More tesseract examples

tesseract/samples/src/main/java/org/bytedeco/javacpp/samples/tesseract/IteratorOverClassifierChoicesExample.java

@@ -0,0 +1,73 @@
+
+package org.bytedeco.javacpp.samples.tesseract;
+
+import static org.bytedeco.javacpp.lept.*;
+import static org.bytedeco.javacpp.tesseract.*;
+
+import java.io.File;
+import java.net.URL;
+
+import org.bytedeco.javacpp.BytePointer;
+import org.bytedeco.javacpp.Loader;
+
+/**
+ * To run this program, you need to configure:
+ * <ul>
+ * <li>An environment variable pointing to the dictionaries installed on the system
+ * TESSDATA_PREFIX=/usr/share/tesseract-ocr/4.00</li>
+ * <li>An environment variable to tweak the Locale
+ * LC_ALL=C</li>
+ * </ul>
+ *
+ * @author Arnaud Jeansen
+ */
+public class IteratorOverClassifierChoicesExample {
+    public static void main(String[] args) throws Exception {
+        BytePointer outText;
+        BytePointer choiceText;
+
+        TessBaseAPI api = new TessBaseAPI();
+        // Initialize tesseract-ocr with English, intializing tessdata path with the standard ENV variable
+        if (api.Init(System.getenv("TESSDATA_PREFIX") + "/tessdata", "eng") != 0) {
+            System.err.println("Could not initialize tesseract.");
+            System.exit(1);
+        }
+
+        // Open input image with leptonica library
+        URL url = new URL("https://upload.wikimedia.org/wikipedia/commons/thumb/7/74/Computer_modern_sample.svg/1920px-Computer_modern_sample.svg.png");
+        File file = Loader.cacheResource(url);
+        PIX image = pixRead(file.getAbsolutePath());
+        api.SetImage(image);
+
+        ETEXT_DESC recoc = TessMonitorCreate();
+        api.Recognize(recoc);
+
+        ResultIterator ri = api.GetIterator();
+        int pageIteratorLevel = RIL_SYMBOL;
+        if (ri != null) {
+            do {
+                outText = ri.GetUTF8Text(pageIteratorLevel);
+                float conf = ri.Confidence(pageIteratorLevel);
+                String symbolInformation = String.format("symbol: '%s';  \tconf: %.2f", outText.getString(), conf);
+                System.out.println(symbolInformation);
+                boolean indent = false;
+                ChoiceIterator ci = TessResultIteratorGetChoiceIterator(ri);
+                do {
+                    if (indent)
+                        System.out.print("\t\t");
+                    System.out.print("\t-");
+                    choiceText = ci.GetUTF8Text();
+                    System.out.println(String.format("%s conf: %f", choiceText.getString(), ci.Confidence()));
+                    indent = true;
+                    choiceText.deallocate();
+                } while (ci.Next());
+
+                outText.deallocate();
+            } while (ri.Next(pageIteratorLevel));
+        }
+
+        // Destroy used object and release memory
+        api.End();
+        pixDestroy(image);
+    }
+}

tesseract/samples/src/main/java/org/bytedeco/javacpp/samples/tesseract/OrientationAndScriptDetectionExample.java

@@ -0,0 +1,60 @@
+
+package org.bytedeco.javacpp.samples.tesseract;
+
+import static org.bytedeco.javacpp.lept.*;
+import static org.bytedeco.javacpp.tesseract.*;
+
+import java.io.File;
+import java.net.URL;
+
+import org.bytedeco.javacpp.BytePointer;
+import org.bytedeco.javacpp.Loader;
+import org.bytedeco.javacpp.tesseract;
+
+/**
+ * To run this program, you need to configure:
+ * <ul>
+ * <li>An environment variable pointing to the dictionaries installed on the system
+ * TESSDATA_PREFIX=/usr/share/tesseract-ocr/4.00</li>
+ * <li>An environment variable to tweak the Locale
+ * LC_ALL=C</li>
+ * </ul>
+ *
+ * @author Arnaud Jeansen
+ */
+public class OrientationAndScriptDetectionExample {
+    public static void main(String[] args) throws Exception {
+        BytePointer outText;
+
+        TessBaseAPI api = new TessBaseAPI();
+        // Initialize tesseract-ocr with English, intializing tessdata path with the standard ENV variable
+        if (api.Init(System.getenv("TESSDATA_PREFIX") + "/tessdata", "eng") != 0) {
+            System.err.println("Could not initialize tesseract.");
+            System.exit(1);
+        }
+
+        // Open input image with leptonica library
+        URL url = new URL("https://upload.wikimedia.org/wikipedia/commons/thumb/7/74/Computer_modern_sample.svg/1920px-Computer_modern_sample.svg.png");
+        File file = Loader.cacheResource(url);
+        PIX image = pixRead(file.getAbsolutePath());
+        api.SetPageSegMode(PSM_AUTO_OSD);
+        api.SetImage(image);
+        tesseract.ETEXT_DESC reco = TessMonitorCreate();
+        api.Recognize(reco);
+
+        tesseract.PageIterator iterator = api.AnalyseLayout();
+        int[] orientation = new int[1];
+        int[] writing_direction = new int[1];
+        int[] textline_order = new int[1];
+        float[] deskew_angle = new float[1];
+
+        iterator.Orientation(orientation, writing_direction, textline_order, deskew_angle);
+        String osdInformation = String.format("Orientation: %d;\nWritingDirection: %d\nTextlineOrder: %d\nDeskew angle: %.4f\n",
+                orientation[0], writing_direction[0], textline_order[0], deskew_angle[0]);
+        System.out.println(osdInformation);
+
+        // Destroy used object and release memory
+        api.End();
+        pixDestroy(image);
+    }
+}

tesseract/samples/src/main/java/org/bytedeco/javacpp/samples/tesseract/ResultIteratorExample.java

@@ -0,0 +1,64 @@
+
+package org.bytedeco.javacpp.samples.tesseract;
+
+import static org.bytedeco.javacpp.lept.*;
+import static org.bytedeco.javacpp.tesseract.*;
+
+import java.io.File;
+import java.net.URL;
+
+import org.bytedeco.javacpp.BytePointer;
+import org.bytedeco.javacpp.Loader;
+import org.bytedeco.javacpp.tesseract;
+
+/**
+ * To run this program, you need to configure:
+ * <ul>
+ * <li>An environment variable pointing to the dictionaries installed on the system
+ * TESSDATA_PREFIX=/usr/share/tesseract-ocr/4.00</li>
+ * <li>An environment variable to tweak the Locale
+ * LC_ALL=C</li>
+ * </ul>
+ *
+ * @author Arnaud Jeansen
+ */
+public class ResultIteratorExample {
+    public static void main(String[] args) throws Exception {
+        BytePointer outText;
+
+        TessBaseAPI api = new TessBaseAPI();
+        // Initialize tesseract-ocr with English, intializing tessdata path with the standard ENV variable
+        if (api.Init(System.getenv("TESSDATA_PREFIX") + "/tessdata", "eng") != 0) {
+            System.err.println("Could not initialize tesseract.");
+            System.exit(1);
+        }
+
+        // Open input image with leptonica library
+        URL url = new URL("https://upload.wikimedia.org/wikipedia/commons/thumb/7/74/Computer_modern_sample.svg/1920px-Computer_modern_sample.svg.png");
+        File file = Loader.cacheResource(url);
+        PIX image = pixRead(file.getAbsolutePath());
+        api.SetImage(image);
+
+        tesseract.ETEXT_DESC recoc = TessMonitorCreate();
+        api.Recognize(recoc);
+
+        tesseract.ResultIterator ri = api.GetIterator();
+        int pageIteratorLevel = RIL_WORD;
+        if (ri != null) {
+            do {
+                outText = ri.GetUTF8Text(pageIteratorLevel);
+                float conf = ri.Confidence(pageIteratorLevel);
+                int[] x1 = new int[1], y1 = new int[1], x2 = new int[1], y2 = new int[1];
+                ri.BoundingBox(pageIteratorLevel, x1, y1, x2, y2);
+                String riInformation = String.format("word: '%s';  \tconf: %.2f; BoundingBox: %d,%d,%d,%d;\n", outText.getString(), conf, x1[0], y1[0], x2[0], y2[0]);
+                System.out.println(riInformation);
+
+                outText.deallocate();
+            } while (ri.Next(pageIteratorLevel));
+        }
+
+        // Destroy used object and release memory
+        api.End();
+        pixDestroy(image);
+    }
+}