Added barcode annotation tool

Author: yushulx

examples/official/annotation_tool/README.md

@@ -0,0 +1,110 @@
+# Barcode Annotation Tool
+
+A desktop application for annotating barcode locations in images. It automatically detects barcodes using **Dynamsoft Barcode Reader** and **ZXing-C++**, then lets you review, edit, and export the annotations.
+
+## Features
+
+- **Dual-engine detection** — Runs both Dynamsoft DBR and ZXing-C++ on every image, showing results side-by-side with differences highlighted.
+- **Drag-and-drop** — Drop images or folders onto the file list or directly onto the image display area.
+- **Adjustable quad overlays** — Every detected barcode quad has draggable corner handles. Drag any vertex to fine-tune the bounding polygon.
+- **4-click manual annotation** — Press **D** to enter draw mode, then click four points to define a custom quad. A dialog lets you enter barcode text and format.
+- **Delete / Delete All** — Remove the current image (and its annotations) or clear everything at once.
+- **Structured results panel** — A tree view for each barcode showing:
+  - Quad point coordinates (updated live when dragging)
+  - Dynamsoft result (text, format, coordinates)
+  - ZXing result (text, format, coordinates)
+  - Differences between the two engines
+- **JSON export** — Save all annotations to a JSON file that includes `total_images` and `total_barcodes` counts.
+
+## Requirements
+
+- Python 3.9+
+- A valid [Dynamsoft Barcode Reader license key](https://www.dynamsoft.com/customer/license/trialLicense/?product=dcv&package=cross-platform) 
+
+## Installation
+
+```bash
+pip install -r requirements.txt
+```
+
+## Usage
+
+```bash
+python main.py
+```
+
+### Keyboard Shortcuts
+
+| Key | Action |
+|---|---|
+| **←** / **→** | Previous / Next image |
+| **D** | Toggle draw mode |
+| **C** | Toggle crop mode |
+| **Esc** | Exit draw / crop mode or cancel current quad |
+| **Delete** | Remove current image |
+| **Scroll wheel** | Zoom in / out |
+| **Right-click** (draw mode) | Cancel current quad-in-progress |
+
+### Workflow
+
+1. **Load images** — Use *Load Files*, *Load Folder*, or drag-and-drop.
+2. **Review detections** — Images are automatically sorted into *Verified*, *Needs Review*, and *No Result* lists.
+3. **Fix mismatches** — Click an image in *Needs Review*, then edit text/format by clicking a polygon or drag its corner handles to correct the quad position.
+4. **Handle missed barcodes** — Click an image in *No Result* to re-run detection. If it still fails, use **Crop Mode** (drag a rectangle over the barcode) or **Draw Mode** (press **D**, click four corners, fill the dialog).
+5. **Verify** — Click *✔ Verify* to manually promote any image to the Verified list.
+6. **Import previous work** — Click *Import JSON* to reload a prior session's annotations onto the loaded images instantly.
+7. **Export / Save dataset** — Click *Export JSON* to save annotations for all images, or *Save Verified Dataset* to write only the Verified images and their annotations to a new folder.
+
+### Export Format
+
+```json
+{
+  "format": "barcode-benchmark/1.0",
+  "dataset": "Annotated Collection",
+  "total_images": 42,
+  "total_barcodes": 87,
+  "images": [
+    {
+      "file": "label_001.png",
+      "barcodes": [
+        {
+          "text": "1Z999AA10123456784",
+          "format": "Code128",
+          "points": [[x1,y1], [x2,y2], [x3,y3], [x4,y4]]
+        }
+      ]
+    }
+  ]
+}
+```
+
+
+## Annotation Viewer (`viewer.py`)
+
+A companion read-only viewer for inspecting exported `barcode-benchmark/1.0` JSON annotation files alongside the original images.
+
+Built with **tkinter** (Python built-in) and **Pillow** only, making it easy to distribute and run anywhere Python is installed.
+
+```bash
+pip install Pillow
+```
+
+### Usage
+
+```bash
+python viewer.py 
+```
+
+### Keyboard Shortcuts
+
+| Key | Action |
+|---|---|
+| **←** / **→** | Previous / Next image |
+| **O** | Open JSON annotation file |
+| **Scroll wheel** | Zoom in / out |
+
+### Workflow
+
+1. **Load images** — Use *Load Images…*, *Load Folder…*, or drag-and-drop image files/folders.
+2. **Load JSON** — Use *Open JSON…* or drag-and-drop a `.json` file exported by the annotation tool.
+3. **Browse** — Select any image in the list. If its filename appears in the JSON, overlays and barcode details are shown automatically.