图片文字识别不出来
收藏
目标图片如图:
使用如下脚本:
def ocr_to_word(source_image, target_image): ocr = PaddleOCR(use_angle_cls=True, lang="ch", use_gpu=False, det_model_dir=os.path.join(OCR_MODEL_PATH, "ch_PP-OCRv3_det_infer"), cls_model_dir=os.path.join(OCR_MODEL_PATH, "ch_ppocr_mobile_v2.0_cls_infer"), rec_model_dir=os.path.join(OCR_MODEL_PATH, "ch_PP-OCRv3_rec_infer")) result = ocr.ocr(source_image, cls=True) for idx in range(len(result)): res = result[idx] for line in res: print(line) # 显示结果 result = result[0] image = Image.open(source_image).convert('RGB') boxes = [line[0] for line in result] txts = [line[1][0] for line in result] scores = [line[1][1] for line in result] im_show = draw_ocr(image, boxes, txts, scores, font_path=f'{OCR_MODEL_PATH}/simfang.ttf') im_show = Image.fromarray(im_show) im_show.save(target_image)
控制台输出:
[2023/03/31 16:21:56] ppocr DEBUG: dt_boxes num : 0, elapse : 0.02344989776611328 [2023/03/31 16:21:56] ppocr DEBUG: cls num : 0, elapse : 0 [2023/03/31 16:21:56] ppocr DEBUG: rec_res num : 0, elapse : 0.0
图片结果为:
当前Python版本为3.8.16,paddleocr 版本2.6.1.3
0
收藏
请登录后评论