def extract_image_data(data, languages=None):
"""Extract text from a binary string of data."""
if TESSDATA_PREFIX is None:
raise ValueError('Env TESSDATA_PREFIX is not set, OCR will not work.')
key, text = get_cache(data)
if text is not None:
return text
try:
img = Image.open(StringIO(data))
except Exception as ex:
log.debug('Failed to parse image internally: %r', ex)
return ''
# TODO: play with contrast and sharpening the images.
try:
languages = _get_languages(languages)
extractor = Tesseract(TESSDATA_PREFIX, lang=languages)
extractor.set_page_seg_mode(PageSegMode.PSM_AUTO_OSD)
text = extractor.ocr_image(img)
log.debug('OCR done: %s, %s characters extracted', languages,
len(text))
set_cache(key, text)
return text
except Exception as ex:
log.exception(ex)
return ''
def extract_image_data(data, languages=None):
"""Extract text from a binary string of data."""
if TESSDATA_PREFIX is None:
raise ValueError("Env TESSDATA_PREFIX is not set, OCR will not work.")
key, text = get_cache(data)
if text is not None:
return text
try:
img = Image.open(StringIO(data))
except Exception as ex:
log.debug("Failed to parse image internally: %r", ex)
return ""
# TODO: play with contrast and sharpening the images.
try:
languages = _get_languages(languages)
extractor = Tesseract(TESSDATA_PREFIX, lang=languages)
extractor.set_page_seg_mode(PageSegMode.PSM_AUTO_OSD)
text = extractor.ocr_image(img)
log.debug("OCR done: %s, %s characters extracted", languages, len(text))
set_cache(key, text)
return text
except Exception as ex:
log.exception(ex)
return ""
def _get_tesseract():
# FIXME: not currently loading small language sets.
languages = _get_languages(None)
if not hasattr(tess, 'instance'):
tess.instance = Tesseract(TESSDATA_PREFIX, languages)
tess.instance.set_page_seg_mode(PageSegMode.PSM_AUTO_OSD)
return tess.instance