2, 1.0, 2.0};
double WHITE[] = {2.0, 1.7, 1.7};
//double BLACK[] = {0.0, 0.3, 0.3};
*/
// define the reference color values
// double RED[] = {0.4, 0.5, 1.8};
// double GREEN[] = {1.0, 1.2, 1.0};
double BLUE[] = {1.75, 1.0, 0.5};
// double YELLOW[] = {0.82, 1.7, 1.7};
double ORANGE[] = {0.2, 1.0, 2.0};
double WHITE[] = {2.0, 1.7, 1.7};
// double BLACK[] = {0.0, 0.3, 0.3};
// compute the square error relative to the reference color values
// double minError = 3.0;
double minError = 2.0;
double errorSqr;
char bestFit = 'x';
// test++;
// System.out.print("\n\n" + test + "\n\n");
// check BLUE fitness
errorSqr = normSqr(BLUE[0], BLUE[1], BLUE[2], bavg, gavg, ravg);
System.out.println("Blue: " + errorSqr);
if (errorSqr < minError) {
minError = errorSqr;
bestFit = COLOR_BLUE;
}
// check ORANGE fitness
errorSqr = normSqr(ORANGE[0], ORANGE[1], ORANGE[2], bavg, gavg, ravg);
System.out.println("Orange: " + errorSqr);
if (errorSqr < minError) {
minError = errorSqr;
bestFit = COLOR_ORANGE;
}
// check WHITE fitness
errorSqr = normSqr(WHITE[0], WHITE[1], WHITE[2], bavg, gavg, ravg);
System.out.println("White: " + errorSqr);
if (errorSqr < minError) {
minError = errorSqr;
bestFit = COLOR_WHITE;
}
// check BLACK fitness
/*errorSqr = normSqr(BLACK[0], BLACK[1], BLACK[2], bavg, gavg, ravg);
System.out.println("Black: " + errorSqr);
if(errorSqr < minError)
{
minError = errorSqr;
bestFit = COLOR_BLACK;
}*/
// return the best fit color label
return bestFit;
}
