def __langoptions__(self):
tesseract_list = pytesseract.get_languages(config='')
supported_langs = []
for k,v in lang_options.items():
if v in tesseract_list:
supported_langs.append(k)
return sorted(supported_langs)
def main(inputfile, args, lang = 'eng', width = 2000):
if lang == None: lang = 'eng' # Default to language to 'eng'
if args.scale: width = int(args.scale) # Set width scale to
if args.verbose:
print('File:', inputfile)
print('Starting ocr with lang:', lang)
if args.get_lang:
print('Languages:')
print(', '.join(pytesseract.get_languages()))
# Load image
image = cv2.imread(inputfile)
# Downscale image with interpolation
(h, w) = image.shape[:2]
r = width / float(w)
dim = (int(width), int(h * r))
image = cv2.resize(image, dim, interpolation = cv2.INTER_AREA)
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
if args.verbose:
print("Downscaled to ", dim)
# Perform OCR
result = pytesseract.image_to_string(image, args.lang).strip()
# Remove all special chars (--remove-special)
if args.remove_special:
if args.verbose:
print("Removing special chars")
for c in SPECIAL_CHARS:
result = result.replace(c, '')
# Output result (--verbose) or (--result)
if args.verbose and args.result:
print('OCR Result:', result)
elif args.result:
print(result)
# Write result to file (--out)
if args.out:
outFileName = inputfile+'.txt'
if args.verbose:
print("Outputting to", outFileName)
with open(outFileName, 'w') as fo:
fo.write(result)
def fetch_text_in_image(self, img):
# List of available languages
print(pytesseract.get_languages(config=''))
# NOTE: You can just load the image if it is one of the supported formats like:
# text = pytesseract.image_to_string(img)
# Load image with OpenCV
img_cv = cv2.imread(img)
# pytesseract needs images in RGB, opencv uses BGR - convert it!
img_rgb = cv2.cvtColor(img_cv, cv2.COLOR_BGR2RGB)
# Other data sources available from https://github.com/tesseract-ocr/tessdata
# tessdata_dir_config = r'--tessdata-dir "/tmp/screenshots" --psm 11'
# text = pytesseract.image_to_string(img_rgb, lang='eng', config=tessdata_dir_config)
# Inform the OCR engine that the "page" is not a document of text and process
tessdata_dir_config = r'--psm 11'
text = pytesseract.image_to_string(img_rgb,
lang='eng',
config=tessdata_dir_config)
return text
def get_tesseract_languages(tessdata_path) -> list[str]:
"""Get info abput tesseract setup."""
configure_tesseract_binary()
try:
languages = sorted(
pytesseract.get_languages(
config=get_tesseract_config(tessdata_path)))
except RuntimeError as e:
traceback.print_tb(e.__traceback__)
raise RuntimeError(
"Couldn't determine Tesseract information. If you pip installed NormCap "
+ "make sure Tesseract is installed and configured correctly."
) from e
if not languages:
raise ValueError(
"Could not load any languages for tesseract. " +
"On Windows, make sure that TESSDATA_PREFIX environment variable is set. "
+
"On Linux/MacOS see if 'tesseract --list-langs' work is the command line."
)
return languages
try:
from PIL import Image
except ImportError:
import Image
import pytesseract
# If you don't have tesseract executable in your PATH, include the following:
pytesseract.pytesseract.tesseract_cmd = r'<full_path_to_your_tesseract_executable>'
# Example tesseract_cmd = r'C:\Program Files (x86)\Tesseract-OCR\tesseract'
# Simple image to string
print(pytesseract.image_to_string(Image.open('test.png')))
# List of available languages
print(pytesseract.get_languages(config=''))
# French text image to string
print(pytesseract.image_to_string(Image.open('test-european.jpg'), lang='fra'))
# In order to bypass the image conversions of pytesseract, just use relative or absolute image path
# NOTE: In this case you should provide tesseract supported images or tesseract will return error
print(pytesseract.image_to_string('test.png'))
# Batch processing with a single file containing the list of multiple image file paths
print(pytesseract.image_to_string('images.txt'))
# Timeout/terminate the tesseract job after a period of time
try:
print(pytesseract.image_to_string('test.jpg',
timeout=2)) # Timeout after 2 seconds
print(pytesseract.image_to_string(
################
# SETUP
################
# create config file if not already present
if not configPath.exists():
print("No config found")
writeConfig()
readConfig()
# pytesseract version
print("[INFO] currently using tesseract: " + str(pytesseract.get_tesseract_version()))
# other languages can be installed by
# sudo apt install tesseract-ocr-[language code]
langs = pytesseract.get_languages(config="")
print(f"[INFO] following languages are availible:\n {langs}")
# get a numerical sorted list of all files and the total count
numbers = re.compile(r"(\d+)") # matches numerical token with multiple digits
files = sorted(listdir(path=src_image_path), key=numericalSort)
totalNumOfImages = len(files)
if enable_debug:
print(f'[INFO] {totalNumOfImages} image files are present in "{src_image_path}"')
print(f"[INFO] Following files are present:\n {files}")
################
# LOOP
################
# CHECK IF OCR CODE IS INCLUDED IN INSTALLED LANG's. NEEDS TO BE PRESENT FOR OCR!
def print_lang_codes():
langs = pytesseract.get_languages(config='')
for l in langs:
print(f"'{l}': ,")
# Text recognition from images using Pytesseract lib
import pytesseract
pytesseract.pytesseract.tesseract_cmd = r'D:\installations\pytesseract\inst\Tesseract-OCR\tesseract.exe'
import cv2
from PIL import Image
import tensorflow as tf
# Read and display image
image = cv2.imread('sample3.jpg')
img_rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
# Some functions of tesseract
print(pytesseract.image_to_string(img_rgb)) # Prints out the text
print(pytesseract.get_languages(config='')) # List of available languages
#--------- Prints out the letters and corresponding coordinates and create BB around them ------
print(pytesseract.image_to_boxes(
img_rgb)) # Prints out the letters and corresponding coordinates
bounding_box_coordinates = pytesseract.image_to_boxes(img_rgb)
image_h, image_w, _ = img_rgb.shape
for bounding_box in bounding_box_coordinates.splitlines():
print(type(bounding_box)) # string class
print(bounding_box)
bounding_box = bounding_box.split(' ')
x1, y1, x2, y2 = int(bounding_box[1]), int(bounding_box[2]), int(
bounding_box[3]), int(bounding_box[4])
cv2.rectangle(img_rgb, (x1, image_h - y1), (x2, image_h - y2), (0, 0, 255),
thickness=3)
def get_available_languages() -> set[str]:
try:
# Will throw an exception if tesseract_cmd is invalid
return set(tess.get_languages())
except EnvironmentError:
return AVAILABLE_LANGUAGES
}
</style>
""",
unsafe_allow_html=True)
st.markdown(f"""
<div class="container">
<img class="logo-img"
src="data:image/png;base64,{base64.b64encode(open(LOGO_IMAGE, "rb").read()).decode()}">
<p class="logo-text">GrumpyOCR</p>
</div>
""",
unsafe_allow_html=True)
tesseract_version = pytesseract.get_tesseract_version()
tesseract_languages = pytesseract.get_languages(config='')
st.markdown(f"Version: {tesseract_version}, Langs: {tesseract_languages}")
st.sidebar.markdown('### Настройки')
languages = ['eng+rus', 'rus', 'eng']
selected_languages = st.sidebar.selectbox('Язык', languages)
# Выбор файла с признаками обязателен для дальнейшего выполнения.
st.markdown("### 1 Выберите или перенесите файл с текстом")
upload_file_object = st.file_uploader('', ['pdf', 'png', 'jpg', 'jpeg'])
if not upload_file_object:
st.stop()
# Небольшой трюк с кэшированием в файл.
filename = upload_file_object.name
if not upload_file_object.closed:
