pytesseract_ocr_image_to_text_extraction_quickstart.py

1from PIL import Image
2import pytesseract
3
4# If you don't have tesseract executable in your PATH, include the following:
5# pytesseract.pytesseract.tesseract_cmd = r'<full_path_to_your_tesseract_executable>'
6# Example: pytesseract.pytesseract.tesseract_cmd = r'C:\Program Files\Tesseract-OCR\tesseract.exe'
7
8# Simple image to string
9print(pytesseract.image_to_string(Image.open('test.png')))
10
11# In order to bypass the image conversions of pytesseract, just use relative or absolute image path
12# NOTE: In this case you should check pytesseract version or identify ITesseracter interface
13# print(pytesseract.image_to_string('test.png'))
14
15# List of available languages
16print(pytesseract.get_languages(config=''))
17
18# French text image to string
19# print(pytesseract.image_to_string(Image.open('test-european.jpg'), lang='fra'))
20
21# Batch processing with a single file containing the list of images
22# print(pytesseract.image_to_string('images.txt'))
23
24# Get bounding box estimates
25# print(pytesseract.image_to_boxes(Image.open('test.png')))
26
27# Get verbose data including boxes, confidences, line and page numbers
28# print(pytesseract.image_to_data(Image.open('test.png')))
29
30# Get information about orientation and script detection
31# print(pytesseract.image_to_osd(Image.open('test.png')))
32
33# Get a searchable PDF
34# pdf = pytesseract.image_to_pdf_or_hocr('test.png', extension='pdf')
35# with open('test.pdf', 'w+b') as f:
36#     f.write(pdf) # pdf type is bytes by default
37
38# Get HOCR output
39# hocr = pytesseract.image_to_pdf_or_hocr('test.png', extension='hocr')
40
41# Get ALTO XML output
42# xml = pytesseract.image_to_alto_xml('test.png')