付_工商银行943825图.yaml 2.2 KB

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465666768
  1. document:
  2. name: "付_工商银行943825图"
  3. base_dir: "/Users/zhch158/workspace/data/流水分析/付_工商银行943825图"
  4. # 🎯 关键改进:定义该文档使用的 OCR 工具及其结果目录
  5. ocr_results:
  6. # bank_statement_yusys_v3
  7. - tool: "mineru"
  8. result_dir: "bank_statement_yusys_v3"
  9. image_dir: "bank_statement_yusys_v3/{{name}}"
  10. description: "YUSYS-OCR框架 v3.0"
  11. enabled: true
  12. # bank_statement_mineru_vl
  13. - tool: "mineru"
  14. result_dir: "bank_statement_mineru_vl"
  15. image_dir: "bank_statement_mineru_vl/{{name}}"
  16. description: "YUSYS统一OCR框架mineru_vl"
  17. enabled: true
  18. # # MinerU
  19. # - tool: "mineru"
  20. # result_dir: "mineru_vllm_results"
  21. # image_dir: "mineru_vllm_results/{{name}}"
  22. # description: "MinerU 图片合成结果"
  23. # enabled: true
  24. # # MinerU (带 cell bbox)
  25. # - tool: "mineru"
  26. # result_dir: "mineru_vllm_results_cell_bbox"
  27. # image_dir: "mineru_vllm_results/{{name}}"
  28. # description: "MinerU + PaddleOCR 坐标"
  29. # enabled: true
  30. # # PaddleOCR-VL
  31. # - tool: "paddleocr_vl"
  32. # result_dir: "paddleocr_vl_results"
  33. # image_dir: "paddleocr_vl_results/{{name}}"
  34. # description: "PaddleOCR VLM 图片合成结果"
  35. # enabled: true
  36. # # PaddleOCR-VL (带 cell bbox)
  37. # - tool: "mineru" # 格式同 MinerU
  38. # result_dir: "paddleocr_vl_results_cell_bbox"
  39. # image_dir: "paddleocr_vl_results/{{name}}"
  40. # description: "PaddleOCR VLM + PaddleOCR 坐标"
  41. # enabled: true
  42. # # DotsOCR
  43. # - tool: "dots_ocr"
  44. # result_dir: "dotsocr_vllm_results"
  45. # image_dir: "dotsocr_vllm_results/{{name}}"
  46. # description: "Dots OCR 图片合成结果"
  47. # enabled: true
  48. # # DotsOCR (带 cell bbox)
  49. # - tool: "mineru"
  50. # result_dir: "dotsocr_vllm_results_cell_bbox"
  51. # image_dir: "dotsocr_vllm_results/{{name}}"
  52. # description: "Dots OCR + PaddleOCR 坐标"
  53. # enabled: true
  54. # # PPStructV3
  55. # - tool: "ppstructv3"
  56. # result_dir: "ppstructurev3_client_results"
  57. # image_dir: "ppstructurev3_client_results/{{name}}"
  58. # description: "PPStructV3 图片合成结果"
  59. # enabled: true