public void templateMatching() {
System.loadLibrary(Core.NATIVE_LIBRARY_NAME);
int match_method = 5;
int max_Trackbar = 5;
Mat data = Highgui.imread("images/training_data/1" + "/data (" + 1 + ").jpg");
Mat temp = Highgui.imread("images/template.jpg");
Mat img = data.clone();
int result_cols = img.cols() - temp.cols() + 1;
int result_rows = img.rows() - temp.rows() + 1;
Mat result = new Mat(result_rows, result_cols, CvType.CV_32FC1);
Imgproc.matchTemplate(img, temp, result, match_method);
Core.normalize(result, result, 0, 1, Core.NORM_MINMAX, -1, new Mat());
double minVal;
double maxVal;
Point minLoc;
Point maxLoc;
Point matchLoc;
// minMaxLoc( result, &minVal, &maxVal, &minLoc, &maxLoc, Mat() );
Core.MinMaxLocResult res = Core.minMaxLoc(result);
if (match_method == Imgproc.TM_SQDIFF || match_method == Imgproc.TM_SQDIFF_NORMED) {
matchLoc = res.minLoc;
} else {
matchLoc = res.maxLoc;
}
// / Show me what you got
Core.rectangle(
img,
matchLoc,
new Point(matchLoc.x + temp.cols(), matchLoc.y + temp.rows()),
new Scalar(0, 255, 0));
// Save the visualized detection.
Highgui.imwrite("images/samp.jpg", img);
}
// Transform the json-type feature to mat-type
public static Mat json2mat(String json) {
JsonParser parser = new JsonParser();
JsonElement parseTree = parser.parse(json);
// Verify the input is JSON type
if (!parseTree.isJsonObject()) {
System.out.println("The input is not a JSON type...\nExiting...");
System.exit(1);
}
JsonObject jobj = parser.parse(json).getAsJsonObject();
if (jobj == null || !jobj.isJsonObject() || jobj.isJsonNull()) {
return null;
}
// Detect broken/null features
JsonElement r = jobj.get("rows");
if (r == null) {
return null;
}
int rows = jobj.get("rows").getAsInt();
int cols = jobj.get("cols").getAsInt();
int type = jobj.get("type").getAsInt();
String data = jobj.get("data").getAsString();
String[] pixs = data.split(",");
Mat descriptor = new Mat(rows, cols, type);
for (String pix : pixs) {
String[] tmp = pix.split(" ");
int r_pos = Integer.valueOf(tmp[0]);
int c_pos = Integer.valueOf(tmp[1]);
double rgb = Double.valueOf(tmp[2]);
descriptor.put(r_pos, c_pos, rgb);
}
return descriptor;
}
public static void extractQueryFeatures2HDFS(String filename, Job job) throws IOException {
// Read the local image.jpg as a Mat
Mat query_mat_float =
Highgui.imread(LOCAL_USER_DIR + ID + INPUT + "/" + filename, CvType.CV_32FC3);
// Convert RGB to GRAY
Mat query_gray = new Mat();
Imgproc.cvtColor(query_mat_float, query_gray, Imgproc.COLOR_RGB2GRAY);
// Convert the float type to unsigned integer(required by SIFT)
Mat query_mat_byte = new Mat();
query_gray.convertTo(query_mat_byte, CvType.CV_8UC3);
// // Resize the image to 1/FACTOR both width and height
// Mat query_mat_byte = FeatureExtraction.resize(query_mat_byte);
// Extract the feature from the (Mat)image
Mat query_features = FeatureExtraction.extractFeature(query_mat_byte);
System.out.println(PREFIX + "Extracting the query image feature...");
System.out.println("query_mat(float,color):" + query_mat_float);
System.out.println("query_mat(float,gray):" + query_gray);
System.out.println("query_mat(byte,gray):" + query_mat_byte);
System.out.println("query_mat_features:" + query_features);
System.out.println();
// Store the feature to the hdfs in order to use it later in different map tasks
System.out.println(PREFIX + "Generating the feature file for the query image in HDFS...");
FileSystem fs = FileSystem.get(job.getConfiguration());
String featureFileName = filename.substring(0, filename.lastIndexOf(".")) + ".json";
FSDataOutputStream fsDataOutputStream =
fs.create(new Path(HDFS_HOME + USER + ID + INPUT + "/" + featureFileName));
BufferedWriter bw =
new BufferedWriter(new OutputStreamWriter(fsDataOutputStream, StandardCharsets.UTF_8));
bw.write(FeatureExtraction.mat2json(query_features));
bw.close();
System.out.println(PREFIX + "Query feature extraction finished...");
System.out.println();
}
public void copyMat(Mat src, Mat dest) {
int srcRows = src.rows();
int srcCols = src.cols();
int destRows = dest.rows();
int destCols = dest.cols();
for (int i = 0; i < srcRows; i++) {
for (int j = 0; j < srcCols; j++) {
double bit = src.get(i, j)[0];
dest.put(i, j, bit);
System.out.println(bit);
}
}
}
public static Mat getCCH(Mat image) {
ArrayList<MatOfPoint> contours = new ArrayList<MatOfPoint>();
Mat hierarchy = new Mat();
Imgproc.findContours(
image, contours, hierarchy, Imgproc.RETR_EXTERNAL, Imgproc.CHAIN_APPROX_NONE);
Mat chainHistogram = Mat.zeros(1, 8, CvType.CV_32F);
int n = 0;
MatOfPoint2f approxCurve = new MatOfPoint2f();
for (MatOfPoint contour : contours) {
// get the freeman chain code from the contours
int rows = contour.rows();
// System.out.println("\nrows"+rows+"\n"+contour.dump());
int direction = 7;
Mat prevPoint = contours.get(0).row(0);
n += rows - 1;
for (int i = 1; i < rows; i++) {
// get the current point
double x1 = contour.get(i - 1, 0)[1];
double y1 = contour.get(i - 1, 0)[0];
// get the second point
double x2 = contour.get(i, 0)[1];
double y2 = contour.get(i, 0)[0];
if (x2 == x1 && y2 == y1 + 1) direction = 0;
else if (x2 == x1 - 1 && y2 == y1 + 1) direction = 1;
else if (x2 == x1 - 1 && y2 == y1) direction = 2;
else if (x2 == x1 - 1 && y2 == y1 - 1) direction = 3;
else if (x2 == x1 && y2 == y1 - 1) direction = 4;
else if (x2 == x1 + 1 && y2 == y1 - 1) direction = 5;
else if (x2 == x1 + 1 && y2 == y1) direction = 6;
else if (x2 == x1 + 1 && y2 == y1 + 1) direction = 7;
else System.out.print("err");
double counter = chainHistogram.get(0, direction)[0];
chainHistogram.put(0, direction, ++counter);
System.out.print(direction);
}
}
System.out.println("\n" + chainHistogram.dump());
Scalar alpha = new Scalar(n); // the factor
Core.divide(chainHistogram, alpha, chainHistogram);
System.out.println("\nrows=" + n + " " + chainHistogram.dump());
return chainHistogram;
}
public void map(Text key, Text value, Context context)
throws InterruptedException, IOException {
String filename = key.toString();
String json = value.toString();
// Make sure the input is valid
if (!(filename.isEmpty() || json.isEmpty())) {
// Change the json-type feature to Mat-type feature
Mat descriptor = json2mat(json);
if (descriptor != null) {
// Read the query feature from the cache in Hadoop
Mat query_features;
String pathStr = context.getConfiguration().get("featureFilePath");
FileSystem fs = FileSystem.get(context.getConfiguration());
FSDataInputStream fsDataInputStream = fs.open(new Path(pathStr));
StringBuilder sb = new StringBuilder();
// Use a buffer to read the query_feature
int remain = fsDataInputStream.available();
while (remain > 0) {
int read;
byte[] buf = new byte[BUF_SIZE];
read = fsDataInputStream.read(buf, fsDataInputStream.available() - remain, BUF_SIZE);
sb.append(new String(buf, 0, read, StandardCharsets.UTF_8));
remain = remain - read;
System.out.println("remain:" + remain + "\tread:" + read + "\tsb.size:" + sb.length());
}
// Read the query_feature line by line
// Scanner sc = new Scanner(fsDataInputStream, "UTF-8");
// StringBuilder sb = new StringBuilder();
// while (sc.hasNextLine()) {
// sb.append(sc.nextLine());
// }
// String query_json = sb.toString();
// String query_json = new String(buf, StandardCharsets.UTF_8);
String query_json = sb.toString();
fsDataInputStream.close();
query_features = json2mat(query_json);
// Get the similarity of the current database image against the query image
DescriptorMatcher matcher = DescriptorMatcher.create(DescriptorMatcher.FLANNBASED);
MatOfDMatch matches = new MatOfDMatch();
// Ensure the two features have same length of cols (the feature extracted are all 128
// cols(at least in this case))
if (query_features.cols() == descriptor.cols()) {
matcher.match(query_features, descriptor, matches);
DMatch[] dMatches = matches.toArray();
// Calculate the max/min distances
// double max_dist = Double.MAX_VALUE;
// double min_dist = Double.MIN_VALUE;
double max_dist = 0;
double min_dist = 100;
for (int i = 0; i < dMatches.length; i++) {
double dist = dMatches[i].distance;
if (min_dist > dist) min_dist = dist;
if (max_dist < dist) max_dist = dist;
}
// Only distances ≤ threshold are good matches
double threshold = max_dist * THRESHOLD_FACTOR;
// double threshold = min_dist * 2;
LinkedList<DMatch> goodMatches = new LinkedList<DMatch>();
for (int i = 0; i < dMatches.length; i++) {
if (dMatches[i].distance <= threshold) {
goodMatches.addLast(dMatches[i]);
}
}
// Get the ratio of good_matches to all_matches
double ratio = (double) goodMatches.size() / (double) dMatches.length;
System.out.println("*** current_record_filename:" + filename + " ***");
System.out.println("feature:" + descriptor + "\nquery_feature:" + query_features);
System.out.println(
"min_dist of keypoints:" + min_dist + " max_dist of keypoints:" + max_dist);
System.out.println(
"total_matches:" + dMatches.length + "\tgood_matches:" + goodMatches.size());
// System.out.println("type:" + descriptor.type() + " channels:" +
// descriptor.channels() + " rows:" + descriptor.rows() + " cols:" + descriptor.cols());
// System.out.println("qtype:" + query_features.type() + "
// qchannels:" + query_features.channels() + " qrows:" + query_features.rows() + "
// qcols:" + query_features.cols());
System.out.println();
if (ratio > PERCENTAGE_THRESHOLD) {
// Key:1 Value:filename|ratio
context.write(ONE, new Text(filename + "|" + ratio));
// context.write(ONE, new Text(filename + "|" +
// String.valueOf(goodMatches.size())));
}
} else {
System.out.println("The size of the features are not equal");
}
} else {
// a null pointer, do nothing
System.out.println("A broken/null feature:" + filename);
System.out.println();
}
}
}
public static void main(String[] args) {
System.loadLibrary(Core.NATIVE_LIBRARY_NAME);
// Mat mat = Mat.eye( 3, 3, CvType.CV_8UC1 );
// System.out.println( "mat = " + mat.dump() );
Sample n = new Sample();
// n.templateMatching();
// put text in image
// Mat data= Highgui.imread("images/erosion.jpg");
// Core.putText(data, "Sample", new Point(50,80), Core.FONT_HERSHEY_SIMPLEX, 1, new
// Scalar(0,0,0),2);
//
// Highgui.imwrite("images/erosion2.jpg", data);
// getting dct of an image
String path = "images/croppedfeature/go (20).jpg";
path = "images/wordseg/img1.png";
Mat image = Highgui.imread(path, Highgui.IMREAD_GRAYSCALE);
ArrayList<MatOfPoint> contours = new ArrayList<MatOfPoint>();
Imgproc.threshold(image, image, 0, 255, Imgproc.THRESH_OTSU);
Imgproc.threshold(image, image, 220, 128, Imgproc.THRESH_BINARY_INV);
Mat newImg = new Mat(45, 100, image.type());
newImg.setTo(new Scalar(0));
n.copyMat(image, newImg);
int vgap = 25;
int hgap = 45 / 3;
Moments m = Imgproc.moments(image, false);
Mat hu = new Mat();
Imgproc.HuMoments(m, hu);
System.out.println(hu.dump());
// //divide the mat into 12 parts then get the features of each part
// int count=1;
// for(int j=0; j<45; j+=hgap){
// for(int i=0;i<100;i+=vgap){
// Mat result = newImg.submat(j, j+hgap, i, i+vgap);
//
//
// Moments m= Imgproc.moments(result, false);
// double m01= m.get_m01();
// double m00= m.get_m00();
// double m10 = m.get_m10();
// int x= m00!=0? (int)(m10/m00):0;
// int y= m00!=0? (int)(m01/m00):0;
// Mat hu= new Mat();
// Imgproc.HuMoments(m, hu);
// System.out.println(hu.dump());
// System.out.println(count+" :"+x+" and "+y);
// Imgproc.threshold(result, result, 0,254, Imgproc.THRESH_BINARY_INV);
// Highgui.imwrite("images/submat/"+count+".jpg", result);
// count++;
//
// }
// }
//
// for(int i=vgap;i<100;i+=vgap){
// Point pt1= new Point(i, 0);
// Point pt2= new Point(i, 99);
// Core.line(newImg, pt1, pt2, new Scalar(0,0,0));
// }
// for(int i=hgap;i<45;i+=hgap){
// Point pt1= new Point(0, i);
// Point pt2= new Point(99, i);
// Core.line(newImg, pt1, pt2, new Scalar(0,0,0));
// }
// Highgui.imwrite("images/submat/copyto.jpg", newImg);
}