chore: update readme

Author: Joker1212

README.md

@@ -15,43 +15,107 @@
 
 - **2024.11.15**
     - 完成初版代码，转换 [UVDoc](https://github.com/tanguymagne/UVDoc) 模型为onnx,完善前后处理
+- **2024.12.15**
+  - 补充去模糊/去阴影/二值化的功能和模型，重新升级为 RapidUndistort
 
 
 ### 简介
 
-本仓库用于进行文档扭曲的修正，同时能一定程度缓解透视和旋转问题
-
+本仓库用于进行文档扭曲的修正/文档去模糊/文档去阴影/文档二值化等问题 \ 
+提供多种模型和自由组合选择task，支持模型自动下载 \
+原始pytorch模型来源参考 [致谢](#致谢) \
+[快速使用](#快速使用) [使用建议](#使用建议) [参数说明](#参数说明) [模型地址](https://www.modelscope.cn/studios/jockerK/DocUnwrap/files)
 
 ### 在线体验
 [modelscope](https://www.modelscope.cn/studios/jockerK/DocUnwrap) [huggingface](https://huggingface.co/spaces/Joker1212/RapidUnwrap)
 ### 效果展示
-![res_show.jpg](preview.jpg)
+![res_show.jpg](preview1.gif)
+![res_show1.jpg](preview2.gif)
 
 ### 安装
 ``` python {linenos=table}
 # 建议使用清华源安装 https://pypi.tuna.tsinghua.edu.cn/simple
-pip install rapid-unwrap
+pip install rapid-undistorted
 ```
 
 ### 快速使用
 
 ``` python {linenos=table}
 import cv2
 
-from rapid_unwrap.inference import DocUnwrapper
-img_path = "img/demo4.jpg"
-doc_unwrapper = DocUnwrapper(img_path)
-unwrapped_img, elapse = doc_unwrapper(img_path)
+from rapid_undistorted.inference import InferenceEngine
+img_path = "img/demo.jpg"
+engine = InferenceEngine()
+# 扭曲修正->去阴影->去模糊 (指定去模糊模型)
+output_img, elapse = engine(img_path, ["unwrap", "unshadow", ("unblur", "OpenCvBilateral")])
+# 去阴影->去模糊 (指定去模糊模型)
+#output_img, elapse = engine(img_path, ["unshadow", ("unblur", "OpenCvBilateral")])
+# 默认选择yaml配置文件中第一个unblur模型
+#output_img, elapse = engine(img_path, ["unshadow", "unblur"])
+# 二值化替代去阴影方法
+#output_img, elapse = engine(img_path, ["unwrap", "binarize", "unblur"])
 print(f"doc unwrap elapse:{elapse}")
-cv2.imwrite("unwarped.png", unwrapped_img)
+cv2.imwrite("result.png", output_img)
 
 ```
 
+### 使用建议
+- 英文及数字去模糊 NAFDPM 模型更好，但中文直接使用opencv方法更佳
+- 去阴影模型相对于二值化功能更丰富，效果更好，不建议直接使用二值化方法
+
+
+### 参数说明
+#### 初始化参数
+支持传入一个config配置文件，声明需要的task类型和对应的模型，以及path
+config_path = "configs/config.yaml"
+engine = InferenceEngine(config_path)
+```yaml
+tasks:
+  unwrap:
+    models:
+      - type: "UVDoc"
+        path:
+        use_cuda: false
+        
+  unshadow:
+    models:
+      - type: "GCDnet"
+        sub_models:
+          - type: "GCDnet"
+            path:
+            use_cuda: false
+            use_dml: false
+          - type: "DRnet"
+            path:
+            use_cuda: false
+
+  binarize:
+    models:
+      - type: "UnetCnn"
+        path:
+        use_cuda: false
+
+  unblur:
+    models:
+      - type: "OpenCvBilateral"
+        path:
+      - type: "NAFDPM"
+        path:
+        use_cuda: false
 
-### 致谢
+```
+#### 执行参数
+```python
+engine(img_path, task_list)
+engine(img_path, ["unwrap", "unshadow", ("unblur", "OpenCvBilateral")])
+```
 
-[UVDoc](https://github.com/tanguymagne/UVDoc)
+### 致谢
 
+unwrap: [UVDoc](https://github.com/tanguymagne/UVDoc)
+unshadow: [GCDnet](https://github.com/ZZZHANG-jx/GCDRNet)
+unblur: [NAFDPM](https://github.com/ispamm/NAF-DPM)
+binarize: [UnetCnn](https://github.com/sajjanvsl/U-Net-CNN-for-binarization-of-Historical-Kannada-Handwritten-Palm-Leaf-Manuscripts)
 
 ### 贡献指南
 

README_en.md

@@ -13,40 +13,109 @@
 
 - **2024.11.15**
   - Completed the initial version of the code, converted the [UVDoc](https://github.com/tanguymagne/UVDoc) model to onnx, and improved pre- and post-processing.
-
+- **2024.12.15**
+  - Added deblurring/shadow removal/binarization functions and models, upgraded to RapidUndistort.
+  
 ### Introduction
 
-This repository is used for correcting document distortions and can also alleviate perspective and rotation issues to some extent.
+This repository is used for correcting document distortion, deblurring documents, shadow removal, and document binarization.
+It provides multiple models and flexible task combinations, supports automatic model downloading.
+Original PyTorch model sources can be found in the [Acknowledgments](#acknowledgments) section.
+[Quick Start](#quick-start) [Usage Suggestions](#usage-suggestions) [Parameter Explanation](#parameter-explanation) [Model Address](https://www.modelscope.cn/studios/jockerK/DocUnwrap/files)
 
-### Online Experience
+### Online Demo
 [modelscope](https://www.modelscope.cn/studios/jockerK/DocUnwrap) [huggingface](https://huggingface.co/spaces/Joker1212/RapidUnwrap)
 
-### Effect Demonstration
-![res_show.jpg](preview.jpg)
+### Effect Showcase
+![res_show.jpg](preview1.gif)
+![res_show1.jpg](preview2.gif)
 
 ### Installation
 ``` python {linenos=table}
-pip install rapid-unwrap
+pip install rapid-undistorted
 ```
 
 ### Quick Start
 
 ``` python {linenos=table}
 import cv2
 
-from rapid_unwrap.inference import DocUnwrapper
-img_path = "img/demo4.jpg"
-doc_unwrapper = DocUnwrapper(img_path)
-unwrapped_img, elapse = doc_unwrapper(img_path)
+from rapid_undistorted.inference import InferenceEngine
+img_path = "img/demo.jpg"
+engine = InferenceEngine()
+# Distortion correction -> Shadow removal -> Deblurring (specify deblurring model)
+output_img, elapse = engine(img_path, ["unwrap", "unshadow", ("unblur", "OpenCvBilateral")])
+# Shadow removal -> Deblurring (specify deblurring model)
+#output_img, elapse = engine(img_path, ["unshadow", ("unblur", "OpenCvBilateral")])
+# Default selection of the first unblur model in the yaml configuration file
+#output_img, elapse = engine(img_path, ["unshadow", "unblur"])
+# Binarization as an alternative to shadow removal method
+#output_img, elapse = engine(img_path, ["unwrap", "binarize", "unblur"])
 print(f"doc unwrap elapse:{elapse}")
-cv2.imwrite("unwarped.png", unwrapped_img)
+cv2.imwrite("result.png", output_img)
+
+```
+
+### Usage Suggestions
+- For English and numeric deblurring, the NAFDPM model is better, but for Chinese text, using the OpenCV method is more suitable.
+- The shadow removal model has richer functionality and better results compared to binarization, so it is not recommended to directly use the binarization method.
 
+
+### Parameter Explanation
+#### Initialization Parameters
+Supports passing a config configuration file to declare the required task types and corresponding models, as well as paths.
+```python
+config_path = "configs/config.yaml"
+engine = InferenceEngine(config_path)
 ```
+```yaml
+tasks:
+  unwrap:
+    models:
+      - type: "UVDoc"
+        path:
+        use_cuda: false
+        
+  unshadow:
+    models:
+      - type: "GCDnet"
+        sub_models:
+          - type: "GCDnet"
+            path:
+            use_cuda: false
+            use_dml: false
+          - type: "DRnet"
+            path:
+            use_cuda: false
+
+  binarize:
+    models:
+      - type: "UnetCnn"
+        path:
+        use_cuda: false
+
+  unblur:
+    models:
+      - type: "OpenCvBilateral"
+        path:
+      - type: "NAFDPM"
+        path:
+        use_cuda: false
 
+```
+#### Execution Parameters
+```python
+engine(img_path, task_list)
+engine(img_path, ["unwrap", "unshadow", ("unblur", "OpenCvBilateral")])
+```
 
 ### Acknowledgments
 
-[UVDoc](https://github.com/tanguymagne/UVDoc)
+unwrap: [UVDoc](https://github.com/tanguymagne/UVDoc)
+unshadow: [GCDnet](https://github.com/ZZZHANG-jx/GCDRNet)
+unblur: [NAFDPM](https://github.com/ispamm/NAF-DPM)
+binarize: [UnetCnn](https://github.com/sajjanvsl/U-Net-CNN-for-binarization-of-Historical-Kannada-Handwritten-Palm-Leaf-Manuscripts)
+
 
 
 ### Contribution Guidelines

preview.jpg

@@ -17,7 +17,7 @@ def __init__(self, config):
     def __call__(self, img):
         s = time.time()
         size = img.shape[:2][::-1]
-        img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB).astype(np.float32) / 255
+        img = img.astype(np.float32) / 255
         inp = self.preprocess(img.copy())
         outputs, _ = self.session([inp])
         elapse = time.time() - s