From f89f68abd167f845a3ac74d4752848e2e9f91a4f Mon Sep 17 00:00:00 2001
From: liuyebo <1515783401@qq.com>
Date: Wed, 10 Jul 2024 17:09:00 +0800
Subject: [PATCH] =?UTF-8?q?=E4=BC=98=E5=8C=96ocr=E6=A8=A1=E5=9E=8B?=
 =?UTF-8?q?=EF=BC=8C=E5=A2=9E=E5=8A=A0=E5=85=B3=E9=94=AE=E8=AF=8D?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 photo_mask.py | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/photo_mask.py b/photo_mask.py
index b041672..2bd4f15 100644
--- a/photo_mask.py
+++ b/photo_mask.py
@@ -9,7 +9,7 @@ from time import sleep
 import cv2
 import numpy as np
 import paddleclas
-from paddleocr import PaddleOCR
+from paddlenlp.utils.doc_parser import DocParser
 from sqlalchemy import update
 
 from auto_email.error_email import send_an_error_email
@@ -20,7 +20,7 @@ from config.ucloud import BUCKET
 from models import ZxPhrec, ZxPhhd
 from ucloud import ucloud
 
-OCR = PaddleOCR(use_angle_cls=False, lang="ch", show_log=False, gpu_id=1)
+DOC_PARSER = DocParser(use_gpu=True, device_id=1)
 
 
 def open_image(img_path):
@@ -258,7 +258,8 @@ def get_mask_layout(image, content):
 
     result = []
     try:
-        layouts = get_ocr_layout(OCR, temp_file.name)
+        # layouts = get_ocr_layout(OCR, temp_file.name)
+        layouts = DOC_PARSER.parse({"doc": temp_file.name})["layout"]
         if not layouts:
             # 无识别结果
             return result
@@ -273,6 +274,8 @@ def get_mask_layout(image, content):
                     result.append(find_box_of_value("交款人", layout))
                 if "文款人" in layout[1]:
                     result.append(find_box_of_value("文款人", layout))
+                if "购买方名称" in layout[1]:
+                    result.append(find_box_of_value("购买方名称", layout))
             return result
     except Exception as e:
         logging.error("涂抹时出错", exc_info=e)