feat: add new models MFD-1.5 & MFR-1.5

Update 2025.07.25: V1.1.4 Released

Major Changes:

• Upgraded the Mathematical Formula Detection (MFD) and Mathematical Formula Recognition (MFR) models to version 1.5. All default configurations, documentation, and examples now use mfd-1.5 and mfr-1.5 as the standard models.

主要变更：

• 数学公式检测（MFD）和数学公式识别（MFR）模型升级到 1.5 版本，所有默认配置、文档和示例均以 mfd-1.5 和 mfr-1.5 为标准模型。

feat: enhance image processing functions and add comprehensive tests for transparency handling

Update 2025.05.06: V1.1.3.2 Released

Major Changes:

• Fixed a potential error when processing transparent images, see #171 for details.

主要变更：

• 修复了处理透明图片时可能出现的错误，具体见 #171 。

fix: update version to 1.1.3.1 for VLM model import fix

Update 2025.04.27: V1.1.3.1 Released

Major Changes:

• Bugfix: Fixed the issue of model import related to VLM.

主要变更：

• 修复了 VLM 相关的模型导入问题。

feat: add VLM support for text and table recognition

Update 2025.04.15: V1.1.3 Released

Major Changes:

• Support for VlmTableOCR and VlmTextFormulaOCR models based on the VLM interface (see LiteLLM documentation) allowing the use of closed-source VLM models. Installation command: pip install pix2text[vlm].
  • Usage examples can be found in tests/test_vlm.py and tests/test_pix2text.py.

主要变更：

• 支持基于 VLM 接口（具体参考 LiteLLM 文档）的 VlmTableOCR 和 VlmTextFormulaOCR 模型，可使用闭源 VLM 模型。安装命令：pip install pix2text[vlm]。
  • 使用方式见 tests/test_vlm.py 和 tests/test_pix2text.py。

Bugfix: Fixed issues related to downloading models on Windows

Update 2024.12.17: V1.1.2.3 Released

Major Changes:

• Bugfix: Fixed issues related to downloading models on Windows.

主要变更：

• 修复了在 Windows 环境下下载模型的问题。

bugfix

Update 2024.12.11: V1.1.2.2 Released

Major Changes:

• Bugfix: Resolved issues related to serialization errors when handling ONNX Runtime session options by ensuring that non-serializable configurations are managed appropriately.

主要变更：

• 修复了与 ONNX Runtime session options 相关的序列化错误，通过确保不可序列化的配置信息在适当的管理下进行处理。

bugfix

Update 2024.12.02: V1.1.2.1 Released

Major Changes:

• Fixed an error in fetch_column_info()@DocYoloLayoutParser, thanks to Bin.

主要变更：

• 修复了 fetch_column_info()@DocYoloLayoutParser 中的错误，感谢网友 Bin 。

Integrated a better layout analysis model DocLayout-YOLO

Update 2024.11.17: V1.1.2 Released

Major Changes:

• A new layout analysis model DocLayout-YOLO has been integrated, improving the accuracy of layout analysis.
• Bug fixes:
  • When the text language is set to English only, a dedicated English OCR model is used to avoid including Chinese in the output.
  • The processing logic for PNG images has been optimized, enhancing recognition performance.

主要变更：

• 版面分析模型加入 DocLayout-YOLO，提升版面分析的准确性。
• 修复 bugs：
  • 在设置文本语言只有英语时，使用专门的英文 OCR 模型，避免输出中包含中文。
  • 对 PNG 图片的处理逻辑进行了优化，提升了识别效果。

Bugfixes

Update 2024.07.18: V1.1.1.2 Released

Major Changes:

• fix bugs：
  • #129
  • #116

主要变更：

• 修复 bugs：
  • #129
  • #116

Fix: some formats of models require fixed-size input images

Update 2024.06.24: V1.1.1.1 Released

Major Changes:

• Added a new parameter static_resized_shape when initializing MathFormulaDetector, which is used to resize the input image to a fixed size. Some formats of models require fixed-size input images during inference, such as CoreML.

主要变更：

• MathFormulaDetector 初始化时加入了参数 static_resized_shape, 用于把输入图片 resize 为固定大小。某些格式的模型在推理时需要固定大小的输入图片，如 CoreML。