Import redaction fixes

redaction-service-v1/redaction-service-server-v1/pom.xml

@@ -20,6 +20,11 @@
     </properties>
 
     <dependencies>
+        <dependency>
+            <groupId>org.openpnp</groupId>
+            <artifactId>opencv</artifactId>
+            <version>4.5.1-2</version>
+        </dependency>
         <dependency>
             <groupId>com.iqser.red.commons</groupId>
             <artifactId>storage-commons</artifactId>

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/redaction/service/RedactionExtractorService.java

@@ -0,0 +1,184 @@
+package com.iqser.red.service.redaction.v1.server.redaction.service;
+
+import static org.opencv.imgproc.Imgproc.COLOR_BGR2GRAY;
+
+import java.awt.Color;
+import java.awt.image.BufferedImage;
+import java.awt.image.DataBufferByte;
+import java.io.ByteArrayOutputStream;
+import java.io.File;
+import java.io.FileOutputStream;
+import java.io.IOException;
+import java.io.InputStream;
+import java.util.ArrayList;
+import java.util.HashMap;
+import java.util.Iterator;
+import java.util.List;
+import java.util.Map;
+import java.util.UUID;
+import java.util.stream.Collectors;
+
+import javax.imageio.ImageIO;
+
+import org.apache.pdfbox.pdmodel.PDDocument;
+import org.apache.pdfbox.pdmodel.PDPage;
+import org.apache.pdfbox.rendering.ImageType;
+import org.apache.pdfbox.rendering.PDFRenderer;
+import org.opencv.core.Core;
+import org.opencv.core.Mat;
+import org.opencv.core.MatOfByte;
+import org.opencv.core.MatOfPoint;
+import org.opencv.core.MatOfPoint2f;
+import org.opencv.core.Rect;
+import org.opencv.core.Scalar;
+import org.opencv.core.Size;
+import org.opencv.imgcodecs.Imgcodecs;
+import org.opencv.imgproc.Imgproc;
+import org.springframework.beans.BeanUtils;
+import org.springframework.stereotype.Service;
+
+import com.iqser.red.service.redaction.v1.model.ImportedRedaction;
+import com.iqser.red.service.redaction.v1.model.ImportedRedactions;
+import com.iqser.red.service.redaction.v1.model.Point;
+import com.iqser.red.service.redaction.v1.model.Rectangle;
+
+import lombok.SneakyThrows;
+
+@Service
+public class RedactionExtractorService {
+
+    @SneakyThrows
+    public ImportedRedactions extractRedaction(InputStream pdfFile) {
+
+        PDDocument pdfDocument = PDDocument.load(pdfFile);
+        pdfDocument.setAllSecurityToBeRemoved(true);
+
+        long pageCount = pdfDocument.getNumberOfPages();
+
+        Map<Integer, List<ImportedRedaction>> importedRedactionsMap = new HashMap<>();
+        for (int pageNumber = 1; pageNumber <= pageCount; pageNumber++) {
+            PDPage page = pdfDocument.getPage(pageNumber - 1);
+            BufferedImage image = pageConvertToImage(pdfDocument, page, 200, ImageType.GRAY);
+
+
+            Mat imageAsMat = bufferedImage2Mat(image);
+
+            Mat imageAsMatINv = new Mat();
+
+            Mat invertcolormatrix= new Mat(imageAsMat.rows(),imageAsMat.cols(), imageAsMat.type(), new Scalar(255,255,255));
+
+            Core.subtract(invertcolormatrix, imageAsMat, imageAsMatINv);
+
+
+
+            Mat blurred = new Mat();
+            Imgproc.GaussianBlur(imageAsMatINv,blurred, new Size(5, 5), 1);
+            Mat thresh = new Mat();
+            Imgproc.threshold(blurred, thresh, 240, 255, Imgproc.THRESH_BINARY);
+
+
+            Mat hierachy = new Mat();
+            List<MatOfPoint> contours = new ArrayList<>();
+            Imgproc.findContours(thresh,contours,hierachy, Imgproc.RETR_TREE, Imgproc.CHAIN_APPROX_NONE);
+
+
+            System.out.println(contours);
+            System.out.println(hierachy.size(3));
+
+            var min_nomralized_area=800;
+
+            if(hierachy.rows() > 0 && hierachy.row(0).rows() > 0) {
+                hierachy = hierachy.row(0).row(0);
+
+                System.out.println("Cont" + contours.size() + "Hier" + hierachy.cols());
+
+//                for(int j = 0; j<= hierachy.cols(); j++){
+//                    System.out.println("----> Col " + j);
+//                    for (double d :hierachy.get(0,j)){
+//                        System.out.println( " " + d);
+//                    }
+//                }
+
+                int i = 0;
+                List<MatOfPoint> filterd = new ArrayList<>();
+                for (MatOfPoint contour : contours) {
+                    if (isLikelyRedaction(contour, hierachy.get(0, i), min_nomralized_area)) {
+                        filterd.add(contour);
+                    }
+                    i++;
+                }
+
+                Imgproc.drawContours(imageAsMat, filterd, -1, new Scalar(180), 2);
+
+                List<ImportedRedaction> importedRedactions = new ArrayList<>();
+                for (MatOfPoint contour : filterd) {
+                    Rect rect = Imgproc.boundingRect(contour);
+
+                    importedRedactions.add(new ImportedRedaction(UUID.randomUUID().toString(), List.of(new Rectangle(new Point((float) (rect.x * page.getCropBox()
+                            .getWidth() / image.getWidth()), (float) (page.getCropBox().getHeight() - rect.y * page.getCropBox()
+                            .getWidth() / image.getWidth())), rect.width * page.getCropBox()
+                            .getWidth() / image.getWidth(), -rect.height * page.getCropBox()
+                            .getWidth() / image.getWidth(), pageNumber))));
+                }
+
+                importedRedactionsMap.put(pageNumber, importedRedactions);
+
+            }
+
+//            BufferedImage gray = new BufferedImage(imageAsMat.width(), imageAsMat.height(), BufferedImage.TYPE_BYTE_GRAY);
+//            byte[] data = ((DataBufferByte) gray.getRaster().getDataBuffer()).getData();
+//            imageAsMat.get(0, 0, data);
+//
+//            File outputfile = new File("/tmp/image.jpg");
+//            ImageIO.write(gray, "jpg", outputfile);
+
+//
+
+//            System.out.println(contours);
+        }
+        pdfDocument.close();
+
+        return new ImportedRedactions(importedRedactionsMap);
+    }
+
+
+
+    private boolean isFilled(double[] hierachy){
+        return hierachy[3] <= 0 && hierachy[2] == -1;
+    }
+
+   private boolean isBoxy(MatOfPoint contour){
+        double epsilon = 0.01 * Imgproc.arcLength(new MatOfPoint2f( contour.toArray()), true);
+       MatOfPoint2f approx = new MatOfPoint2f();
+        Imgproc.approxPolyDP(new MatOfPoint2f( contour.toArray()), approx, epsilon, true);
+            return approx.toArray().length <= 10;
+    }
+
+
+    private boolean largeEnough(float minArea, MatOfPoint contour){
+        return Imgproc.contourArea(contour, false) > minArea;
+    }
+
+    private boolean isLikelyRedaction(MatOfPoint contour, double[] hierachy , float minArea) {
+       return isFilled(hierachy) && isBoxy(contour) && largeEnough(minArea, contour);
+    }
+
+
+
+    public static Mat bufferedImage2Mat(BufferedImage image) throws IOException {
+
+        ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream();
+        ImageIO.write(image, "jpg", byteArrayOutputStream);
+        byteArrayOutputStream.flush();
+        return Imgcodecs.imdecode(new MatOfByte(byteArrayOutputStream.toByteArray()), Imgcodecs.IMREAD_UNCHANGED);
+    }
+
+
+    public static BufferedImage pageConvertToImage(PDDocument doc, PDPage page, int dpi,
+                                                   ImageType imageType) throws IOException {
+
+        PDFRenderer renderer = new PDFRenderer(doc);
+        return renderer.renderImageWithDPI(doc.getPages().indexOf(page), dpi, imageType);
+    }
+
+}

redaction-service-v1/redaction-service-server-v1/src/test/java/com/iqser/red/service/redaction/v1/server/RedactionIntegrationTest.java

@@ -37,6 +37,7 @@ import org.kie.api.builder.KieBuilder;
 import org.kie.api.builder.KieFileSystem;
 import org.kie.api.builder.KieModule;
 import org.kie.api.runtime.KieContainer;
+import org.opencv.core.Core;
 import org.springframework.amqp.rabbit.core.RabbitTemplate;
 import org.springframework.beans.factory.annotation.Autowired;
 import org.springframework.boot.autoconfigure.EnableAutoConfiguration;
@@ -66,6 +67,7 @@ import com.iqser.red.service.redaction.v1.model.AnalyzeResult;
 import com.iqser.red.service.redaction.v1.model.AnnotateRequest;
 import com.iqser.red.service.redaction.v1.model.AnnotateResponse;
 import com.iqser.red.service.redaction.v1.model.FileAttribute;
+import com.iqser.red.service.redaction.v1.model.ImportedRedactions;
 import com.iqser.red.service.redaction.v1.model.RedactionLogEntry;
 import com.iqser.red.service.redaction.v1.model.RedactionRequest;
 import com.iqser.red.service.redaction.v1.model.RedactionResult;
@@ -78,6 +80,7 @@ import com.iqser.red.service.redaction.v1.server.controller.RedactionController;
 import com.iqser.red.service.redaction.v1.server.memory.MemoryStats;
 import com.iqser.red.service.redaction.v1.server.redaction.service.AnalyzeService;
 import com.iqser.red.service.redaction.v1.server.redaction.service.ManualRedactionSurroundingTextService;
+import com.iqser.red.service.redaction.v1.server.redaction.service.RedactionExtractorService;
 import com.iqser.red.service.redaction.v1.server.redaction.utils.OsUtils;
 import com.iqser.red.service.redaction.v1.server.redaction.utils.ResourceLoader;
 import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizationUtilities;
@@ -554,6 +557,45 @@ public class RedactionIntegrationTest {
     }
 
 
+    @Autowired
+    private RedactionExtractorService redactionExtractorService;
+
+//    static{ System.loadLibrary(Core.NATIVE_LIBRARY_NAME); }
+
+    @Test
+    @SneakyThrows
+    public void testImportRedactions() {
+        nu.pattern.OpenCV.loadShared();
+
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Metolachlor/S-Metolachlor_RAR_01_Volume_1_2018-09-06_redacted1.pdf");
+
+        ImportedRedactions importedRedactions = redactionExtractorService.extractRedaction(pdfFileResource.getInputStream());
+
+
+        String outputFileName = OsUtils.getTemporaryDirectory() + "/Annotated.pdf";
+
+        ClassPathResource pdfFileResource2 = new ClassPathResource("files/Metolachlor/S-Metolachlor_RAR_01_Volume_1_2018-09-06.pdf");
+
+        AnalyzeRequest request = prepareStorage(pdfFileResource2.getInputStream());
+        storageService.storeObject(RedactionStorageService.StorageIdUtils.getStorageId(TEST_DOSSIER_ID, TEST_FILE_ID, FileType.IMPORTED_REDACTIONS), objectMapper.writeValueAsBytes(importedRedactions));
+
+        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        AnalyzeResult result = analyzeService.analyze(request);
+
+        var redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
+
+        AnnotateResponse annotateResponse = redactionController.annotate(AnnotateRequest.builder()
+                .dossierId(TEST_DOSSIER_ID)
+                .fileId(TEST_FILE_ID)
+                .build());
+
+        try (FileOutputStream fileOutputStream = new FileOutputStream(outputFileName)) {
+            fileOutputStream.write(annotateResponse.getDocument());
+        }
+
+    }
+
+
     @Test
     @Ignore
     public void testLargeScannedFileOOM() {