Histórico de Commits

Autor SHA1 Mensagem Data
  zhch158_admin dd92babb27 feat: 改进文档处理流程,支持从 PDF 提取文本并与 OCR 结果对比,添加调试模式 há 3 dias atrás
  zhch158_admin dc9a615776 fix: 修复加载和分类文档时的参数传递,添加 renderer_used 参数 há 3 dias atrás
  zhch158_admin 85f5dfa1f4 feat: 更新 process_text_element 方法,改进预匹配 spans 的处理逻辑,支持 OCR 和 PDF 文本提取来源 há 3 dias atrás
  zhch158_admin 43d0e1c5d3 feat: 在 bank_statement_wired_unet 配置中启用调试模式 há 3 dias atrás
  zhch158_admin db1a81a141 feat: 添加 PDF 文档类型检测功能,支持 pypdfium2 和 fitz 渲染引擎,优化文本提取过程 há 3 dias atrás
  zhch158_admin 939c825128 feat: Add `.gitignore` to exclude common development artifacts and specific project paths, and update `main_v2.py`. há 3 dias atrás
  zhch158_admin ae0a19dc4d chore: Add .gitignore to exclude various development and output files, and update main_v2.py. há 4 dias atrás
  zhch158_admin 481b5ea371 feat: add .gitignore to exclude common development artifacts, temporary files, and specific project outputs. há 4 dias atrás
  zhch158_admin a4b8405df5 feat: 添加 bank_statement_wired_unet OCR 工具配置,支持结果目录和图像目录定义 há 4 dias atrás
  zhch158_admin f90f868f20 feat: 添加边缘线过滤功能,优化线段提取过程以减少噪声 há 4 dias atrás
  zhch158_admin b1d0bc2173 feat: 添加倾斜矫正单元测试,验证倾斜检测和矫正功能的正确性 há 4 dias atrás
  zhch158_admin fd2b0bf294 fix: 修复倾斜角度计算和矫正逻辑,确保符号约定一致性 há 4 dias atrás
  zhch158_admin dce163b944 fix: 修复路径导入问题,确保项目根目录被正确添加到系统路径 há 4 dias atrás
  zhch158_admin 3263321e84 feat: 添加统一的PDF图像加载函数,支持多种渲染引擎 há 1 semana atrás
  zhch158_admin 4e6c855b17 feat: 添加PDF渲染引擎对比分析工具,支持分析图像属性和差异 há 1 semana atrás
  zhch158_admin 95e2272ed9 fix: 添加PDF文档处理时保存页面图像的功能 há 1 semana atrás
  zhch158_admin e21b57e051 fix: 更新输出配置,启用保存布局和OCR图像 há 1 semana atrás
  zhch158_admin f75f1bb639 Fix typo in documentation regarding U-Net line detection and comparison with existing methods. há 1 semana atrás
  zhch158_admin af1c467c48 fix: 优化UNet推理调试信息记录,增强尺寸一致性验证,返回检测到的倾斜角度 há 1 semana atrás
  zhch158_admin 6be6c8c3bb fix: 移除UNet预处理的宽高缩放因子参数,优化坐标转换的调试信息记录 há 1 semana atrás
  zhch158_admin ef447c6c7b fix: 从识别结果中获取倾斜角度,优化表格处理的准确性 há 1 semana atrás
  zhch158_admin 64652051e4 fix: 更新示例输入输出路径,修正注释以提高代码可读性 há 1 semana atrás
  zhch158_admin ca720abd31 fix: 增强UNet预处理的缩放因子验证,优化预测结果的尺寸一致性检查,记录详细的调试信息以确保坐标转换的准确性 há 1 semana atrás
  zhch158_admin 1fbcf06f4a fix: 增强文本填充器的OCR检测能力,支持跨单元格检测和输出调试图像,优化重叠检测逻辑 há 1 semana atrás
  zhch158_admin bb0acb2afc fix: 增强网格结构恢复中的坐标转换精度,添加调试信息以验证缩放比例和单元格覆盖情况 há 1 semana atrás
  zhch158_admin 3cf3aa5085 fix: 调整表格处理中的padding策略,优化边缘保护与噪声控制的平衡 há 1 semana atrás
  zhch158_admin 2f5c74136e fix: 优化 crop_region 方法中的代码格式,提升可读性 há 1 semana atrás
  zhch158_admin 0102386803 fix: Update OCR confidence threshold in bank_statement_wired_unet.yaml to improve cell recognition accuracy há 1 semana atrás
  zhch158_admin 652b321bd6 feat: Update batch processing in main_v2.py to include output directory parameter for document processing, enhancing flexibility in file management. há 1 semana atrás
  zhch158_admin 1bb438fba3 fix: Improve coordinate transformation accuracy in WiredTableVisualizer to reduce cumulative errors and enhance debugging with detailed logging of cell coordinates during visualization. há 1 semana atrás