Spaces:
Build error
Build error
Create README.md
Browse files
README.md
CHANGED
|
@@ -1,13 +1,62 @@
|
|
| 1 |
---
|
| 2 |
title: Subdivision Plan Analyzer
|
| 3 |
-
emoji:
|
| 4 |
colorFrom: blue
|
| 5 |
-
colorTo:
|
| 6 |
sdk: gradio
|
| 7 |
-
sdk_version:
|
| 8 |
app_file: app.py
|
| 9 |
pinned: false
|
| 10 |
license: mit
|
| 11 |
---
|
| 12 |
|
| 13 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
title: Subdivision Plan Analyzer
|
| 3 |
+
emoji: 📐
|
| 4 |
colorFrom: blue
|
| 5 |
+
colorTo: green
|
| 6 |
sdk: gradio
|
| 7 |
+
sdk_version: 4.19.2
|
| 8 |
app_file: app.py
|
| 9 |
pinned: false
|
| 10 |
license: mit
|
| 11 |
---
|
| 12 |
|
| 13 |
+
# 📐 Subdivision Plan Analyzer
|
| 14 |
+
|
| 15 |
+
An AI-powered tool to extract lot information from subdivision plans using OCR and computer vision.
|
| 16 |
+
|
| 17 |
+
## Features
|
| 18 |
+
|
| 19 |
+
- **Automatic Lot Detection**: Uses computer vision to identify lot boundaries
|
| 20 |
+
- **OCR Text Extraction**: Extracts lot numbers, dimensions, and areas
|
| 21 |
+
- **Smart Association**: Matches text to corresponding lot boundaries
|
| 22 |
+
- **Visual Annotation**: Shows detected lots with colored overlays
|
| 23 |
+
- **Data Export**: Download results as CSV file
|
| 24 |
+
- **Summary Statistics**: Provides lot count, total area, and averages
|
| 25 |
+
|
| 26 |
+
## How to Use
|
| 27 |
+
|
| 28 |
+
1. **Upload Image**: Upload a subdivision plan in PNG or JPG format
|
| 29 |
+
2. **Adjust Settings**:
|
| 30 |
+
- Scale: Set the plan scale (default 1:1000)
|
| 31 |
+
- Confidence: Adjust OCR confidence threshold (0.5-0.95)
|
| 32 |
+
3. **Process**: Click "Extract Lots" to analyze the plan
|
| 33 |
+
4. **Review Results**: Check the extracted data in the table
|
| 34 |
+
5. **Export**: Download the results as a CSV file
|
| 35 |
+
|
| 36 |
+
## What Gets Extracted
|
| 37 |
+
|
| 38 |
+
- **Lot Numbers**: 3-4 digit identifiers (100-9999)
|
| 39 |
+
- **Dimensions**: Frontage and depth measurements in meters
|
| 40 |
+
- **Areas**: Lot areas in square meters (m²)
|
| 41 |
+
- **Lot Types**: Classification as Standard, Corner, or Small lots
|
| 42 |
+
|
| 43 |
+
## Tips for Best Results
|
| 44 |
+
|
| 45 |
+
- Use high-resolution images with clear text
|
| 46 |
+
- Ensure lot numbers and measurements are legible
|
| 47 |
+
- Plans with consistent formatting work best
|
| 48 |
+
- Adjust confidence threshold if too many/few items detected
|
| 49 |
+
|
| 50 |
+
## Technical Details
|
| 51 |
+
|
| 52 |
+
- **OCR Engine**: EasyOCR with English language model
|
| 53 |
+
- **Image Processing**: OpenCV for boundary detection
|
| 54 |
+
- **Pattern Matching**: Regular expressions for data extraction
|
| 55 |
+
- **UI Framework**: Gradio for web interface
|
| 56 |
+
|
| 57 |
+
## Note
|
| 58 |
+
|
| 59 |
+
First run may take a few minutes as the OCR models download (approximately 64MB).
|
| 60 |
+
|
| 61 |
+
---
|
| 62 |
+
Made with ❤️ using Gradio and EasyOCR
|