一个强大的屏幕文字识别和选择工具 / A powerful screen text recognition and selection tool
- 按行智能识别文本块 / Smart text block recognition by line
- 支持5像素垂直容差 / 5-pixel vertical tolerance support
- 自动合并相邻文本 / Auto-merge adjacent text
- 保持自然阅读顺序 / Maintain natural reading order
- 不添加空格的情况 / No Space Cases:
- 中文字符前后 / Around Chinese characters
- 标点符号前后 / Around punctuation marks
- 连续数字之间 / Between consecutive numbers
- 添加空格的情况 / Add Space Cases:
- 字母和数字之间(间距>10px)/ Between letters and numbers (gap>10px)
- 不同类型字符之间 / Between different character types
- 中英文混排时 / In Chinese-English mixed text
- 统一半透明高亮层 / Unified semi-transparent highlight layer
- 颜色:#4D94FF / Color: #4D94FF
- 透明度:30% / Opacity: 30%
- 流畅的选择体验 / Smooth selection experience
- 即时视觉反馈 / Instant visual feedback
- 按住Alt键0.3秒触发 / Hold Alt for 0.3s to trigger
- 拖动选择文本 / Drag to select text
- 自动复制到剪贴板 / Auto-copy to clipboard
- 主要引擎:PaddleOCR(支持中文)/ Primary: PaddleOCR (Chinese support)
- 备选引擎:Tesseract / Secondary: Tesseract
- Python 3.6+
- tkinter
- win32api/win32gui
- pytesseract
- PaddleOCR
- PIL (Pillow)
- BeautifulSoup
- 操作系统:Windows 10/11 / OS: Windows 10/11
- 需要安装Tesseract OCR / Requires Tesseract OCR installation
- 最小化CPU使用 / Minimal CPU usage
- 动态DPI缩放支持 / Dynamic DPI scaling support
- 高效的文本块管理 / Efficient text block management
- 优化的图形渲染 / Optimized graphics rendering
- 使用Windows低级键盘钩子 / Uses Windows low-level keyboard hooks
- 最小化持久状态 / Minimal persistent state
- 无外部数据传输 / No external data transmission
- 高DPI环境可能有坐标偏差 / High DPI environments might have coordinate variations
- OCR识别准确度依赖图像质量 / OCR accuracy depends on image quality
- 性能因系统配置而异 / Performance varies with system specifications
- 提升OCR准确度 / Improve OCR accuracy
- 增强多显示器支持 / Enhance multi-monitor support
- 添加可配置热键 / Add configurable hotkeys
- 实现更强大的错误处理 / Implement robust error handling
- 扩展语言支持 / Expand language support
欢迎提交Issue和Pull Request / Issues and Pull Requests are welcome