shiftlab-ocr 0.3.1

Generates text from image

shiftlab_ocr

SHIFT OCR is a library fo handwriting text segmentation and character recognition.

Get Started

Doc2Text

it transcribes an image into text

pip install shiftlab_ocr

from shiftlab_ocr import Scanner

PATH_TO_IMAGE = 'test.jpg'
scanner = Scanner(ocr_model='hw-cyr')
result = scanner.doc2text(PATH_TO_IMAGE)

('Директору Заявление 10 январе 2019г. Ирл Иванов А.П. ',
 [<shift_ocr.crop.Crop at 0x7f158cffd1d0>,
  <shift_ocr.crop.Crop at 0x7f158cffd610>,
  <shift_ocr.crop.Crop at 0x7f158cffd790>,
  <shift_ocr.crop.Crop at 0x7f158cffd8d0>,
  <shift_ocr.crop.Crop at 0x7f158cffdd50>,
  <shift_ocr.crop.Crop at 0x7f158cffd910>])

Generator of handwriting

It generates handwriting script with random backgrounds and handwriting fonts with given list of strings

from shiftlab_ocr import Generator

g = Generator(lang='ru')
g.upload_source('/content/source.txt')

s = g.generate_from_string('Москва',min_length=4,max_length=24) # get from a string
s

b = g.generate_batch(12,4,13) # get batch of random samples from source.txt
fig=plt.figure(figsize=(10, 10))
rows = int(len(b)/4) + 2
columns = int(len(b)/8) + 2
for i in range(len(b)):
  fig.add_subplot(rows, columns, i+1)
  plt.imshow(np.asarray(b[i][0])) 

Also, see Google Colab Demo