Răsfoiți Sursa

refactor(bank_statement_configs): enhance bank statement configuration and remove obsolete file

- Updated the bank_statement_paddle_vl.yaml configuration to include additional supported formats and improved layout detection settings.
- Enhanced output configurations by adding options for saving page-specific outputs and enabling debug modes.
- Removed the deprecated bank_statement_wired_unet.yaml file to streamline the project structure and eliminate redundancy.
zhch158_admin 2 zile în urmă
părinte
comite
b8a691eb95

+ 40 - 32
ocr_tools/universal_doc_parser/config/bank_statement_paddle_vl.yaml

@@ -3,8 +3,8 @@ scene_name: "bank_statement"
 description: "银行交易流水、对账单等场景"
 
 input:
-  supported_formats: [".pdf", ".png", ".jpg"]
-  dpi: 200
+  supported_formats: [".pdf", ".png", ".jpg", ".jpeg", ".bmp", ".tiff"]
+  dpi: 200  # PDF转图片的DPI
 
 preprocessor:
   module: "mineru"
@@ -19,19 +19,39 @@ layout_detection:
   # module: "paddle"
   # model_name: "RT-DETR-H_layout_17cls"
   # model_dir: /Users/zhch158/workspace/repository.git/PaddleX/zhch/unified_pytorch_models/Layout/RT-DETR-H_layout_17cls.onnx  # 使用默认路径,或指定: "./Layout/RT-DETR-H_layout_17cls.onnx"
-  module: "mineru"
-  model_name: "layout"
-  model_dir: null  # 使用默认路径
-  device: "cpu"
   # batch_size: 4
   # conf: 0.1
   # iou: 0.45
+  module: "docling"
+  model_name: "docling-layout-old"
+  model_dir: ds4sd/docling-layout-old
+  device: "cpu"
+  conf: 0.3
+  num_threads: 4
+  
+  # 后处理配置
+  post_process:
+    # 将大面积文本块转换为表格(后处理)
+    convert_large_text_to_table: true  # 是否启用
+    min_text_area_ratio: 0.25         # 最小面积占比(25%)
+    min_text_width_ratio: 0.4         # 最小宽度占比(40%)
+    min_text_height_ratio: 0.3        # 最小高度占比(30%)
 
+  # Debug 可视化配置(与 MinerUWiredTableRecognizer.DebugOptions 对齐)
+  # 默认关闭。开启后将保存:layout检测结果
+  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
+    prefix: ""                  # 保存文件名前缀(如设置为页码)
+
+# ============================================================
+# VL识别配置(表格、公式)
+# ============================================================
 vl_recognition:
   module: "paddle"
   backend: "http-client"
   model_name: "PaddleOCR-VL-0.9B"
-  server_url: "http://10.192.72.11:8110"
+  server_url: "http://10.192.72.11:20016"
   max_image_size: 4096  # 🔧 添加:最大图片尺寸
   resize_mode: 'max'    # 🔧 添加:缩放模式 ('max' 保持宽高比, 'fixed' 固定尺寸)
   device: "cpu"
@@ -48,39 +68,27 @@ vl_recognition:
 ocr_recognition:
   module: "mineru" 
   language: "ch"
-  det_threshold: 0.3
-  unclip_ratio: 1.8
+  det_threshold: 0.6
+  unclip_ratio: 1.5
+  enable_merge_det_boxes: false
   batch_size: 8
   device: "cpu"
 
+# ============================================================
+# 输出配置
+# ============================================================
 output:
+  create_subdir: false
+  save_pdf_images: true
   save_json: true
+  save_page_json: true
   save_markdown: true
+  save_page_markdown: true
   save_html: true
   save_layout_image: true
   save_ocr_image: true
   draw_type_label: true
   draw_bbox_number: true
-  
-# 场景特定配置
-scene_config:
-  bank_statement:
-    table_structure: "single_column_list"
-    merged_cells: false
-    expected_columns: ["日期", "摘要", "收入", "支出", "余额"]
-    amount_validation: true
-    date_validation: true
-    
-  processing_rules:
-    # 表格处理规则
-    table_rules:
-      - detect_table_type: ["wired", "wireless"]  
-      - extract_header_automatically: true
-      - validate_amount_format: true
-      - merge_continuation_rows: true
-      
-    # OCR后处理规则  
-    ocr_rules:
-      - filter_low_confidence: 0.7
-      - merge_adjacent_text: true
-      - number_format_normalization: true
+  save_enhanced_json: true
+  normalize_numbers: true
+  debug_mode: true

+ 0 - 0
ocr_tools/universal_doc_parser/config/bank_statement_wired_unet.yaml → ocr_tools/universal_doc_parser/config/bank_statement_yusys_v3.yaml