pdf 识别 2023年4月1日11:06:23

README.md

@@ -100,6 +100,12 @@
 
 ![](./src/main/resources/static/redame/img_41.png)
 
+#### 3.14 pdf内容识别
+
+> 开源项目 itext
+
+![](./src/main/resources/static/redame/img_42.png)
+
 
 ### 4. 编解码工具
 

pom.xml

@@ -344,6 +344,18 @@
             <artifactId>tess4j</artifactId>
             <version>4.5.2</version>
         </dependency>
+
+        <dependency>
+            <groupId>com.lowagie</groupId>
+            <artifactId>itext</artifactId>
+            <version>4.2.1</version>
+        </dependency>
+        <dependency>
+            <groupId>com.github.codemonstur</groupId>
+            <artifactId>jpedal</artifactId>
+            <version>4.92-p13</version>
+        </dependency>
+
     </dependencies>
 
     <build>

src/main/java/com/zhangmeng/tools/controller/BarCodeController.java

@@ -95,7 +95,7 @@ public class BarCodeController {
     }
 
     public void init_tess4j(File file) {
-        String result = PlateUtil.zxing_bar_code(file.getPath(), 1, 0);
+        String result = PlateUtil.zxing_bar_code(file.getPath());
         res_view.setText(result);
     }
 }

src/main/java/com/zhangmeng/tools/controller/HomeController.java

@@ -496,4 +496,8 @@ public class HomeController implements Serializable {
     public void bar_code_menu_item(ActionEvent event) {
         load_small_tools(12);
     }
+
+    public void pdf_menu_item(ActionEvent event) {
+        load_small_tools(13);
+    }
 }

src/main/java/com/zhangmeng/tools/controller/PdfOcrController.java

@@ -0,0 +1,94 @@
+package com.zhangmeng.tools.controller;
+
+import com.zhangmeng.tools.utils.AlertUtils;
+import com.zhangmeng.tools.utils.ImagePath;
+import com.zhangmeng.tools.utils.PlateUtil;
+import javafx.application.Platform;
+import javafx.beans.property.SimpleObjectProperty;
+import javafx.fxml.FXML;
+import javafx.scene.control.Button;
+import javafx.scene.control.TextArea;
+import javafx.scene.control.TextField;
+import javafx.scene.image.Image;
+import javafx.scene.image.ImageView;
+import javafx.stage.FileChooser;
+import javafx.stage.Stage;
+import lombok.extern.slf4j.Slf4j;
+
+import java.io.File;
+
+/**
+ * @author : 芊芊墨客
+ * @version : 1.0
+ * @date : 2023-04-01 10:46
+ */
+@Slf4j
+public class PdfOcrController {
+
+    public static String tessdata = System.getProperty("user.dir");
+
+    @FXML
+    public Button file_choose_button;
+
+    @FXML
+    public TextArea res_view;
+
+    @FXML
+    public Button cover;
+
+    @FXML
+    public TextField file_path;
+
+    public static final SimpleObjectProperty<File> choose_file = new SimpleObjectProperty<>();
+
+    @FXML
+    public void initialize() {
+
+        file_choose_button.setText(null);
+        ImageView iv = new ImageView(new Image(ImagePath.path(ImagePath.ImagePathType.IMAGE_FILE)));
+        iv.setPreserveRatio(true);
+        iv.setFitWidth(18);
+        file_choose_button.setGraphic(iv);
+
+        cover.setText("识别");
+
+        file_choose_button.setOnAction(event -> {
+            choose_file();
+        });
+
+        cover.setOnAction(event -> {
+            if (choose_file.getValue() == null) {
+                AlertUtils.alert_warning("请选择将要识别的pdf再试!");
+                return;
+            }
+
+            File file = choose_file.getValue();
+            Stage alert = AlertUtils.alert_loading(cover.getScene().getWindow());
+            new Thread(() -> {
+                Platform.runLater(() -> {
+                    init_pdf(file);
+                    alert.close();
+                });
+            }).start();
+        });
+    }
+
+    public void choose_file() {
+        Stage stage = new Stage();
+        FileChooser dc = new FileChooser();
+        dc.setTitle("文件选择");
+        dc.getExtensionFilters().addAll(new FileChooser.ExtensionFilter("类型", "*.pdf"));
+        File file = dc.showOpenDialog(stage);
+        if (file != null) {
+            String path = file.getAbsolutePath();
+            file_path.setText(path);
+            log.info("file_path:{}", path);
+            choose_file.set(file);
+        }
+    }
+
+    public void init_pdf(File file) {
+        String result = PlateUtil.getPdfContent(file);
+        res_view.setText(result);
+    }
+}

src/main/java/com/zhangmeng/tools/controller/SmallToolsController.java

@@ -80,6 +80,7 @@ public class SmallToolsController {
     private AnchorPane maven_install_jar;
     private AnchorPane word_ocr;
     private AnchorPane bar_code;
+    private AnchorPane pdf_ocr;
 
     @FXML
     private ListView<ResourcesUtils.SmallTools> listView;
@@ -385,6 +386,13 @@ public class SmallToolsController {
                     }
                     bar_code(flag);
                 }
+
+                if (newValue.getIndex() == 13) {
+                    if (pdf_ocr != null) {
+                        flag = true;
+                    }
+                    pdf_ocr(flag);
+                }
             }
         });
     }
@@ -407,6 +415,7 @@ public class SmallToolsController {
             case Maven_Install_Jar -> new Image(ImagePath.path(ImagePath.ImagePathType.Qr_CODE));
             case Word_ocr ->  new Image(ImagePath.path(ImagePath.ImagePathType.Qr_CODE));
             case Bar_Code ->  new Image(ImagePath.path(ImagePath.ImagePathType.Qr_CODE));
+            case Pdf_Ocr ->  new Image(ImagePath.path(ImagePath.ImagePathType.Qr_CODE));
         };
     }
 
@@ -677,6 +686,23 @@ public class SmallToolsController {
         common_method();
     }
 
+    public void pdf_ocr(boolean flag) {
+        //默认选择第一个
+        listView.getSelectionModel().select(13);
+
+        if (!flag) {
+            try {
+                root = FXMLLoader.load(ResourcesUtils.getResource("pdf-ocr"));
+            } catch (IOException e) {
+                e.printStackTrace();
+            }
+            pdf_ocr = root;
+        } else {
+            root = pdf_ocr;
+        }
+        common_method();
+    }
+
 
     private void common_method() {
         splitPane.getItems().remove(1);

src/main/java/com/zhangmeng/tools/utils/PlateUtil.java

@@ -4,6 +4,9 @@ package com.zhangmeng.tools.utils;
 import com.google.zxing.*;
 import com.google.zxing.client.j2se.BufferedImageLuminanceSource;
 import com.google.zxing.common.HybridBinarizer;
+import com.itextpdf.text.pdf.PdfReader;
+import com.itextpdf.text.pdf.parser.PdfTextExtractor;
+
 import lombok.extern.slf4j.Slf4j;
 import org.opencv.core.Core;
 import org.opencv.core.Mat;
@@ -11,9 +14,13 @@ import org.opencv.core.MatOfByte;
 import org.opencv.core.MatOfInt;
 import org.opencv.imgcodecs.Imgcodecs;
 import org.opencv.imgproc.Imgproc;
+import org.springframework.web.multipart.MultipartFile;
 
 import java.awt.image.BufferedImage;
 import java.awt.image.DataBufferByte;
+import java.io.File;
+import java.io.FileInputStream;
+import java.io.IOException;
 import java.util.HashMap;
 import java.util.Map;
 
@@ -170,446 +177,17 @@ public class PlateUtil {
             IMWRITE_PNG_STRATEGY_FIXED = 4,
             IMWRITE_PXM_BINARY = 32;
 
-
-    //
-    // C++:  Mat imdecode(Mat buf, int flags)
-    //
-
     /**
-     * <p>Reads an image from a buffer in memory.</p>
+     * 条形码识别
-     *
-     * <p>The function reads an image from the specified buffer in the memory.
-     * If the buffer is too short or contains invalid data, the empty matrix/image
-     * is returned.</p>
-     *
-     * <p>See "imread" for the list of supported formats and flags description.</p>
-     *
-     * <p>Note: In the case of color images, the decoded images will have the channels
-     * stored in <code>B G R</code> order.</p>
-     *
-     * @param buf   Input array or vector of bytes.
-     * @param flags The same flags as in "imread".
-     * @see <a href="http://docs.opencv.org/modules/highgui/doc/reading_and_writing_images_and_video.html#imdecode">org.opencv.highgui.Highgui.imdecode</a>
      */
-    public static Mat imdecode(Mat buf, int flags) {
+    public static String zxing_bar_code(String sourcePath) {
-
-        Mat retVal = new Mat(imdecode_0(buf.nativeObj, flags));
-
-        return retVal;
-    }
-
-
-    //
-    // C++:  bool imencode(string ext, Mat img, vector_uchar& buf, vector_int params = vector<int>())
-    //
-
-    /**
-     * <p>Encodes an image into a memory buffer.</p>
-     *
-     * <p>The function compresses the image and stores it in the memory buffer that is
-     * resized to fit the result.
-     * See "imwrite" for the list of supported formats and flags description.</p>
-     *
-     * <p>Note: <code>cvEncodeImage</code> returns single-row matrix of type
-     * <code>CV_8UC1</code> that contains encoded image as array of bytes.</p>
-     *
-     * @param ext    File extension that defines the output format.
-     * @param img    Image to be written.
-     * @param buf    Output buffer resized to fit the compressed image.
-     * @param params Format-specific parameters. See "imwrite".
-     * @see <a href="http://docs.opencv.org/modules/highgui/doc/reading_and_writing_images_and_video.html#imencode">org.opencv.highgui.Highgui.imencode</a>
-     */
-    public static boolean imencode(String ext, Mat img, MatOfByte buf, MatOfInt params) {
-        Mat buf_mat = buf;
-        Mat params_mat = params;
-        boolean retVal = imencode_0(ext, img.nativeObj, buf_mat.nativeObj, params_mat.nativeObj);
-
-        return retVal;
-    }
-
-    /**
-     * <p>Encodes an image into a memory buffer.</p>
-     *
-     * <p>The function compresses the image and stores it in the memory buffer that is
-     * resized to fit the result.
-     * See "imwrite" for the list of supported formats and flags description.</p>
-     *
-     * <p>Note: <code>cvEncodeImage</code> returns single-row matrix of type
-     * <code>CV_8UC1</code> that contains encoded image as array of bytes.</p>
-     *
-     * @param ext File extension that defines the output format.
-     * @param img Image to be written.
-     * @param buf Output buffer resized to fit the compressed image.
-     * @see <a href="http://docs.opencv.org/modules/highgui/doc/reading_and_writing_images_and_video.html#imencode">org.opencv.highgui.Highgui.imencode</a>
-     */
-    public static boolean imencode(String ext, Mat img, MatOfByte buf) {
-        Mat buf_mat = buf;
-        boolean retVal = imencode_1(ext, img.nativeObj, buf_mat.nativeObj);
-
-        return retVal;
-    }
-
-
-    //
-    // C++:  Mat imread(string filename, int flags = 1)
-    //
-
-    /**
-     * <p>Loads an image from a file.</p>
-     *
-     * <p>The function <code>imread</code> loads an image from the specified file and
-     * returns it. If the image cannot be read (because of missing file, improper
-     * permissions, unsupported or invalid format), the function returns an empty
-     * matrix (<code>Mat.data==NULL</code>). Currently, the following file formats
-     * are supported:</p>
-     * <ul>
-     *   <li> Windows bitmaps - <code>*.bmp, *.dib</code> (always supported)
-     *   <li> JPEG files - <code>*.jpeg, *.jpg, *.jpe</code> (see the *Notes*
-     * section)
-     *   <li> JPEG 2000 files - <code>*.jp2</code> (see the *Notes* section)
-     *   <li> Portable Network Graphics - <code>*.png</code> (see the *Notes*
-     * section)
-     *   <li> Portable image format - <code>*.pbm, *.pgm, *.ppm</code> (always
-     * supported)
-     *   <li> Sun rasters - <code>*.sr, *.ras</code> (always supported)
-     *   <li> TIFF files - <code>*.tiff, *.tif</code> (see the *Notes* section)
-     * </ul>
-     *
-     * <p>Note:</p>
-     * <ul>
-     *   <li> The function determines the type of an image by the content, not by
-     * the file extension.
-     *   <li> On Microsoft Windows* OS and MacOSX*, the codecs shipped with an
-     * OpenCV image (libjpeg, libpng, libtiff, and libjasper) are used by default.
-     * So, OpenCV can always read JPEGs, PNGs, and TIFFs. On MacOSX, there is also
-     * an option to use native MacOSX image readers. But beware that currently these
-     * native image loaders give images with different pixel values because of the
-     * color management embedded into MacOSX.
-     *   <li> On Linux*, BSD flavors and other Unix-like open-source operating
-     * systems, OpenCV looks for codecs supplied with an OS image. Install the
-     * relevant packages (do not forget the development files, for example,
-     * "libjpeg-dev", in Debian* and Ubuntu*) to get the codec support or turn on
-     * the <code>OPENCV_BUILD_3RDPARTY_LIBS</code> flag in CMake.
-     * </ul>
-     *
-     * <p>Note: In the case of color images, the decoded images will have the channels
-     * stored in <code>B G R</code> order.</p>
-     *
-     * @param filename Name of file to be loaded.
-     * @param flags    Flags specifying the color type of a loaded image:
-     *                 <ul>
-     *                   <li> CV_LOAD_IMAGE_ANYDEPTH - If set, return 16-bit/32-bit image when the
-     *                 input has the corresponding depth, otherwise convert it to 8-bit.
-     *                   <li> CV_LOAD_IMAGE_COLOR - If set, always convert image to the color one
-     *                   <li> CV_LOAD_IMAGE_GRAYSCALE - If set, always convert image to the
-     *                 grayscale one
-     *                   <li> >0 Return a 3-channel color image.
-     *                 </ul>
-     *                 <p>Note: In the current implementation the alpha channel, if any, is stripped
-     *                 from the output image. Use negative value if you need the alpha channel.</p>
-     *                 <ul>
-     *                   <li> =0 Return a grayscale image.
-     *                   <li> <0 Return the loaded image as is (with alpha channel).
-     *                 </ul>
-     * @see <a href="http://docs.opencv.org/modules/highgui/doc/reading_and_writing_images_and_video.html#imread">org.opencv.highgui.Highgui.imread</a>
-     */
-    public static Mat imread(String filename, int flags) {
-
-        Mat retVal = new Mat(imread_0(filename, flags));
-
-        return retVal;
-    }
-
-    /**
-     * <p>Loads an image from a file.</p>
-     *
-     * <p>The function <code>imread</code> loads an image from the specified file and
-     * returns it. If the image cannot be read (because of missing file, improper
-     * permissions, unsupported or invalid format), the function returns an empty
-     * matrix (<code>Mat.data==NULL</code>). Currently, the following file formats
-     * are supported:</p>
-     * <ul>
-     *   <li> Windows bitmaps - <code>*.bmp, *.dib</code> (always supported)
-     *   <li> JPEG files - <code>*.jpeg, *.jpg, *.jpe</code> (see the *Notes*
-     * section)
-     *   <li> JPEG 2000 files - <code>*.jp2</code> (see the *Notes* section)
-     *   <li> Portable Network Graphics - <code>*.png</code> (see the *Notes*
-     * section)
-     *   <li> Portable image format - <code>*.pbm, *.pgm, *.ppm</code> (always
-     * supported)
-     *   <li> Sun rasters - <code>*.sr, *.ras</code> (always supported)
-     *   <li> TIFF files - <code>*.tiff, *.tif</code> (see the *Notes* section)
-     * </ul>
-     *
-     * <p>Note:</p>
-     * <ul>
-     *   <li> The function determines the type of an image by the content, not by
-     * the file extension.
-     *   <li> On Microsoft Windows* OS and MacOSX*, the codecs shipped with an
-     * OpenCV image (libjpeg, libpng, libtiff, and libjasper) are used by default.
-     * So, OpenCV can always read JPEGs, PNGs, and TIFFs. On MacOSX, there is also
-     * an option to use native MacOSX image readers. But beware that currently these
-     * native image loaders give images with different pixel values because of the
-     * color management embedded into MacOSX.
-     *   <li> On Linux*, BSD flavors and other Unix-like open-source operating
-     * systems, OpenCV looks for codecs supplied with an OS image. Install the
-     * relevant packages (do not forget the development files, for example,
-     * "libjpeg-dev", in Debian* and Ubuntu*) to get the codec support or turn on
-     * the <code>OPENCV_BUILD_3RDPARTY_LIBS</code> flag in CMake.
-     * </ul>
-     *
-     * <p>Note: In the case of color images, the decoded images will have the channels
-     * stored in <code>B G R</code> order.</p>
-     *
-     * @param filename Name of file to be loaded.
-     * @see <a href="http://docs.opencv.org/modules/highgui/doc/reading_and_writing_images_and_video.html#imread">org.opencv.highgui.Highgui.imread</a>
-     */
-    public static Mat imread(String filename) {
-
-        Mat retVal = new Mat(imread_1(filename));
-
-        return retVal;
-    }
-
-
-    //
-    // C++:  bool imwrite(string filename, Mat img, vector_int params = vector<int>())
-    //
-
-    /**
-     * <p>Saves an image to a specified file.</p>
-     *
-     * <p>The function <code>imwrite</code> saves the image to the specified file. The
-     * image format is chosen based on the <code>filename</code> extension (see
-     * "imread" for the list of extensions). Only 8-bit (or 16-bit unsigned
-     * (<code>CV_16U</code>) in case of PNG, JPEG 2000, and TIFF) single-channel or
-     * 3-channel (with 'BGR' channel order) images can be saved using this function.
-     * If the format, depth or channel order is different, use "Mat.convertTo", and
-     * "cvtColor" to convert it before saving. Or, use the universal "FileStorage"
-     * I/O functions to save the image to XML or YAML format.
-     * It is possible to store PNG images with an alpha channel using this function.
-     * To do this, create 8-bit (or 16-bit) 4-channel image BGRA, where the alpha
-     * channel goes last. Fully transparent pixels should have alpha set to 0, fully
-     * opaque pixels should have alpha set to 255/65535. The sample below shows how
-     * to create such a BGRA image and store to PNG file. It also demonstrates how
-     * to set custom compression parameters <code></p>
-     *
-     * <p>// C++ code:</p>
-     *
-     * <p>#include <vector></p>
-     *
-     * <p>#include <stdio.h></p>
-     *
-     * <p>#include <opencv2/opencv.hpp></p>
-     *
-     * <p>using namespace cv;</p>
-     *
-     * <p>using namespace std;</p>
-     *
-     * <p>void createAlphaMat(Mat &mat)</p>
-     *
-     *
-     * <p>CV_Assert(mat.channels() == 4);</p>
-     *
-     * <p>for (int i = 0; i < mat.rows; ++i) {</p>
-     *
-     * <p>for (int j = 0; j < mat.cols; ++j) {</p>
-     *
-     * <p>Vec4b& bgra = mat.at<Vec4b>(i, j);</p>
-     *
-     * <p>bgra[0] = UCHAR_MAX; // Blue</p>
-     *
-     * <p>bgra[1] = saturate_cast<uchar>((float (mat.cols - j)) / ((float)mat.cols) *
-     * UCHAR_MAX); // Green</p>
-     *
-     * <p>bgra[2] = saturate_cast<uchar>((float (mat.rows - i)) / ((float)mat.rows) *
-     * UCHAR_MAX); // Red</p>
-     *
-     * <p>bgra[3] = saturate_cast<uchar>(0.5 * (bgra[1] + bgra[2])); // Alpha</p>
-     *
-     *
-     *
-     *
-     * <p>int main(int argv, char argc)</p>
-     *
-     *
-     * <p>// Create mat with alpha channel</p>
-     *
-     * <p>Mat mat(480, 640, CV_8UC4);</p>
-     *
-     * <p>createAlphaMat(mat);</p>
-     *
-     * <p>vector<int> compression_params;</p>
-     *
-     * <p>compression_params.push_back(CV_IMWRITE_PNG_COMPRESSION);</p>
-     *
-     * <p>compression_params.push_back(9);</p>
-     *
-     * <p>try {</p>
-     *
-     * <p>imwrite("alpha.png", mat, compression_params);</p>
-     *
-     *
-     * <p>catch (runtime_error& ex) {</p>
-     *
-     * <p>fprintf(stderr, "Exception converting image to PNG format: %sn", ex.what());</p>
-     *
-     * <p>return 1;</p>
-     *
-     *
-     * <p>fprintf(stdout, "Saved PNG file with alpha data.n");</p>
-     *
-     * <p>return 0;</p>
-     *
-     * @param filename Name of the file.
-     * @param img      a img
-     * @param params   Format-specific save parameters encoded as pairs
-     *                 <code>paramId_1, paramValue_1, paramId_2, paramValue_2,...</code>. The
-     *                 following parameters are currently supported:
-     *                 <ul>
-     *                   <li> For JPEG, it can be a quality (<code>CV_IMWRITE_JPEG_QUALITY</code>)
-     *                 from 0 to 100 (the higher is the better). Default value is 95.
-     *                   <li> For PNG, it can be the compression level (<code>CV_IMWRITE_PNG_COMPRESSION</code>)
-     *                 from 0 to 9. A higher value means a smaller size and longer compression time.
-     *                 Default value is 3.
-     *                   <li> For PPM, PGM, or PBM, it can be a binary format flag (<code>CV_IMWRITE_PXM_BINARY</code>),
-     *                 0 or 1. Default value is 1.
-     *                 </ul>
-     * @see <a href="http://docs.opencv.org/modules/highgui/doc/reading_and_writing_images_and_video.html#imwrite">org.opencv.highgui.Highgui.imwrite</a>
-     */
-    public static boolean imwrite(String filename, Mat img, MatOfInt params) {
-        Mat params_mat = params;
-        boolean retVal = imwrite_0(filename, img.nativeObj, params_mat.nativeObj);
-
-        return retVal;
-    }
-
-    /**
-     * <p>Saves an image to a specified file.</p>
-     *
-     * <p>The function <code>imwrite</code> saves the image to the specified file. The
-     * image format is chosen based on the <code>filename</code> extension (see
-     * "imread" for the list of extensions). Only 8-bit (or 16-bit unsigned
-     * (<code>CV_16U</code>) in case of PNG, JPEG 2000, and TIFF) single-channel or
-     * 3-channel (with 'BGR' channel order) images can be saved using this function.
-     * If the format, depth or channel order is different, use "Mat.convertTo", and
-     * "cvtColor" to convert it before saving. Or, use the universal "FileStorage"
-     * I/O functions to save the image to XML or YAML format.
-     * It is possible to store PNG images with an alpha channel using this function.
-     * To do this, create 8-bit (or 16-bit) 4-channel image BGRA, where the alpha
-     * channel goes last. Fully transparent pixels should have alpha set to 0, fully
-     * opaque pixels should have alpha set to 255/65535. The sample below shows how
-     * to create such a BGRA image and store to PNG file. It also demonstrates how
-     * to set custom compression parameters <code></p>
-     *
-     * <p>// C++ code:</p>
-     *
-     * <p>#include <vector></p>
-     *
-     * <p>#include <stdio.h></p>
-     *
-     * <p>#include <opencv2/opencv.hpp></p>
-     *
-     * <p>using namespace cv;</p>
-     *
-     * <p>using namespace std;</p>
-     *
-     * <p>void createAlphaMat(Mat &mat)</p>
-     *
-     *
-     * <p>CV_Assert(mat.channels() == 4);</p>
-     *
-     * <p>for (int i = 0; i < mat.rows; ++i) {</p>
-     *
-     * <p>for (int j = 0; j < mat.cols; ++j) {</p>
-     *
-     * <p>Vec4b& bgra = mat.at<Vec4b>(i, j);</p>
-     *
-     * <p>bgra[0] = UCHAR_MAX; // Blue</p>
-     *
-     * <p>bgra[1] = saturate_cast<uchar>((float (mat.cols - j)) / ((float)mat.cols) *
-     * UCHAR_MAX); // Green</p>
-     *
-     * <p>bgra[2] = saturate_cast<uchar>((float (mat.rows - i)) / ((float)mat.rows) *
-     * UCHAR_MAX); // Red</p>
-     *
-     * <p>bgra[3] = saturate_cast<uchar>(0.5 * (bgra[1] + bgra[2])); // Alpha</p>
-     *
-     *
-     *
-     *
-     * <p>int main(int argv, char argc)</p>
-     *
-     *
-     * <p>// Create mat with alpha channel</p>
-     *
-     * <p>Mat mat(480, 640, CV_8UC4);</p>
-     *
-     * <p>createAlphaMat(mat);</p>
-     *
-     * <p>vector<int> compression_params;</p>
-     *
-     * <p>compression_params.push_back(CV_IMWRITE_PNG_COMPRESSION);</p>
-     *
-     * <p>compression_params.push_back(9);</p>
-     *
-     * <p>try {</p>
-     *
-     * <p>imwrite("alpha.png", mat, compression_params);</p>
-     *
-     *
-     * <p>catch (runtime_error& ex) {</p>
-     *
-     * <p>fprintf(stderr, "Exception converting image to PNG format: %sn", ex.what());</p>
-     *
-     * <p>return 1;</p>
-     *
-     *
-     * <p>fprintf(stdout, "Saved PNG file with alpha data.n");</p>
-     *
-     * <p>return 0;</p>
-     *
-     * @param filename Name of the file.
-     * @param img      a img
-     * @see <a href="http://docs.opencv.org/modules/highgui/doc/reading_and_writing_images_and_video.html#imwrite">org.opencv.highgui.Highgui.imwrite</a>
-     */
-    public static boolean imwrite(String filename, Mat img) {
-
-        boolean retVal = imwrite_1(filename, img.nativeObj);
-
-        return retVal;
-    }
-
-
-    // C++:  Mat imdecode(Mat buf, int flags)
-    private static native long imdecode_0(long buf_nativeObj, int flags);
-
-    // C++:  bool imencode(string ext, Mat img, vector_uchar& buf, vector_int params = vector<int>())
-    private static native boolean imencode_0(String ext, long img_nativeObj, long buf_mat_nativeObj, long params_mat_nativeObj);
-
-    private static native boolean imencode_1(String ext, long img_nativeObj, long buf_mat_nativeObj);
-
-    // C++:  Mat imread(string filename, int flags = 1)
-    private static native long imread_0(String filename, int flags);
-
-    private static native long imread_1(String filename);
-
-    // C++:  bool imwrite(string filename, Mat img, vector_int params = vector<int>())
-    private static native boolean imwrite_0(String filename, long img_nativeObj, long params_mat_nativeObj);
-
-    private static native boolean imwrite_1(String filename, long img_nativeObj);
-
-    /**
-     * @param binaryType     二值化类型
-     * @param adaptiveMethod
-     */
-    public static String zxing_bar_code(String sourcePath,Integer adaptiveMethod, Integer binaryType) {
 
         System.loadLibrary(Core.NATIVE_LIBRARY_NAME);
 
         // 加载为灰度图显示
         Mat source = Imgcodecs.imread(sourcePath,Imgcodecs.IMREAD_GRAYSCALE);
         Mat destination = new Mat(source.rows(), source.cols(), source.type());
+        //二值化类型
         Imgproc.threshold(source, destination, 190, 255, Imgproc.THRESH_BINARY);
         String result = parseCode(destination);
         log.info("result:{}",result);
@@ -629,11 +207,9 @@ public class PlateUtil {
             hints.put(DecodeHintType.CHARACTER_SET, "UTF-8");
 
             Result result = formatReader.decode(binaryBitmap, hints);
-            StringBuffer sbuffer = new StringBuffer();
+            resultText = "解析结果 = " + result.toString() + "\n" +
-            sbuffer.append("解析结果 = " + result.toString() + "\n");
+                    "格式类型 = " + result.getBarcodeFormat() + "\n" +
-            sbuffer.append("格式类型 = " + result.getBarcodeFormat() + "\n");
+                    "文本内容 = " + result.getText() + "\n";
-            sbuffer.append("文本内容 = " + result.getText() + "\n");
-            resultText = sbuffer.toString();
         } catch (Exception e) {
             e.printStackTrace();
         }
@@ -653,4 +229,20 @@ public class PlateUtil {
         System.arraycopy(b, 0, targetPixels, 0, b.length);
         return image;
     }
+
+    public static String getPdfContent(File file) {
+        PdfReader reader = null;
+        StringBuilder buff = new StringBuilder();
+        try {
+            reader = new PdfReader(new FileInputStream(file));
+            int num = reader.getNumberOfPages();// 获得页数
+            for (int i = 1; i <= num; i++) {
+                String textFromPage = PdfTextExtractor.getTextFromPage(reader, i);
+                buff.append(textFromPage);
+            }
+        } catch (IOException e) {
+            e.printStackTrace();
+        }
+        return buff.toString();
+    }
 }

src/main/java/com/zhangmeng/tools/utils/ResourcesUtils.java

@@ -123,6 +123,7 @@ public class ResourcesUtils {
         Maven_Install_Jar("maven安装jar",10),
         Word_ocr("文字识别ocr",11),
         Bar_Code("条形码识别",12),
+        Pdf_Ocr("pdf识别",13),
         ;
 
         SmallTools(String title, int index) {

src/main/resources/fxml/home.fxml

@@ -42,7 +42,8 @@
                         <MenuItem mnemonicParsing="false" text="JsonView" onAction="#JsonView_menu_item"/>
                         <MenuItem mnemonicParsing="false" text="maven-jar-install" onAction="#maven_jar_install_menu_item"/>
                         <MenuItem mnemonicParsing="false" text="word-ocr" onAction="#word_ocr_menu_item"/>
-                        <MenuItem mnemonicParsing="false" text="条形码识别r" onAction="#bar_code_menu_item"/>
+                        <MenuItem mnemonicParsing="false" text="条形码识别" onAction="#bar_code_menu_item"/>
+                        <MenuItem mnemonicParsing="false" text="pdf识别" onAction="#pdf_menu_item"/>
                     </items>
                 </Menu>
 

src/main/resources/fxml/pdf-ocr.fxml

@@ -0,0 +1,18 @@
+<?xml version="1.0" encoding="UTF-8"?>
+
+<?import javafx.scene.control.Button?>
+<?import javafx.scene.control.Label?>
+<?import javafx.scene.control.TextArea?>
+<?import javafx.scene.control.TextField?>
+<?import javafx.scene.layout.AnchorPane?>
+
+<AnchorPane prefHeight="649.0" prefWidth="1200.0" xmlns="http://javafx.com/javafx/19" xmlns:fx="http://javafx.com/fxml/1" fx:controller="com.zhangmeng.tools.controller.PdfOcrController">
+   <children>
+      <TextArea fx:id="res_view" layoutX="727.0" layoutY="112.0" prefHeight="340.0" prefWidth="956.0" AnchorPane.bottomAnchor="197.0" AnchorPane.leftAnchor="201.0" AnchorPane.rightAnchor="43.0" AnchorPane.topAnchor="112.0" />
+      <Button fx:id="file_choose_button" layoutX="202.0" layoutY="67.0" mnemonicParsing="false" text="Button" />
+      <Label layoutX="58.0" layoutY="71.0" text="请选择将要识别的pdf:" />
+      <TextField fx:id="file_path" layoutX="280.0" layoutY="67.0" prefHeight="25.0" prefWidth="430.0" AnchorPane.leftAnchor="280.0" AnchorPane.rightAnchor="490.0" />
+      <Button fx:id="cover" layoutX="390.0" layoutY="492.0" mnemonicParsing="false" text="Button" AnchorPane.bottomAnchor="134.0" />
+      <Label layoutX="102.0" layoutY="112.0" text="识别内容:" />
+   </children>
+</AnchorPane>