Bläddra i källkod

feat(config): add table classification debug options and update VLM server port

- Add debug_options to table_classification config for visualization control.
- Explicitly set VLM model_name to 'PaddleOCR-VL-0.9B'.
- Update VLM server_url port from 20016 to 20116.
zhch158_admin 2 dagar sedan
förälder
incheckning
a880f363ac
1 ändrade filer med 11 tillägg och 1 borttagningar
  1. 11 1
      ocr_tools/universal_doc_parser/config/bank_statement_wired_unet.yaml

+ 11 - 1
ocr_tools/universal_doc_parser/config/bank_statement_wired_unet.yaml

@@ -47,6 +47,15 @@ table_classification:
   confidence_threshold: 0.5   # 分类置信度阈值
   batch_size: 16              # 批处理大小
 
+  # Debug 可视化配置(与 MinerUWiredTableRecognizer.DebugOptions 对齐)
+  # 默认关闭。开启后将保存:表格线
+  debug_options:
+    enabled: true               # 是否开启调试可视化输出
+    output_dir: null             # 调试输出目录;null不输出
+    save_table_lines: true       # 保存表格线可视化(unet横线/竖线叠加)
+    image_format: "png"          # 可视化图片格式:png/jpg
+    prefix: ""                  # 保存文件名前缀(如设置为页码/表格序号)
+
 # 有线表格识别专用配置
 table_recognition_wired:
   use_wired_unet: true
@@ -77,10 +86,11 @@ table_recognition_wired:
 vl_recognition:
   # 可选: "mineru" (MinerU VLM) 或 "paddle" (PaddleOCR-VL)
   module: "paddle"
+  model_name: "PaddleOCR-VL-0.9B"
   
   # 后端配置
   backend: "http-client"  # 可选: "http-client", "vllm-engine", "transformers"
-  server_url: "http://10.192.72.11:20016"  # PaddleOCR-VL 服务地址
+  server_url: "http://10.192.72.11:20116"  # PaddleOCR-VL 服务地址
   
   # 图片尺寸限制(避免序列长度超限)
   max_image_size: 4096