List of usage examples for org.opencv.core MatOfPoint2f MatOfPoint2f
public MatOfPoint2f()
From source file:ImageReade.java
public static void detectLetter(Mat img) { ArrayList<Rect> boundRect = new ArrayList<>(); Mat img_gray, img_sobel, img_threshold, element; img_gray = new Mat(); img_sobel = new Mat(); img_threshold = new Mat(); element = new Mat(); Imgproc.cvtColor(img, img_gray, Imgproc.COLOR_BGRA2GRAY); imshow("Rec img_gray", img_gray); Imgproc.Sobel(img_gray, img_sobel, CvType.CV_8U, 1, 0, 3, 1, 0, Imgproc.BORDER_DEFAULT); imshow("Rec img_sobel", img_sobel); Imgproc.threshold(img_sobel, img_threshold, 0, 255, CV_THRESH_OTSU + CV_THRESH_BINARY); imshow("Rec img_threshold", img_threshold); element = Imgproc.getStructuringElement(Imgproc.MORPH_RECT, new Size(16, 6)); Imgproc.morphologyEx(img_threshold, img_threshold, CV_MOP_CLOSE, element); imshow("Rec img_threshold second", img_threshold); List<MatOfPoint> contours = new ArrayList<MatOfPoint>(); //Imgproc.findContours(img_threshold, contours, new Mat(), Imgproc.RETR_LIST,Imgproc.CHAIN_APPROX_SIMPLE); Imgproc.findContours(img_threshold, contours, new Mat(), 0, 1); for (int i = 0; i < contours.size(); i++) { System.out.println(Imgproc.contourArea(contours.get(i))); // if (Imgproc.contourArea(contours.get(i)) > 100) { // //Imgproc.approxPolyDP( contours.get(i), contours_poly[i], 3, true ); // Rect rect = Imgproc.boundingRect(contours.get(i)); // System.out.println(rect.height); // if (rect.width > rect.height) { // //System.out.println(rect.x +","+rect.y+","+rect.height+","+rect.width); // Core.rectangle(img, new Point(rect.x,rect.y), new Point(rect.x+rect.width,rect.y+rect.height),new Scalar(0,0,255)); // } // // // } if (Imgproc.contourArea(contours.get(i)) > 100) { MatOfPoint2f mMOP2f1 = new MatOfPoint2f(); MatOfPoint2f mMOP2f2 = new MatOfPoint2f(); contours.get(i).convertTo(mMOP2f1, CvType.CV_32FC2); Imgproc.approxPolyDP(mMOP2f1, mMOP2f2, 3, true); mMOP2f2.convertTo(contours.get(i), CvType.CV_32S); Rect rect = Imgproc.boundingRect(contours.get(i)); if (rect.width > rect.height) { Core.rectangle(img, new Point(rect.x, rect.y), new Point(rect.x + rect.width, rect.y + rect.height), new Scalar(0, 0, 255)); }//from w ww . j av a 2 s . c om } } imshow("Rec Detected", img); }
From source file:ThirdTry.java
public static void detectLetter(Mat img, Mat m2) { ArrayList<Rect> boundRect = new ArrayList<>(); Mat img_gray, img_sobel, img_threshold, element; img_gray = new Mat(); img_sobel = new Mat(); img_threshold = new Mat(); element = new Mat(); Imgproc.cvtColor(img, img_gray, Imgproc.COLOR_BGRA2GRAY); //imshow("Rec img_gray", img_gray); Imgproc.Sobel(img_gray, img_sobel, CvType.CV_8UC1, 1, 0, 3, 1, 0, Imgproc.BORDER_DEFAULT); //imshow("Rec img_sobel", img_sobel); Imgproc.threshold(m2, img_threshold, 0, 255, CV_THRESH_OTSU + CV_THRESH_BINARY); //imshow("Rec img_threshold", img_threshold); element = Imgproc.getStructuringElement(Imgproc.MORPH_RECT, new Size(3, 2)); Imgproc.morphologyEx(m2, img_threshold, CV_MOP_CLOSE, element); imshow("Rec img_threshold second", img_threshold); element = Imgproc.getStructuringElement(Imgproc.MORPH_RECT, new Size(12, 12)); Imgproc.morphologyEx(img_threshold, img_threshold, CV_MOP_CLOSE, element); //imshow("Rec img_threshold second", img_threshold); List<MatOfPoint> contours = new ArrayList<MatOfPoint>(); //Imgproc.findContours(img_threshold, contours, new Mat(), Imgproc.RETR_LIST,Imgproc.CHAIN_APPROX_SIMPLE); Imgproc.findContours(img_threshold, contours, new Mat(), 0, 1); for (int i = 0; i < contours.size(); i++) { System.out.println(Imgproc.contourArea(contours.get(i))); // if (Imgproc.contourArea(contours.get(i)) > 100) { // //Imgproc.approxPolyDP( contours.get(i), contours_poly[i], 3, true ); // Rect rect = Imgproc.boundingRect(contours.get(i)); // System.out.println(rect.height); // if (rect.width > rect.height) { // //System.out.println(rect.x +","+rect.y+","+rect.height+","+rect.width); // Core.rectangle(img, new Point(rect.x,rect.y), new Point(rect.x+rect.width,rect.y+rect.height),new Scalar(0,0,255)); // } // // // } if (Imgproc.contourArea(contours.get(i)) > 100) { MatOfPoint2f mMOP2f1 = new MatOfPoint2f(); MatOfPoint2f mMOP2f2 = new MatOfPoint2f(); contours.get(i).convertTo(mMOP2f1, CvType.CV_32FC2); Imgproc.approxPolyDP(mMOP2f1, mMOP2f2, 3, true); mMOP2f2.convertTo(contours.get(i), CvType.CV_32S); Rect rect = Imgproc.boundingRect(contours.get(i)); if (rect.width > rect.height) { Core.rectangle(img, new Point(rect.x, rect.y), new Point(rect.x + rect.width, rect.y + rect.height), new Scalar(0, 0, 255)); }/*w ww . ja va 2s.c om*/ } } //imshow("Rec Detected", img); }
From source file:OctoEye.java
License:Open Source License
private void detectPupil() { // min and max pupil radius int r_min = 2; int r_max = 45; // min and max pupil diameter int d_min = 2 * r_min; int d_max = 2 * r_max; // min and max pupil area double area;//from ww w .j a v a 2 s .c o m double a_min = Math.PI * r_min * r_min; double a_max = Math.PI * r_max * r_max; // histogram stuff List<Mat> images; MatOfInt channels; Mat mask; Mat hist; MatOfInt mHistSize; MatOfFloat mRanges; // contour and circle stuff Rect rect = null; Rect rectMin; Rect rectMax; List<MatOfPoint> contours; MatOfPoint3 circles; // pupil center Point p; // ellipse test points Point v; Point r; Point s; // rect points Point tl; Point br; // pupil edge detection Vector<Point> pointsTest; Vector<Point> pointsEllipse; Vector<Point> pointsRemoved; // temporary variables double distance; double rad; double length; int x; int y; int tmp; byte buff[]; // ------------------------------------------------------------------------------------------------------------- // step 1 // blur the image to reduce noise Imgproc.medianBlur(src, tmp1, 25); // ------------------------------------------------------------------------------------------------------------- // step 2 // locate the pupil with feature detection and compute a histogram for each, // the best feature will be used as rough pupil location (rectMin) int score = 0; int winner = 0; // feature detection MatOfKeyPoint matOfKeyPoints = new MatOfKeyPoint(); FeatureDetector blobDetector = FeatureDetector.create(FeatureDetector.MSER); // Maximal Stable Extremal Regions blobDetector.detect(tmp1, matOfKeyPoints); List<KeyPoint> keyPoints = matOfKeyPoints.toList(); // histogram calculation for (int i = 0; i < keyPoints.size(); i++) { x = (int) keyPoints.get(i).pt.x; y = (int) keyPoints.get(i).pt.y; tl = new Point(x - 5 >= 0 ? x - 5 : 0, y - 5 >= 0 ? y - 5 : 0); br = new Point(x + 5 < WIDTH ? x + 5 : WIDTH - 1, y + 5 < HEIGHT ? y + 5 : HEIGHT - 1); images = new ArrayList<Mat>(); images.add(tmp1.submat(new Rect(tl, br))); channels = new MatOfInt(0); mask = new Mat(); hist = new Mat(); mHistSize = new MatOfInt(256); mRanges = new MatOfFloat(0f, 256f); Imgproc.calcHist(images, channels, mask, hist, mHistSize, mRanges); tmp = 0; for (int j = 0; j < 256 / 3; j++) { tmp += (256 / 3 - j) * (int) hist.get(j, 0)[0]; } if (tmp >= score) { score = tmp; winner = i; rect = new Rect(tl, br); } if (debug) { // show features (orange) Core.circle(dbg, new Point(x, y), 3, ORANGE); } } if (rect == null) { return; } rectMin = rect.clone(); if (debug) { // show rectMin (red) Core.rectangle(dbg, rectMin.tl(), rect.br(), RED, 1); } // ------------------------------------------------------------------------------------------------------------- // step 3 // compute a rectMax (blue) which is larger than the pupil int margin = 32; rect.x = rect.x - margin; rect.y = rect.y - margin; rect.width = rect.width + 2 * margin; rect.height = rect.height + 2 * margin; rectMax = rect.clone(); if (debug) { // show features (orange) Core.rectangle(dbg, rectMax.tl(), rectMax.br(), BLUE); } // ------------------------------------------------------------------------------------------------------------- // step 4 // blur the image again Imgproc.medianBlur(src, tmp1, 7); Imgproc.medianBlur(tmp1, tmp1, 3); Imgproc.medianBlur(tmp1, tmp1, 3); Imgproc.medianBlur(tmp1, tmp1, 3); // ------------------------------------------------------------------------------------------------------------- // step 5 // detect edges Imgproc.Canny(tmp1, tmp2, 40, 50); // ------------------------------------------------------------------------------------------------------------- // step 6 // from pupil center to maxRect borders, find all edge points, compute a first ellipse p = new Point(rectMin.x + rectMin.width / 2, rectMin.y + rectMin.height / 2); pointsTest = new Vector<Point>(); pointsEllipse = new Vector<Point>(); pointsRemoved = new Vector<Point>(); buff = new byte[tmp2.rows() * tmp2.cols()]; tmp2.get(0, 0, buff); length = Math.min(p.x - rectMax.x - 3, p.y - rectMax.y - 3); length = Math.sqrt(2 * Math.pow(length, 2)); Point z = new Point(p.x, p.y - length); for (int i = 0; i < 360; i += 15) { rad = Math.toRadians(i); x = (int) (p.x + Math.cos(rad) * (z.x - p.x) - Math.sin(rad) * (z.y - p.y)); y = (int) (p.y + Math.sin(rad) * (z.x - p.x) - Math.cos(rad) * (z.y - p.y)); pointsTest.add(new Point(x, y)); } if (debug) { for (int i = 0; i < pointsTest.size(); i++) { Core.line(dbg, p, pointsTest.get(i), GRAY, 1); Core.rectangle(dbg, rectMin.tl(), rectMin.br(), GREEN, 1); Core.rectangle(dbg, rectMax.tl(), rectMax.br(), BLUE, 1); } Core.rectangle(dbg, rectMin.tl(), rectMin.br(), BLACK, -1); Core.rectangle(dbg, rectMin.tl(), rectMin.br(), RED, 1); Core.rectangle(dbg, rectMax.tl(), rectMax.br(), BLUE); } // p: Ursprung ("Mittelpunkt" der Ellipse) // v: Zielpunkt (Testpunkt) // r: Richtungsvektor PV for (int i = 0; i < pointsTest.size(); i++) { v = new Point(pointsTest.get(i).x, pointsTest.get(i).y); r = new Point(v.x - p.x, v.y - p.y); length = Math.sqrt(Math.pow(p.x - v.x, 2) + Math.pow(p.y - v.y, 2)); boolean found = false; for (int j = 0; j < Math.round(length); j++) { s = new Point(Math.rint(p.x + (double) j / length * r.x), Math.rint(p.y + (double) j / length * r.y)); s.x = Math.max(1, Math.min(s.x, WIDTH - 2)); s.y = Math.max(1, Math.min(s.y, HEIGHT - 2)); tl = new Point(s.x - 1, s.y - 1); br = new Point(s.x + 1, s.y + 1); buff = new byte[3 * 3]; rect = new Rect(tl, br); try { (tmp2.submat(rect)).get(0, 0, buff); for (int k = 0; k < 3 * 3; k++) { if (Math.abs(buff[k]) == 1) { pointsEllipse.add(s); found = true; break; } } } catch (Exception e) { break; } if (found) { break; } } } double e_min = Double.POSITIVE_INFINITY; double e_max = 0; double e_med = 0; for (int i = 0; i < pointsEllipse.size(); i++) { v = pointsEllipse.get(i); length = Math.sqrt(Math.pow(p.x - v.x, 2) + Math.pow(p.y - v.y, 2)); e_min = (length < e_min) ? length : e_min; e_max = (length > e_max) ? length : e_max; e_med = e_med + length; } e_med = e_med / pointsEllipse.size(); if (pointsEllipse.size() >= 5) { Point[] points1 = new Point[pointsEllipse.size()]; for (int i = 0; i < pointsEllipse.size(); i++) { points1[i] = pointsEllipse.get(i); } MatOfPoint2f points2 = new MatOfPoint2f(); points2.fromArray(points1); pupil = Imgproc.fitEllipse(points2); } if (pupil.center.x == 0 && pupil.center.y == 0) { // something went wrong, return null reset(); return; } if (debug) { Core.ellipse(dbg, pupil, PURPLE, 2); } // ------------------------------------------------------------------------------------------------------------- // step 7 // remove some outlier points and compute the ellipse again try { for (int i = 1; i <= 4; i++) { distance = 0; int remove = 0; for (int j = pointsEllipse.size() - 1; j >= 0; j--) { v = pointsEllipse.get(j); length = Math.sqrt(Math.pow(v.x - pupil.center.x, 2) + Math.pow(v.y - pupil.center.y, 2)); if (length > distance) { distance = length; remove = j; } } v = pointsEllipse.get(remove); pointsEllipse.removeElementAt(remove); pointsRemoved.add(v); } } catch (Exception e) { // something went wrong, return null reset(); return; } if (pointsEllipse.size() >= 5) { Point[] points1 = new Point[pointsEllipse.size()]; for (int i = 0; i < pointsEllipse.size(); i++) { points1[i] = pointsEllipse.get(i); } MatOfPoint2f points2 = new MatOfPoint2f(); points2.fromArray(points1); pupil = Imgproc.fitEllipse(points2); Point[] vertices = new Point[4]; pupil.points(vertices); double d1 = Math .sqrt(Math.pow(vertices[1].x - vertices[0].x, 2) + Math.pow(vertices[1].y - vertices[0].y, 2)); double d2 = Math .sqrt(Math.pow(vertices[2].x - vertices[1].x, 2) + Math.pow(vertices[2].y - vertices[1].y, 2)); if (d1 >= d2) { pupilMajorAxis = (int) (d1 / 2); pupilMinorAxis = (int) (d2 / 2); axisA = new Point(vertices[1].x + (vertices[2].x - vertices[1].x) / 2, vertices[1].y + (vertices[2].y - vertices[1].y) / 2); axisB = new Point(vertices[0].x + (vertices[1].x - vertices[0].x) / 2, vertices[0].y + (vertices[1].y - vertices[0].y) / 2); } else { pupilMajorAxis = (int) (d2 / 2); pupilMinorAxis = (int) (d1 / 2); axisB = new Point(vertices[1].x + (vertices[2].x - vertices[1].x) / 2, vertices[1].y + (vertices[2].y - vertices[1].y) / 2); axisA = new Point(vertices[0].x + (vertices[1].x - vertices[0].x) / 2, vertices[0].y + (vertices[1].y - vertices[0].y) / 2); } } double ratio = (double) pupilMinorAxis / (double) pupilMajorAxis; if (ratio < 0.75 || 2 * pupilMinorAxis <= d_min || 2 * pupilMajorAxis >= d_max) { // something went wrong, return null reset(); return; } // pupil found if (debug) { Core.ellipse(dbg, pupil, GREEN, 2); Core.line(dbg, pupil.center, axisA, RED, 2); Core.line(dbg, pupil.center, axisB, BLUE, 2); Core.circle(dbg, pupil.center, 1, GREEN, 0); x = 5; y = 5; Core.rectangle(dbg, new Point(x, y), new Point(x + 80 + 4, y + 10), BLACK, -1); Core.rectangle(dbg, new Point(x + 2, y + 2), new Point(x + 2 + pupilMajorAxis, y + 4), RED, -1); Core.rectangle(dbg, new Point(x + 2, y + 6), new Point(x + 2 + pupilMinorAxis, y + 8), BLUE, -1); for (int i = pointsEllipse.size() - 1; i >= 0; i--) { Core.circle(dbg, pointsEllipse.get(i), 2, ORANGE, -1); } for (int i = pointsRemoved.size() - 1; i >= 0; i--) { Core.circle(dbg, pointsRemoved.get(i), 2, PURPLE, -1); } } Core.ellipse(dst, pupil, GREEN, 2); Core.circle(dst, pupil.center, 1, GREEN, 0); }
From source file:OCV_MinAreaRect.java
License:Open Source License
@Override public void run(ImageProcessor ip) { byte[] byteArray = (byte[]) ip.getPixels(); int w = ip.getWidth(); int h = ip.getHeight(); int num_slice = ip.getSliceNumber(); ArrayList<Point> lstPt = new ArrayList<Point>(); MatOfPoint2f pts = new MatOfPoint2f(); for (int y = 0; y < h; y++) { for (int x = 0; x < w; x++) { if (byteArray[x + w * y] != 0) { lstPt.add(new Point((double) x, (double) y)); }//from ww w.j av a2 s .co m } } if (lstPt.isEmpty()) { return; } pts.fromList(lstPt); RotatedRect rect = Imgproc.minAreaRect(pts); showData(rect, num_slice); }
From source file:OCV_MinEnclosingCircle.java
License:Open Source License
@Override public void run(ImageProcessor ip) { byte[] byteArray = (byte[]) ip.getPixels(); int w = ip.getWidth(); int h = ip.getHeight(); int num_slice = ip.getSliceNumber(); ArrayList<Point> lstPt = new ArrayList<Point>(); MatOfPoint2f pts = new MatOfPoint2f(); for (int y = 0; y < h; y++) { for (int x = 0; x < w; x++) { if (byteArray[x + w * y] != 0) { lstPt.add(new Point((double) x, (double) y)); }//from www . j a v a2 s. c o m } } if (lstPt.isEmpty()) { return; } pts.fromList(lstPt); float[] radius = new float[1]; Point center = new Point(); Imgproc.minEnclosingCircle(pts, center, radius); showData(center.x, center.y, (double) radius[0], num_slice); }
From source file:OCV_GetPerspectiveTransform.java
License:Open Source License
@Override public void run(ImageProcessor ip) { MatOfPoint2f matPt_src = new MatOfPoint2f(); MatOfPoint2f matPt_dst = new MatOfPoint2f(); matPt_src.fromList(lstPt_src);/*from www . j a va 2 s . c o m*/ matPt_dst.fromList(lstPt_dst); Mat mat = Imgproc.getPerspectiveTransform(matPt_src, matPt_dst); if (mat == null || mat.rows() <= 0 || mat.cols() <= 0) { IJ.showMessage("Output is null or error"); return; } ResultsTable rt = OCV__LoadLibrary.GetResultsTable(true); for (int i = 0; i < 3; i++) { rt.incrementCounter(); rt.addValue("Column01", String.valueOf(mat.get(i, 0)[0])); rt.addValue("Column02", String.valueOf(mat.get(i, 1)[0])); rt.addValue("Column03", String.valueOf(mat.get(i, 2)[0])); } rt.show("Results"); }
From source file:OCV_FitEllipse.java
License:Open Source License
@Override public void run(ImageProcessor ip) { byte[] byteArray = (byte[]) ip.getPixels(); int w = ip.getWidth(); int h = ip.getHeight(); int num_slice = ip.getSliceNumber(); ArrayList<Point> lstPt = new ArrayList<Point>(); MatOfPoint2f pts = new MatOfPoint2f(); for (int y = 0; y < h; y++) { for (int x = 0; x < w; x++) { if (byteArray[x + w * y] != 0) { lstPt.add(new Point((double) x, (double) y)); }//from ww w. j a v a 2s . c o m } } if (lstPt.isEmpty()) { return; } pts.fromList(lstPt); RotatedRect rect = Imgproc.fitEllipse(pts); showData(rect, num_slice); }
From source file:OCV_GetAffineTransform.java
License:Open Source License
@Override public void run(ImageProcessor ip) { MatOfPoint2f matPt_src = new MatOfPoint2f(); MatOfPoint2f matPt_dst = new MatOfPoint2f(); matPt_src.fromList(lstPt_src);/*from w w w .jav a2 s . c om*/ matPt_dst.fromList(lstPt_dst); Mat mat = Imgproc.getAffineTransform(matPt_src, matPt_dst); if (mat == null || mat.rows() <= 0 || mat.cols() <= 0) { IJ.showMessage("Output is null or error"); return; } ResultsTable rt = OCV__LoadLibrary.GetResultsTable(true); for (int i = 0; i < 2; i++) { rt.incrementCounter(); rt.addValue("Column01", String.valueOf(mat.get(i, 0)[0])); rt.addValue("Column02", String.valueOf(mat.get(i, 1)[0])); rt.addValue("Column03", String.valueOf(mat.get(i, 2)[0])); } rt.show("Results"); }
From source file:arlocros.ComputePose.java
License:Apache License
public boolean computePose(Mat rvec, Mat tvec, Mat image2) throws NyARException, FileNotFoundException { // convert image to NyAR style for processing final INyARRgbRaster imageRaster = NyARImageHelper.createFromMat(image2); // create new marker system configuration i_config = new NyARMarkerSystemConfig(i_param); markerSystemState = new NyARMarkerSystem(i_config); // Create wrapper that passes cam pictures to marker system cameraSensorWrapper = new NyARSensor(i_screen_size); ids = new int[markerPatterns.size()]; patternmap = new HashMap<>(); for (int i = 0; i < markerPatterns.size(); i++) { // create marker description from pattern file and add to marker // system ids[i] = markerSystemState.addARMarker(arCodes.get(i), 25, markerConfig.getMarkerSize()); patternmap.put(ids[i], markerPatterns.get(i)); }/*w w w .jav a2 s . c om*/ cameraSensorWrapper.update(imageRaster); markerSystemState.update(cameraSensorWrapper); // init 3D point list final List<Point3> points3dlist = new ArrayList<>(); final List<Point> points2dlist = new ArrayList<>(); for (final int id : ids) { // process only if this marker has been detected if (markerSystemState.isExistMarker(id) && markerSystemState.getConfidence(id) > 0.7) { // read and add 2D points final NyARIntPoint2d[] vertex2d = markerSystemState.getMarkerVertex2D(id); Point p = new Point(vertex2d[0].x, vertex2d[0].y); points2dlist.add(p); p = new Point(vertex2d[1].x, vertex2d[2].y); points2dlist.add(p); p = new Point(vertex2d[2].x, vertex2d[2].y); points2dlist.add(p); p = new Point(vertex2d[3].x, vertex2d[3].y); points2dlist.add(p); // final MatOfPoint mop = new MatOfPoint(); // mop.fromList(points2dlist); // final List<MatOfPoint> pts = new ArrayList<>(); // pts.add(mop); // read and add corresponding 3D points points3dlist.addAll(markerConfig.create3dpointlist(patternmap.get(id))); if (visualization) { // draw red rectangle around detected marker Core.rectangle(image2, new Point(vertex2d[0].x, vertex2d[0].y), new Point(vertex2d[2].x, vertex2d[2].y), new Scalar(0, 0, 255)); final String markerFile = patternmap.get(id).replaceAll(".*4x4_", "").replace(".patt", ""); Core.putText(image2, markerFile, new Point((vertex2d[2].x + vertex2d[0].x) / 2.0, vertex2d[0].y - 5), 4, 1, new Scalar(250, 0, 0)); } } } // load 2D and 3D points to Mats for solvePNP final MatOfPoint3f objectPoints = new MatOfPoint3f(); objectPoints.fromList(points3dlist); final MatOfPoint2f imagePoints = new MatOfPoint2f(); imagePoints.fromList(points2dlist); if (visualization) { // show image with markers detected Imshow.show(image2); } // do not call solvePNP with empty intput data (no markers detected) if (points2dlist.size() == 0) { objectPoints.release(); imagePoints.release(); return false; } // uncomment these lines if using RANSAC-based pose estimation (more // shaking) Mat inliers = new Mat(); Calib3d.solvePnPRansac(objectPoints, imagePoints, cameraMatrix, distCoeffs, rvec, tvec, false, 300, 5, 16, inliers, Calib3d.CV_P3P); ArMarkerPoseEstimator.getLog() .info("Points detected: " + points2dlist.size() + " inliers: " + inliers.size()); objectPoints.release(); imagePoints.release(); // avoid publish zero pose if localization failed if (inliers.rows() == 0) { inliers.release(); return false; } inliers.release(); return true; }
From source file:com.android.cts.verifier.sensors.RVCVXCheckAnalyzer.java
License:Apache License
/** * Analyze video frames using computer vision approach and generate a ArrayList<AttitudeRec> * * @param recs output ArrayList of AttitudeRec * @return total number of frame of the video *///w w w.j a v a2s . c o m private int analyzeVideo(ArrayList<AttitudeRec> recs) { VideoMetaInfo meta = new VideoMetaInfo(new File(mPath, "videometa.json")); int decimation = 1; boolean use_timestamp = true; // roughly determine if decimation is necessary if (meta.fps > DECIMATION_FPS_TARGET) { decimation = (int) (meta.fps / DECIMATION_FPS_TARGET); meta.fps /= decimation; } VideoDecoderForOpenCV videoDecoder = new VideoDecoderForOpenCV(new File(mPath, "video.mp4"), decimation); Mat frame; Mat gray = new Mat(); int i = -1; Size frameSize = videoDecoder.getSize(); if (frameSize.width != meta.frameWidth || frameSize.height != meta.frameHeight) { // this is very unlikely return -1; } if (TRACE_VIDEO_ANALYSIS) { Debug.startMethodTracing("cvprocess"); } Size patternSize = new Size(4, 11); float fc = (float) (meta.frameWidth / 2.0 / Math.tan(meta.fovWidth / 2.0)); Mat camMat = cameraMatrix(fc, new Size(frameSize.width / 2, frameSize.height / 2)); MatOfDouble coeff = new MatOfDouble(); // dummy MatOfPoint2f centers = new MatOfPoint2f(); MatOfPoint3f grid = asymmetricalCircleGrid(patternSize); Mat rvec = new MatOfFloat(); Mat tvec = new MatOfFloat(); MatOfPoint2f reprojCenters = new MatOfPoint2f(); if (LOCAL_LOGV) { Log.v(TAG, "Camera Mat = \n" + camMat.dump()); } long startTime = System.nanoTime(); long[] ts = new long[1]; while ((frame = videoDecoder.getFrame(ts)) != null) { if (LOCAL_LOGV) { Log.v(TAG, "got a frame " + i); } if (use_timestamp && ts[0] == -1) { use_timestamp = false; } // has to be in front, as there are cases where execution // will skip the later part of this while i++; // convert to gray manually as by default findCirclesGridDefault uses COLOR_BGR2GRAY Imgproc.cvtColor(frame, gray, Imgproc.COLOR_RGB2GRAY); boolean foundPattern = Calib3d.findCirclesGrid(gray, patternSize, centers, Calib3d.CALIB_CB_ASYMMETRIC_GRID); if (!foundPattern) { // skip to next frame continue; } if (OUTPUT_DEBUG_IMAGE) { Calib3d.drawChessboardCorners(frame, patternSize, centers, true); } // figure out the extrinsic parameters using real ground truth 3D points and the pixel // position of blobs found in findCircleGrid, an estimated camera matrix and // no-distortion are assumed. boolean foundSolution = Calib3d.solvePnP(grid, centers, camMat, coeff, rvec, tvec, false, Calib3d.CV_ITERATIVE); if (!foundSolution) { // skip to next frame if (LOCAL_LOGV) { Log.v(TAG, "cannot find pnp solution in frame " + i + ", skipped."); } continue; } // reproject points to for evaluation of result accuracy of solvePnP Calib3d.projectPoints(grid, rvec, tvec, camMat, coeff, reprojCenters); // error is evaluated in norm2, which is real error in pixel distance / sqrt(2) double error = Core.norm(centers, reprojCenters, Core.NORM_L2); if (LOCAL_LOGV) { Log.v(TAG, "Found attitude, re-projection error = " + error); } // if error is reasonable, add it into the results. use ratio to frame height to avoid // discriminating higher definition videos if (error < REPROJECTION_THREASHOLD_RATIO * frameSize.height) { double[] rv = new double[3]; double timestamp; rvec.get(0, 0, rv); if (use_timestamp) { timestamp = (double) ts[0] / 1e6; } else { timestamp = (double) i / meta.fps; } if (LOCAL_LOGV) Log.v(TAG, String.format("Added frame %d ts = %f", i, timestamp)); recs.add(new AttitudeRec(timestamp, rodr2rpy(rv))); } if (OUTPUT_DEBUG_IMAGE) { Calib3d.drawChessboardCorners(frame, patternSize, reprojCenters, true); Imgcodecs.imwrite(Environment.getExternalStorageDirectory().getPath() + "/RVCVRecData/DebugCV/img" + i + ".png", frame); } } if (LOCAL_LOGV) { Log.v(TAG, "Finished decoding"); } if (TRACE_VIDEO_ANALYSIS) { Debug.stopMethodTracing(); } if (LOCAL_LOGV) { // time analysis double totalTime = (System.nanoTime() - startTime) / 1e9; Log.i(TAG, "Total time: " + totalTime + "s, Per frame time: " + totalTime / i); } return i; }