Ein OCR Modell auf meine Schrift trainieren

Englisches Modell

Für das Englische Modell habe ich bisher eine meiner Einglisch Klausuren in etwa 220 Teile unterteilt, welche vorerst verwendet werden um das Modell zu trainieren.

Um es zu testen, kann tesseract /mnt/d/Programming/tesstrain/data/eng_Erik-ground-truth/eva1_slice1.tif stdout --tessdata-dir /mnt/d/Programming/tesstrain/data -l eng_Erik --psm 7 ausgeführt werden.

Um zu trainieren in /mnt/d/Programming/tesstrain sein und make training MODEL_NAME=eng_Erik MAX_ITERATIONS=300000 ausführen

Timeline (in 10.000er Schritten)

Satz: The term postcolonialism means firstly the historic time and secondly

1. whes term posterlanialine man foriotly the intinnc bimm oad ay eeentle
2. Whwu tars postaolonallinm means fiestly the histinc thme and secorphld
3. Tiy tears posifaolioicalimm meams fiexstl the titisic tme and segorall
4. Thr term postficaolonialism meams fisitly the histac time and secoroll.
5. Tie term postoolonialism yeans fixistly the istiic time and sucarile,
6. TBVhe term postoclnalisms means fisstly the histisic thrme and segorsle
7. The ters postcoloniabiss wmeass frstty the hetoric tirse and secenedh
8. The temm posicolonialiasm means fistlty the tistomic tirme and seconsle
9. The tem postoeclonialiasm means fiat the iistoric time and seconoll
10. The term postcolonialism means frsstl the histosic time and sucoreole
11. The tems postoclonialism means fixstly the histosic thime and secorole
12. Bhe tm postaclonialisms means fisstt the tistom time and sweormls,
13. The therm postaclonialisrm means firty the thistoric thrme and secrowolld
14. The term postcolonialisms means fiisty ithe historic thame and secenol
15. Tdhe nsm postcolonialisms means firsttky the histric ptire and secoenoll
16. Te temm postcolonialism means firstly the tistosc time and secoreldl
17. The tamm postaclonialisms means firtty the histosic tim and secoroll
18. The tem postcclonialisms yeans firstty the historic thre and seconrold
19. The term pGostooclonialisms means fis iy the ristoric tire and ecrald
20. The term postoclonialism meass firsly the historic the and seceraled
21. The tesm postoclonialismm means firrtly the tistoric thime and seconcle
22. The temm postaclonialism means fisrttly the tistoric time and seconcle
23. TVhe term postcolonialisms means fxtly the historic time and secorcole
24. The term postcolonialims means frmesly the historic time and secorole
25. The ters postcolonialims means fisssly the tistoric time and seconole
26. Thie term postcolonialisss means ficstly the historic time and secanole
27. The term postcolomnialims vyeans firisttly the tistoric tirme and seconol
28. he ters postcclonialisss meams fisstly the tistoric time and seconcle
29. Tie term postoclonialisms means fiissly the historic time and secondl
30. The trrm postcoclonialisss means firstly the tietoric time and secondl

Und hier das normale Tesseract eng Ergebniss: Che Jom pethecheiwbly) on Aas, poe Nefevi. Jin, cath vec.

Nützliche Python scripts

Ein Datei Umbenenner

import os
import sys

if __name__ == '__main__':
    if len(sys.argv) not in [4, 3]:
        print('Usage: python file-renamer.py <search_string> <replace_string> [<directory>]')
        sys.exit(1)

    search_string = sys.argv[1]
    replace_string = sys.argv[2]
    directory = sys.argv[3] if len(sys.argv) == 4 else os.getcwd()

    count = 0
    for filename in os.listdir(directory):
        file_path = os.path.join(directory, filename)
        if os.path.isfile(file_path):
            new_filename = filename.replace(search_string, replace_string)
            new_file_path = os.path.join(directory, new_filename)
            if filename != new_filename:
                os.rename(file_path, new_file_path)
                count += 1

    print('Renamed {} files.'.format(count))

Dateien ohne File-Extension mit .gt.txt versehen

import os
import sys

def rename_files_in_directory(directory):
    """
    Renames files in a given directory by appending '.gt.txt' to their names.

    Args:
        directory (str): The path to the directory containing the files.

    Returns:
        None
    """
    current_directory = directory
    file_list = os.listdir(current_directory)

    for file_name in file_list:
        if not 'gt' in file_name and (file_name.endswith('.txt') or '.' not in file_name):
            if '.' not in file_name:
                new_name = file_name + '.gt.txt'
            else:
                new_name = file_name.replace('.txt', '.gt.txt')
            os.rename(os.path.join(current_directory, file_name),
                      os.path.join(current_directory, new_name))


if __name__ == "__main__":
    if len(sys.argv) > 1:
        directory = sys.argv[1]
        rename_files_in_directory(directory)
    else:
        rename_files_in_directory(os.getcwd())