// Scan the photo for text using the tess-two API
public static String scanPhoto(Bitmap bitmap) {
TessBaseAPI baseApi = new TessBaseAPI();
baseApi.setDebug(true);
baseApi.init(DATA_PATH, lang);
// set the black list
baseApi.setVariable("tessedit_char_blacklist", "':;,.?/\\}][{!@#$%^&*()-_=+~");
baseApi.setVariable("save_blob_choices", "T");
baseApi.setImage(bitmap);
String recognizedText = baseApi.getUTF8Text();
// Iterate over the results and print confidence values for debugging purposes
final ResultIterator iterator = baseApi.getResultIterator();
String lastUTF8Text;
float lastConfidence;
iterator.begin();
do {
lastUTF8Text = iterator.getUTF8Text(PageIteratorLevel.RIL_WORD);
lastConfidence = iterator.confidence(PageIteratorLevel.RIL_WORD);
if (lastConfidence > 50) {
Log.d(TAG, String.format("%s => %.2f", lastUTF8Text, lastConfidence));
}
} while (iterator.next(PageIteratorLevel.RIL_WORD));
baseApi.end();
Log.d(TAG, recognizedText);
return recognizedText;
}
public void ocr_main() {
Log.v(TAG, "entering tess");
// Getting uri of image/cropped image
try {
myimage = MediaStore.Images.Media.getBitmap(this.getContentResolver(), image_uri);
} catch (FileNotFoundException e) {
e.printStackTrace();
} catch (IOException e) {
e.printStackTrace();
}
// Log.v(TAG, "bitmap " + myimage.getByteCount());
// Bitmap argb = myimage;
// argb = argb.copy(Bitmap.Config.ARGB_8888, true);
// Log.v(TAG, "bitmap after argb:" + argb.getByteCount());
//
BitmapFactory.Options opt = new BitmapFactory.Options();
opt.inSampleSize = 2;
myimage = BitmapFactory.decodeFile(filepath, opt);
// Log.v(TAG, "bitmap after comp:" + myimage.getByteCount());
// TessBase starts
TessBaseAPI baseApi = new TessBaseAPI();
baseApi.setDebug(true);
baseApi.init(DATA_PATH, lang);
Log.v(TAG, "Before baseApi");
baseApi.setImage(myimage);
Log.v(TAG, "Before baseApi2");
String recognizedText = baseApi.getUTF8Text();
Log.v(TAG, "Before baseApi3");
baseApi.end();
Log.v(TAG, "OCRED TEXT: " + recognizedText);
if (lang.equalsIgnoreCase("eng")) {
recognizedText = recognizedText.replaceAll("[^a-zA-Z0-9]+", " ");
}
// recognizedText is the final OCRed text
recognizedText = recognizedText.trim();
// String ocrtext = "And...BAM! OCRed: " + recognizedText;
// Toast toast = Toast.makeText(this.getApplicationContext(), ocrtext,
// Toast.LENGTH_LONG);
// toast.show();
// deleting temporary crop file created
File file =
new File(
Environment.getExternalStoragePublicDirectory(Environment.DIRECTORY_PICTURES),
"temp.bmp");
boolean deleted = file.delete();
Log.i(TAG, "File deleted: " + deleted);
Intent intent = new Intent(this, ResultActivity.class);
intent.putExtra("ocrText", recognizedText);
startActivity(intent);
}
public String getTextContent() {
Log.d("whathaeljf", "The second image path is " + imagePath);
Bitmap bitmap = BitmapFactory.decodeFile(imagePath);
/*
Matrix matrix = new Matrix();
matrix.postRotate(180);
Bitmap rotatedBitmap = Bitmap.createBitmap(bitmap, 0, 0, bitmap.getWidth(), bitmap.getHeight(), matrix, true);
*/
TessBaseAPI baseApi = new TessBaseAPI();
baseApi.setDebug(true);
baseApi.init(DATA_PATH, lang);
baseApi.setImage(bitmap);
String recognizedText = baseApi.getUTF8Text();
return recognizedText;
}
/*Begin citation: http://gaut.am/making-an-ocr-android-app-using-tesseract/
gets the image from the external storage directory and uses the tess-two library
to process the image and return the recognized text
*/
public String getOCRText() {
Bitmap bitmap = BitmapFactory.decodeFile(DIR + "ocrImage.jpg");
TessBaseAPI baseApi = new TessBaseAPI();
baseApi.setDebug(true);
baseApi.init(DIR, "eng");
baseApi.setImage(bitmap);
String recognizedText = baseApi.getUTF8Text();
baseApi.end();
if ("eng".equalsIgnoreCase("eng")) {
recognizedText = recognizedText.replaceAll("[^a-zA-Z0-9 ]+", " ");
}
Log.d(DolchListActivity.TAG, recognizedText);
return recognizedText.trim();
}