def test_ocr():
ocr_file = lambda fn, mode: ocr(
imread(resource_filename('tests', 'data/%s' % fn)), mode)
s = ocr_file('tesseract-test1.jpg', False)
assert s.startswith(
'The (quick) [brown] {fox} jumps!\nOver the $43,456.78 <lazy> #90 dog\n'
)
assert s.endswith('preguicoso.') or s.endswith('preguieoso.') or s.endswith(
'preguigoso.') # NB: This actually depends on the version of Tesseract
s = ocr_file('tesseract-test2.png', False)
assert s.startswith('This is a lot of 12 point text to test the\nocr code and see if') or \
s.startswith('This is a lot of 12 point text to test the\ncor code and see if')
assert s.endswith('The quick\nbrown dog jumped over the lazy fox.')
def test_ocr():
ocr_file = lambda fn, mode: ocr(
imread(resource_filename('tests', 'data/%s' % fn)), mode)
s = ocr_file('tesseract-test1.jpg', False)
assert s.startswith(
'The (quick) [brown] {fox} jumps!\nOver the $43,456.78 <lazy> #90 dog\n'
)
assert s.endswith('preguicoso.') or s.endswith(
'preguieoso.') # NB: This actually depends on the version of Tesseract
s = ocr_file('tesseract-test2.png', False)
assert s.startswith(
'This is a lot of 12 point text to test the\nocr code and see if'
) or s.startswith(
'This is a lot of 12 point text to test the\ncor code and see if')
assert s.endswith('The quick\nbrown dog jumped over the lazy fox.')
# Since Tesseract 4.0 this test does not work, at least I did not find the way to limit the output character set
#s = ocr_file('tesseract-test1.jpg', True)
#assert s.startswith('T116 10111610 1111011111 110111') or s.startswith('T116 111111610 1111011111 110111')
def test_issue34():
ocr(np.asarray([]))
ocr(np.asarray([[]]))
ocr(np.asarray([[0]], dtype=np.uint8))