Develop Reference

Java - extract text from pdf from selected area to txt

The idea is next,
user selects a pdf file, and then this file converted into an image and such an image is displayed in the application.
In the image the user can choose positions that wants to read from a pdf file, and when the finish with selection position in the background program reads the original pdf and text stored in a txt file.
It is important that the resulting image from pdf file is the same size as himself pdf file
The next code convert pdf to image. I use pdfrenderer-0.9.1.jar
import java.awt.Rectangle;
import java.awt.image.BufferedImage;
import java.io.File;
import java.io.FileNotFoundException;
import java.io.IOException;
import java.io.RandomAccessFile;
import java.nio.ByteBuffer;
import java.nio.channels.FileChannel;
import javax.imageio.ImageIO;
import com.sun.pdfview.PDFFile;
import com.sun.pdfview.PDFPage;
public class Pdf2Image {
public static void main(String[] args) {
File file = new File("E:\\invoice-template-1.pdf");
RandomAccessFile raf;
try {
raf = new RandomAccessFile(file, "r");
FileChannel channel = raf.getChannel();
ByteBuffer buf = channel.map(FileChannel.MapMode.READ_ONLY, 0, channel.size());
PDFFile pdffile = new PDFFile(buf);
// draw the first page to an image
int num=pdffile.getNumPages();
for(int i=0;i<num;i++)
{
PDFPage page = pdffile.getPage(i);
//get the width and height for the doc at the default zoom
int width=(int)page.getBBox().getWidth();
int height=(int)page.getBBox().getHeight();
Rectangle rect = new Rectangle(0,0,width,height);
int rotation=page.getRotation();
Rectangle rect1=rect;
if(rotation==90 || rotation==270)
rect1=new Rectangle(0,0,rect.height,rect.width);
//generate the image
BufferedImage img = (BufferedImage)page.getImage(
rect.width, rect.height, //width & height
rect1, // clip rect
null, // null for the ImageObserver
true, // fill background with white
true // block until drawing is done
);
ImageIO.write(img, "png", new File("E:/invoice-template-"+i+".png"));
}
}
catch (FileNotFoundException e1) {
System.err.println(e1.getLocalizedMessage());
} catch (IOException e) {
System.err.println(e.getLocalizedMessage());
}
}
}
Then the image is displayed to the user in JavaFX application in ImageView components.
Can you help me to get the exact position of the mouse, the mouse when the user selects a portion of the image from which you want to read the text in the pdf file?
With this code I read pdf file and get text from the set position, only I must to manually input position:( . I use pdfbox-1.3.1.jar.
I would like to position the client chooses to keep a picture in the list and read the text from the pdf file with all of these positions.
File file = new File("E:/invoice-template-1.pdf");
PDDocument document = PDDocument.load(file);
PDFTextStripperByArea stripper = new PDFTextStripperByArea();
stripper.setSortByPosition(true);
Rectangle rect1 = new Rectangle(38, 275, 15, 100);
Rectangle rect2 = new Rectangle(54, 275, 40, 100);
stripper.addRegion("row1column1", rect1);
stripper.addRegion("row1column2", rect2);
List allPages = document.getDocumentCatalog().getAllPages();
List<PDPage> pages = document.getDocumentCatalog().getAllPages();
int j = 0;
for (PDPage page : pages) {
stripper.extractRegions(page);
stripper.setSortByPosition(true);
List<String> regions = stripper.getRegions();
for (String region : regions) {
String text = stripper.getTextForRegion(region);
System.out.println("Region: " + region + " on Page " + j);
System.out.println("\tText: \n" + text);
}
For example,
in the next invoice, I want to select the 4 positions to export the text, and when you select the picture, the dimensions of keeping in the list, then go through the list and from those positions export text from pdf file.

OpenCV 3 (Java Binding) : Apply CLAHE to image

I try to use the java bindings of open cv to apply an non-global contrast (histogram) optimization for a (color) png image, but I fail to get it to work.
import java.awt.image.BufferedImage;
import java.awt.image.DataBufferByte;
import java.io.File;
import javax.imageio.ImageIO;
import org.opencv.core.Core;
import org.opencv.core.CvType;
import org.opencv.core.Mat;
import org.opencv.imgcodecs.Imgcodecs;
import org.opencv.imgproc.CLAHE;
import org.opencv.imgproc.Imgproc;
public class Main {
public static void main( String[] args ) {
try {
System.loadLibrary( Core.NATIVE_LIBRARY_NAME );
// fetch the png
File input = new File("test.png");
BufferedImage buffImage = ImageIO.read(input);
byte[] data = ((DataBufferByte) buffImage.getRaster().getDataBuffer()).getData();
// build MAT for original image
Mat orgImage = new Mat(buffImage.getHeight(),buffImage.getWidth(), CvType.CV_8UC3);
orgImage.put(0, 0, data);
// transform from to LAB
Mat labImage = new Mat(buffImage.getHeight(), buffImage.getWidth(), CvType.CV_8UC4);
Imgproc.cvtColor(orgImage, labImage, Imgproc.COLOR_BGR2Lab);
// apply CLAHE
CLAHE clahe = Imgproc.createCLAHE()
Mat destImage = new Mat(buffImage.getHeight(),buffImage.getWidth(), CvType.CV_8UC4);
clahe.apply(labImage, destImage);
Imgcodecs.imwrite("test_clahe.png", destImage);
} catch (Exception e) {
System.out.println("Error: " + e.getMessage());
}
}
I get the exception:
Error: cv::Exception: C:\builds\master_PackSlaveAddon-win64-vc12-static\opencv\modules\imgproc\src\clahe.cpp:354: error: (-215) _src.type() == CV_8UC1 || _src.type() == CV_16UC1 in function `anonymous
-namespace'::CLAHE_Impl::apply
I guess I need to work with the individual channels, but I cannot figure out how. The code is inspired from this c++ example, but somehow I fail to extract the corresponding layers (I guess I need only L chanel for clahe.apply())

This Example just splits the Lab image and applies Clahe on the L channel which is the intensity channel. So just use this code for java.
List<Mat> channels = new LinkedList();
Core.split(labImage, channels);
CLAHE clahe = Imgproc.createCLAHE()
Mat destImage = new Mat(buffImage.getHeight(),buffImage.getWidth(), CvType.CV_8UC4);
clahe.apply(channels.get(0), destImage);
Core.merge(channels, labImage);
and finally merge the intensity channel to the other channels. I haven't changed any parameters as I don't know how your image looks but I guess that isn't the problem. Hope it helps!

JavaCL - flip input image

I am using JavaCL and I want to rotate the image and save image back to "out.png".
Unfortunately the line of code:
write_imagef(output, (int2){coords.y, coords.x}, pixel );
(x and y coordinates are switched) seems to have no any effect. I can do whatever I want but the result is still the original image! Why the output is not affected?
CopyImagesExample.java
import static java.lang.Math.cos;
import static java.lang.Math.sin;
import static org.bridj.Pointer.allocateFloats;
import java.awt.Image;
import java.awt.image.BufferedImage;
import java.io.File;
import java.io.FileInputStream;
import java.io.IOException;
import java.nio.ByteOrder;
import javax.imageio.ImageIO;
import org.bridj.Pointer;
import com.nativelibs4java.opencl.CLBuffer;
import com.nativelibs4java.opencl.CLContext;
import com.nativelibs4java.opencl.CLDevice;
import com.nativelibs4java.opencl.CLEvent;
import com.nativelibs4java.opencl.CLImage2D;
import com.nativelibs4java.opencl.CLKernel;
import com.nativelibs4java.opencl.CLMem.Usage;
import com.nativelibs4java.opencl.CLSampler;
import com.nativelibs4java.opencl.CLSampler.AddressingMode;
import com.nativelibs4java.opencl.CLSampler.FilterMode;
import com.nativelibs4java.opencl.CLProgram;
import com.nativelibs4java.opencl.CLQueue;
import com.nativelibs4java.opencl.JavaCL;
import com.nativelibs4java.util.IOUtils;
public class CopyImagesExample {
public static void main(String[] args) throws IOException {
CLContext context = JavaCL.createBestContext();
CLQueue queue = context.createDefaultQueue();
ByteOrder byteOrder = context.getByteOrder();
CLDevice[] devices = context.getDevices();
System.out.println("Devices count " + context.getDeviceCount());
System.out.println(devices[0].getMaxComputeUnits());
System.out.println(devices[0].getOpenCLVersion());
int n = 1024;
Pointer<Float> aPtr = allocateFloats(n).order(byteOrder);
Pointer<Float> bPtr = allocateFloats(n).order(byteOrder);
Pointer<Float> oPtr = allocateFloats(n).order(byteOrder);
for (int i = 0; i < n; i++) {
aPtr.set(i, (float) cos(i));
bPtr.set(i, (float) sin(i));
}
// Create OpenCL input buffers (using the native memory pointers aPtr
// and bPtr) :
BufferedImage img = ImageIO.read(new FileInputStream("images/lena256.png"));
CLImage2D inputImage = context.createImage2D(Usage.Input, img, false);
// Create an OpenCL output buffer :
CLImage2D outImage = context.createImage2D(Usage.Output, img, false);
CLSampler sampler = context.createSampler(false, AddressingMode.ClampToEdge, FilterMode.Nearest);
// Read the program sources and compile them :
String src = IOUtils.readText(CopyImagesExample.class
.getResource("CopyImage.cl"));
CLProgram program = context.createProgram(devices, src);
// Get and call the kernel :
CLKernel nnKernel = program.createKernel("rotate_image");
nnKernel.setArgs(inputImage, outImage, sampler);
CLEvent addEvt = nnKernel.enqueueNDRange(queue, new int[] { img.getWidth()*img.getHeight() });
BufferedImage outPtr = outImage.read(queue, addEvt); // blocks until
// add_floats
// finished
ImageIO.write(outPtr, "png", new File("out.png"));
}
}
CopyImage.cl
__kernel void rotate_image(
__read_only image2d_t input,
__write_only image2d_t output,
sampler_t sampler)
{
// Store each work-items unique row and column
int2 coords = (int2){get_global_id(0), get_global_id(1)};
float4 pixel = read_imagef(input, sampler, coords);
write_imagef(output, (int2){coords.y, coords.x}, pixel );
}
Input:
Output:
But desired output is that the image is rotated (x and y coordinates of each pixel is switched).

Quick fix: replace code :
CLEvent addEvt = nnKernel.enqueueNDRange(queue, new int[] { img.getWidth()*img.getHeight() });
with:
CLEvent addEvt = nnKernel.enqueueNDRange(queue, new int[] { img.getWidth(), img.getHeight() });
The reason is, that in OpenCL code I am using
get_global_id(0)
and
get_global_id(1)
So I need enqueue 2D range.

Determine whether an image is grayscale in Java

I'm using something along the lines of this to do a (naive, apparently) check for the color space of JPEG images:
import java.io.*;
import java.awt.color.*;
import java.awt.image.*;
import javax.imageio.*;
class Test
{
public static void main(String[] args) throws java.lang.Exception
{
File f = new File(args[0]);
if (f.exists())
{
BufferedImage bi = ImageIO.read(f);
ColorSpace cs = bi.getColorModel().getColorSpace();
boolean isGrayscale = cs.getType() == ColorSpace.TYPE_GRAY;
System.out.println(isGrayscale);
}
}
}
Unfortunately this reports false for images that (visually) appear gray-only.
What check would do the right thing?

You can use this code:
File input = new File("inputImage.jpg");
BufferedImage image = ImageIO.read(input);
Raster ras = image.getRaster();
int elem = ras.getNumDataElements();
System.out.println("Number of Elems: " + elem);
If the number of elems returns 1, then its a greyscale image. If it returns 3, then its a color image.

the image looks like gray beacuse the r=g=b but actually it is a full color image, it has three channel r g b and the real gray image only have one channel

Incorrect number of Contours on javacv?

I have write simple code to retrieve number of Contours in a image and get the number of Contours in the image. But it always gives incorrect answer. Please can some one explain about this ?
import com.googlecode.javacpp.Loader;
import com.googlecode.javacv.CanvasFrame;
import static com.googlecode.javacpp.Loader.*;
import static com.googlecode.javacv.cpp.opencv_core.*;
import static com.googlecode.javacv.cpp.opencv_imgproc.*;
import static com.googlecode.javacv.cpp.opencv_highgui.*;
import java.io.File;
import javax.swing.JFileChooser;
public class TestBeam {
public static void main(String[] args) {
CvMemStorage storage=CvMemStorage.create();
CvSeq squares = new CvContour();
squares = cvCreateSeq(0, sizeof(CvContour.class), sizeof(CvSeq.class), storage);
JFileChooser f=new JFileChooser();
int result=f.showOpenDialog(f);//show dialog box to choose files
File myfile=null;
String path="";
if(result==0){
myfile=f.getSelectedFile();//selected file taken to myfile
path=myfile.getAbsolutePath();//get the path of the file
}
IplImage src = cvLoadImage(path);//hear path is actual path to image
IplImage grayImage = IplImage.create(src.width(), src.height(), IPL_DEPTH_8U, 1);
cvCvtColor(src, grayImage, CV_RGB2GRAY);
cvThreshold(grayImage, grayImage, 127, 255, CV_THRESH_BINARY);
CvSeq cvSeq=new CvSeq();
CvMemStorage memory=CvMemStorage.create();
cvFindContours(grayImage, memory, cvSeq, Loader.sizeof(CvContour.class), CV_RETR_EXTERNAL, CV_CHAIN_APPROX_SIMPLE);
System.out.println(cvSeq.elem_size());
CanvasFrame cnvs=new CanvasFrame("Beam");
cnvs.setDefaultCloseOperation(javax.swing.JFrame.EXIT_ON_CLOSE);
cnvs.showImage(src);
//cvShowImage("Final ", src);
}
}
This is the sample image that I used
But Code always returns output as 8. Please can someone explain this?

cvSeq.elem_size() will return size of sequence element in bytes and not the number of contours. That is why output is 8 every time.
Please refer following link for more information.
http://opencv.willowgarage.com/documentation/dynamic_structures.html#cvseq
For finding number of contours you can use following snippet
int i = 0;
while(cvSeq != null){
i = i + 1;
cvSeq = cvSeq.h_next();
}
System.out.println(i);
With the parameters you have provided CV_RETR_EXTERNAL will only provide external contour that is image boundary in your image(provided you are not inverting the image). You can use CV_RETR_LIST to get all the contours. Visit following link for more information on the parameters.
http://opencv.willowgarage.com/documentation/structural_analysis_and_shape_descriptors.html#findcontours