Loading imgutils/ocr/recognize.py +5 −2 Original line number Diff line number Diff line Loading @@ -61,8 +61,11 @@ def decode(text_index, model: str, text_prob=None, is_remove_duplicate=False): def _text_recognize(image: ImageTyping, model: str = 'ch_PP-OCRv4_rec', is_remove_duplicate: bool = False) -> Tuple[str, float]: _ort_session = _open_ocr_recognition_model(model) expected_height = _ort_session.get_inputs()[0].shape[2] image = load_image(image, force_background='white', mode='RGB') r = 48 / image.height r = expected_height / image.height new_height = int(round(image.height * r)) new_width = int(round(image.width * r)) image = image.resize((new_width, new_height)) Loading @@ -70,7 +73,7 @@ def _text_recognize(image: ImageTyping, model: str = 'ch_PP-OCRv4_rec', input_ = np.array(image).transpose((2, 0, 1)).astype(np.float32) / 255.0 input_ = ((input_ - 0.5) / 0.5)[None, ...].astype(np.float32) _ort_session = _open_ocr_recognition_model(model) _input_name = _ort_session.get_inputs()[0].name _output_name = _ort_session.get_outputs()[0].name output, = _ort_session.run([_output_name], {_input_name: input_}) Loading Loading
imgutils/ocr/recognize.py +5 −2 Original line number Diff line number Diff line Loading @@ -61,8 +61,11 @@ def decode(text_index, model: str, text_prob=None, is_remove_duplicate=False): def _text_recognize(image: ImageTyping, model: str = 'ch_PP-OCRv4_rec', is_remove_duplicate: bool = False) -> Tuple[str, float]: _ort_session = _open_ocr_recognition_model(model) expected_height = _ort_session.get_inputs()[0].shape[2] image = load_image(image, force_background='white', mode='RGB') r = 48 / image.height r = expected_height / image.height new_height = int(round(image.height * r)) new_width = int(round(image.width * r)) image = image.resize((new_width, new_height)) Loading @@ -70,7 +73,7 @@ def _text_recognize(image: ImageTyping, model: str = 'ch_PP-OCRv4_rec', input_ = np.array(image).transpose((2, 0, 1)).astype(np.float32) / 255.0 input_ = ((input_ - 0.5) / 0.5)[None, ...].astype(np.float32) _ort_session = _open_ocr_recognition_model(model) _input_name = _ort_session.get_inputs()[0].name _output_name = _ort_session.get_outputs()[0].name output, = _ort_session.run([_output_name], {_input_name: input_}) Loading