Delete .windsurf-rules

.windsurf-rules

@@ -1,109 +0,0 @@
-
-# Combined ANPR and Helmet Detection System
-
-## Overview
-
-This system integrates Automatic Number Plate Recognition (ANPR) for Indian vehicles with helmet detection for two-wheeler riders. It aims to enhance traffic safety monitoring by identifying vehicle registration numbers and checking for helmet usage in a single interface.
-
-## Rules and Guidelines
-
-1. **Input**: The system accepts images or video frames containing vehicles, preferably motorcycles or scooters.
-
-2. **ANPR Functionality**:
-   - Detects and reads license plates of Indian vehicles.
-   - Supports various Indian license plate formats.
-   - Provides the recognized license plate number as text.
-
-3. **Helmet Detection**:
-   - Identifies if the rider (and pillion rider, if present) is wearing a helmet.
-   - Returns a boolean value: True if helmet(s) detected, False otherwise.
-
-4. **Combined Output**:
-   - License Plate Number
-   - Helmet Status (Yes/No)
-   - Confidence scores for both detections
-
-5. **Error Handling**:
-   - If no license plate is detected, return "No plate detected"
-   - If no person is detected for helmet check, return "No rider detected"
-
-## Workflow
-
-1. User uploads an image or video frame to the system.
-2. System processes the image through both ANPR and helmet detection models simultaneously.
-3. ANPR model identifies and reads the license plate.
-4. Helmet detection model checks for the presence of helmets on riders.
-5. Results from both models are combined into a single output.
-6. The system displays the results to the user.
-
-## Usage Examples
-
-### Example 1: Compliant Rider
-
-**Input**: Image of a motorcycle with a clearly visible license plate and rider wearing a helmet.
-
-**Output**:
-```
-
-License Plate: DL 5S AB 1234
-Helmet Detected: Yes
-ANPR Confidence: 98%
-Helmet Detection Confidence: 95%
-
-```
-
-### Example 2: Non-compliant Rider
-
-**Input**: Image of a scooter with visible license plate but rider not wearing a helmet.
-
-**Output**:
-```
-
-License Plate: MH 01 AB 5678
-Helmet Detected: No
-ANPR Confidence: 97%
-Helmet Detection Confidence: 99%
-
-```
-
-### Example 3: Multiple Riders
-
-**Input**: Image of a motorcycle with two riders, both wearing helmets.
-
-**Output**:
-```
-
-License Plate: KA 01 EF 9876
-Helmet Detected: Yes
-ANPR Confidence: 96%
-Helmet Detection Confidence: 98%
-Note: Multiple helmets detected
-
-```
-
-### Example 4: Unclear Image
-
-**Input**: Blurry image of a vehicle with partially visible license plate.
-
-**Output**:
-```
-
-License Plate: ?N 02 X? 43??
-Helmet Detected: Uncertain
-ANPR Confidence: 60%
-Helmet Detection Confidence: 40%
-Note: Low quality image, results may be inaccurate
-
-```
-
-## Best Practices
-
-1. Use high-resolution images for better accuracy.
-2. Ensure proper lighting conditions in the input images.
-3. For video processing, select frames with clear views of both license plate and rider(s).
-4. Regularly update the model with new training data to improve accuracy.
-5. Use the confidence scores to filter out low-confidence detections if needed.
-
-By following these guidelines and understanding the workflow, users can effectively utilize this combined ANPR and helmet detection system for traffic safety monitoring and enforcement.
-```
-