# platedet
`platedet` is a lightweight package for detecting license plates in images using an ONNX model. It is designed to be part of a pipeline for detecting, cropping, and reading license plates. The underlying model is just YOLO, converted to ONNX format. The training data comprises primarily Brazilian license plates, sourced from internet images.
## Installation
To install the required dependencies, use the following command:
For cpu
```bash
pip install "platedet[cpu]"
```
For cuda 11.X
```bash
pip install "platedet[gpu]"
```
For cuda 12.X
```bash
pip install "platedet[gpu]" --extra-index-url https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/onnxruntime-cuda-12/pypi/simple/
```
## Usage
### Command Line Interface
You can use the command line interface to detect license plates in an image:
```bash
platedet image_path [--model_path MODEL_PATH] [--return_types RETURN_TYPES] [--providers PROVIDERS] [--save_images] [--show_result]
```
#### Arguments
- `image_path`: Path to the input image.
- `--model_path`: Path to the ONNX model (default: `artifacts/platedet.onnx`).
- `--return_types`: Output formats (choices: `raw`, `pil`, `np_array`, `boxes`).
- `--providers`: ONNX Runtime providers (default: `CPUExecutionProvider`).
- `--save_images`: Save cropped images of detected plates.
- `--show_result`: Display results with bounding boxes and confidence scores.
### Example
To detect license plates and save the cropped images:
```bash
platedet path/to/image.png --save_images
```
To display the results with bounding boxes:
```bash
platedet detect.py path/to/image.png --show_result
```
### Using in Code
```python
from PIL import Image
from platedet import Platedet
platedet = Platedet()
image = Image.open('examples/1.jpg')
crops = platedet.inference(image, return_types=['pil'])
for idx, crop in enumerate(crops['pil']['images']):
crop.save(f'{idx}.jpg')
```
If you want to use CUDA:
```python
from PIL import Image
from platedet import Platedet
platedet = Platedet(providers=["CUDAExecutionProvider"])
image = Image.open('examples/1.jpg')
crops = platedet.inference(image, return_types=['pil'])
for idx, crop in enumerate(crops['pil']['images']):
crop.save(f'{idx}.jpg')
```
Check all execution providers [here](https://onnxruntime.ai/docs/execution-providers/).
Raw data
{
"_id": null,
"home_page": "https://github.com/pstwh/platedet",
"name": "platedet",
"maintainer": null,
"docs_url": null,
"requires_python": "<4,>=3.5",
"maintainer_email": null,
"keywords": "plate, detection, brazilian",
"author": null,
"author_email": null,
"download_url": null,
"platform": null,
"description": "\n# platedet\n\n`platedet` is a lightweight package for detecting license plates in images using an ONNX model. It is designed to be part of a pipeline for detecting, cropping, and reading license plates. The underlying model is just YOLO, converted to ONNX format. The training data comprises primarily Brazilian license plates, sourced from internet images.\n\n## Installation\n\nTo install the required dependencies, use the following command:\n\nFor cpu\n\n```bash\npip install \"platedet[cpu]\"\n```\n\nFor cuda 11.X\n```bash\npip install \"platedet[gpu]\"\n```\n\nFor cuda 12.X\n```bash\npip install \"platedet[gpu]\" --extra-index-url https://aiinfra.pkgs.visualstudio.com/PublicPackages/_packaging/onnxruntime-cuda-12/pypi/simple/\n```\n\n## Usage\n\n### Command Line Interface\n\nYou can use the command line interface to detect license plates in an image:\n\n```bash\nplatedet image_path [--model_path MODEL_PATH] [--return_types RETURN_TYPES] [--providers PROVIDERS] [--save_images] [--show_result]\n```\n\n#### Arguments\n\n- `image_path`: Path to the input image.\n- `--model_path`: Path to the ONNX model (default: `artifacts/platedet.onnx`).\n- `--return_types`: Output formats (choices: `raw`, `pil`, `np_array`, `boxes`).\n- `--providers`: ONNX Runtime providers (default: `CPUExecutionProvider`).\n- `--save_images`: Save cropped images of detected plates.\n- `--show_result`: Display results with bounding boxes and confidence scores.\n\n### Example\n\nTo detect license plates and save the cropped images:\n\n```bash\nplatedet path/to/image.png --save_images\n```\n\nTo display the results with bounding boxes:\n\n```bash\nplatedet detect.py path/to/image.png --show_result\n```\n\n### Using in Code\n\n```python\nfrom PIL import Image\nfrom platedet import Platedet\n\nplatedet = Platedet()\nimage = Image.open('examples/1.jpg')\ncrops = platedet.inference(image, return_types=['pil'])\nfor idx, crop in enumerate(crops['pil']['images']):\n crop.save(f'{idx}.jpg')\n```\n\nIf you want to use CUDA:\n```python\nfrom PIL import Image\nfrom platedet import Platedet\n\nplatedet = Platedet(providers=[\"CUDAExecutionProvider\"])\nimage = Image.open('examples/1.jpg')\ncrops = platedet.inference(image, return_types=['pil'])\nfor idx, crop in enumerate(crops['pil']['images']):\n crop.save(f'{idx}.jpg')\n```\n\nCheck all execution providers [here](https://onnxruntime.ai/docs/execution-providers/).\n",
"bugtrack_url": null,
"license": null,
"summary": "Detect license plates",
"version": "0.0.2",
"project_urls": {
"Homepage": "https://github.com/pstwh/platedet"
},
"split_keywords": [
"plate",
" detection",
" brazilian"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "50370ecc8ab264aff7c5e6478edbd35fc4cbd15374f7272f22669aea8a188f2d",
"md5": "2d78951b386d45ba77b6ba827b943636",
"sha256": "c39c602c4824809fa2e2f6c213f1fb1211b966bc99776c16c41d80caadf80d4f"
},
"downloads": -1,
"filename": "platedet-0.0.2-py3-none-any.whl",
"has_sig": false,
"md5_digest": "2d78951b386d45ba77b6ba827b943636",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<4,>=3.5",
"size": 21922638,
"upload_time": "2024-07-26T21:12:33",
"upload_time_iso_8601": "2024-07-26T21:12:33.726625Z",
"url": "https://files.pythonhosted.org/packages/50/37/0ecc8ab264aff7c5e6478edbd35fc4cbd15374f7272f22669aea8a188f2d/platedet-0.0.2-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-07-26 21:12:33",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "pstwh",
"github_project": "platedet",
"github_not_found": true,
"lcname": "platedet"
}
JSON
