LicensePlateFinder

LicensePlateFinder implements YOLOv5 to build a high-precision License Plate Detection Model. The model was fine-tuned on a custom dataset and achieved impressive performance:

• mAP@0.5:0.95: 0.762
• mAP@0.5: 0.831 (VOC metric)

---

Table of Contents

1. About the Dataset
2. Model and Preprocessing
3. Training Configuration
4. Evaluation Metrics
5. Inference Results
6. Real-Time Inference with OCR

---

About the Dataset

The dataset comprises 5,694 images of license plates with bounding box annotations provided in [xmin, ymin, xmax, ymax] format. It is organized as follows:

Dataset
├── train
│  └── Vehicle registration plate
│      └── Label
└── validation
    └── Vehicle registration plate
        └── Label

• Training Samples: 5,308
• Validation Samples: 386

Sample Images

Below are some images with ground truth bounding boxes:

---

Model and Preprocessing

The YOLOv5 model was chosen for its balance between speed and accuracy. Key steps in preprocessing included:

1. Annotation Conversion: Converted [xmin, ymin, xmax, ymax] Pascal VOC bounding box annotations into [x, y, width, height] YOLO format.
2. Data Configuration File: Prepared a data.yaml file with the following details:

train: /content/License_Plate_Detection/train/images
val: /content/License_Plate_Detection/validation/images
nc: 1
names: ['Reg-plate']

---

Training Configuration

The model was fine-tuned with the following hyperparameters:

• Batch Size: 32
• Epochs: 25
• Iterations: 4,146
  • calculated as int(epochs * train_img_count / batch_size)
• Initial Learning Rate: 1e-3

---

Evaluation Metrics

The final COCO primary metric (mAP@0.5:0.95) was 0.762, while the mAP@0.5 (VOC metric) was 0.831. The overall metric statistics is:

 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.762
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.831
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.830
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.222
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.512
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.862
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.669
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.798
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.798
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.233
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.577
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.895

The WandB logs can be found here.

---

Inference Results

---

Real-Time Inference with OCR

To further enhance the functionality of the license plate detection system, PaddleOCR was integrated to extract and display license plate text in real time. This allows for seamless detection and recognition of license plates.

Workflow

1. Detection: YOLOv5 detects license plates in images or video frames.
2. Cropping: The detected bounding box regions are extracted.
3. Text Recognition: PaddleOCR processes the cropped regions and extracts text.
4. Display: The extracted text is overlaid on the original frames for visualization.

Key Features

• Multi-Language Support: PaddleOCR supports various languages for text recognition.
• Real-Time Capability: The pipeline is optimized for real-time performance, making it suitable for applications such as traffic monitoring and automated toll systems.

Algorithm

The following steps outline the algorithm for real-time inference:

1. Input Frame:

   • Capture a video frame or load an image.

2. License Plate Detection:

   • Use YOLOv5 to detect bounding boxes for license plates.

3. Region Cropping:

   • Extract the regions corresponding to the detected bounding boxes.

4. OCR Processing:

   • Pass the cropped regions to PaddleOCR for text recognition.

5. Visualization:

   • Overlay the detected text on the original frame near the corresponding bounding box.

6. Output Frame:

   • Display or save the annotated frame with license plate numbers.

This integration showcases the power of combining advanced object detection with OCR for practical and impactful applications.

Final Output