def extract_image_data(data, languages=None):
"""Extract text from a binary string of data."""
tessdata_prefix = get_config('TESSDATA_PREFIX')
if tessdata_prefix is None:
raise IngestorException("TESSDATA_PREFIX is not set, OCR won't work.")
languages = get_languages_iso3(languages)
text = Cache.get_ocr(data, languages)
if text is not None:
return text
try:
img = Image.open(StringIO(data))
except DecompressionBombWarning as dce:
log.debug("Image too large: %", dce)
return None
except IOError as ioe:
log.info("Unknown image format: %r", ioe)
return None
# TODO: play with contrast and sharpening the images.
extractor = Tesseract(tessdata_prefix, lang=languages)
extractor.set_page_seg_mode(PageSegMode.PSM_AUTO_OSD)
text = extractor.ocr_image(img)
extractor.clear()
log.debug('OCR done: %s, %s characters extracted',
languages, len(text))
Cache.set_ocr(data, languages, text)
return text
def extract_image_data(data, languages=None):
"""Extract text from a binary string of data."""
tessdata_prefix = get_config('TESSDATA_PREFIX')
if tessdata_prefix is None:
raise IngestorException("TESSDATA_PREFIX is not set, OCR won't work.")
languages = get_languages_iso3(languages)
text = Cache.get_ocr(data, languages)
if text is not None:
return text
try:
img = Image.open(StringIO(data))
except DecompressionBombWarning as dce:
log.debug("Image too large: %", dce)
return None
except IOError as ioe:
log.info("Unknown image format: %r", ioe)
return None
# TODO: play with contrast and sharpening the images.
extractor = Tesseract(tessdata_prefix, lang=languages)
extractor.set_image(img)
extractor.set_page_seg_mode(PageSegMode.PSM_AUTO_OSD)
text = extractor.get_text() or ''
text = text.decode(encoding="UTF-8")
# extractor.clear()
log.debug('OCR done: %s, %s characters extracted', languages, len(text))
Cache.set_ocr(data, languages, text)
return text
def extract_image_data(data, languages=None):
"""Extract text from a binary string of data."""
tessdata_prefix = get_config('TESSDATA_PREFIX')
if tessdata_prefix is None:
raise IngestorException("TESSDATA_PREFIX is not set, OCR won't work.")
languages = get_languages_iso3(languages)
text = Cache.get_ocr(data, languages)
if text is not None:
return text
img = Image.open(StringIO(data))
# TODO: play with contrast and sharpening the images.
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))
Cache.set_ocr(data, languages, text)
return text
def extract_image_data(data, languages=None):
"""Extract text from a binary string of data."""
tessdata_prefix = get_config('TESSDATA_PREFIX')
if tessdata_prefix is None:
raise IngestorException("TESSDATA_PREFIX is not set, OCR won't work.")
languages = get_languages_iso3(languages)
text = Cache.get_ocr(data, languages)
if text is not None:
return text
img = Image.open(StringIO(data))
# TODO: play with contrast and sharpening the images.
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))
Cache.set_ocr(data, languages, text)
return text