Skip to content

周伟奇 / bmw-ocr

Go to a project
Toggle navigation
Toggle navigation pinning
7c50d819 by 周伟奇
Options

fix bug

1 parent 779b9a10
Inline Side-by-side
Showing 2 changed files with 21 additions and 20 deletions
src/apps/doc/management/commands/ocr_process.py
......@@ -102,10 +102,10 @@ class Command(BaseCommand, LoggerMixin):
raise EDMSException(edms_exc)
self.cronjob_log.info('{0} [edms download success] [pdf_path={1}]'.format(self.log_base, pdf_path))
def bs_process(self, wb, ocr_data, bs_summary, unknown_summary, classify, res_list, pno, ino):
def bs_process(self, wb, ocr_data, bs_summary, unknown_summary, classify, res_list, pno, ino, part_idx):
sheets = ocr_data.get('data', [])
if not sheets:
res_list.append((pno, ino, consts.RES_SUCCESS_EMPTY))
res_list.append((pno, ino, part_idx, consts.RES_SUCCESS_EMPTY))
return
# confidence = ocr_data.get('confidence', 1)
img_name = 'page_{0}_img_{1}'.format(pno, ino)
......@@ -171,22 +171,22 @@ class Command(BaseCommand, LoggerMixin):
ed_list.append(summary[6])
if cells_exists:
res_list.append((pno, ino, consts.RES_SUCCESS))
res_list.append((pno, ino, part_idx, consts.RES_SUCCESS))
else:
res_list.append((pno, ino, consts.RES_SUCCESS_EMPTY))
res_list.append((pno, ino, part_idx, consts.RES_SUCCESS_EMPTY))
def license1_process(self, ocr_data, license_summary, classify, res_list, pno, ino):
def license1_process(self, ocr_data, license_summary, classify, res_list, pno, ino, part_idx):
# 类别：'0'身份证， '1'居住证
license_data = ocr_data.get('data', [])
if not license_data:
res_list.append((pno, ino, consts.RES_SUCCESS_EMPTY))
res_list.append((pno, ino, part_idx, consts.RES_SUCCESS_EMPTY))
return
res_list.append((pno, ino, consts.RES_SUCCESS))
res_list.append((pno, ino, part_idx, consts.RES_SUCCESS))
license_summary.setdefault(classify, []).extend(license_data)
def license2_process(self, ocr_res_2, license_summary, pid, classify, res_list, pno, ino):
def license2_process(self, ocr_res_2, license_summary, pid, classify, res_list, pno, ino, part_idx):
if ocr_res_2.get('ErrorCode') in consts.SUCCESS_CODE_SET:
res_list.append((pno, ino, consts.RES_SUCCESS))
res_list.append((pno, ino, part_idx, consts.RES_SUCCESS))
if pid == consts.BC_PID:
# 银行卡
# res_dict = {}
......@@ -201,7 +201,7 @@ class Command(BaseCommand, LoggerMixin):
res_dict[field_dict.get('chn_key', '')] = field_dict.get('value', '')
license_summary.setdefault(classify, []).append(res_dict)
else:
res_list.append((pno, ino, consts.RES_FAILED_2))
res_list.append((pno, ino, part_idx, consts.RES_FAILED_2))
@staticmethod
def parse_img_path(img_path):
......@@ -493,26 +493,27 @@ class Command(BaseCommand, LoggerMixin):
wb = BSWorkbook(interest_keyword, salary_keyword, loan_keyword)
for img_path, res in ocr_1_res.items():
pno, ino = self.parse_img_path(img_path)
part_idx = 1
if res.get('code') == 1:
ocr_data_list = res.get('data', [])
if not isinstance(ocr_data_list, list):
res_list.append((pno, ino, consts.RES_FAILED_3))
res_list.append((pno, ino, part_idx, consts.RES_FAILED_3))
self.cronjob_log.info('{0} [ocr_1 res error] [img={1}] [res={2}]'.format(
self.log_base, img_path, res))
else:
for part_idx, ocr_data in enumerate(ocr_data_list):
ino_part = '{0}-{1}'.format(ino, part_idx)
part_idx = part_idx + 1
classify = ocr_data.get('classify')
if classify is None:
res_list.append((pno, ino_part, consts.RES_FAILED_3))
res_list.append((pno, ino, part_idx, consts.RES_FAILED_3))
self.cronjob_log.info('{0} [ocr_1 res error] [img={1}] [res={2}]'.format(
self.log_base, img_path, res))
continue
elif classify in consts.OTHER_CLASSIFY_SET: # 其他类
res_list.append((pno, ino_part, consts.RES_SUCCESS_OTHER))
res_list.append((pno, ino, part_idx, consts.RES_SUCCESS_OTHER))
continue
elif classify in consts.LICENSE_CLASSIFY_SET_1: # 证件1
self.license1_process(ocr_data, license_summary, classify, res_list, pno, ino_part)
self.license1_process(ocr_data, license_summary, classify, res_list, pno, ino, part_idx)
elif classify in consts.LICENSE_CLASSIFY_SET_2: # 证件2
pid, _, _, _, _, _ = consts.LICENSE_CLASSIFY_MAPPING.get(classify)
file_data = ocr_data.get('section_img')
......@@ -557,16 +558,16 @@ class Command(BaseCommand, LoggerMixin):
card_name_res.get('data', {}).get('is_exists_name') == 0:
name = '无'
ocr_2_res['Name'] = name
self.license2_process(ocr_2_res, license_summary, pid, classify, res_list, pno, ino_part)
self.license2_process(ocr_2_res, license_summary, pid, classify, res_list, pno, ino, part_idx)
break
else:
res_list.append((pno, ino_part, consts.RES_FAILED_2))
res_list.append((pno, ino, part_idx, consts.RES_FAILED_2))
self.cronjob_log.warn(
'{0} [ocr_2 failed] [img_path={1}]'.format(self.log_base, img_path))
else: # 流水处理
self.bs_process(wb, ocr_data, bs_summary, unknown_summary, classify, res_list, pno, ino_part)
self.bs_process(wb, ocr_data, bs_summary, unknown_summary, classify, res_list, pno, ino, part_idx)
else:
res_list.append((pno, ino, consts.RES_FAILED_1))
res_list.append((pno, ino, part_idx, consts.RES_FAILED_1))
self.cronjob_log.info('{0} [ocr_1 res error] [img={1}] [res={2}]'.format(
self.log_base, img_path, res))
......
src/apps/doc/ocr/wb.py
......@@ -582,7 +582,7 @@ class BSWorkbook(Workbook):
def res_sheet(self, res_list):
if res_list:
res_list.sort(key=lambda x: (x[0], x[1]))
res_list.sort(key=lambda x: (x[0], x[1], x[2]))
ws = self.create_sheet(consts.RES_SHEET_NAME)
ws.append(consts.RES_SHEET_HEADER)
for res_tuple in res_list:
......
Styling with Markdown is supported
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!