# Remote Sensing Analysis
This package provides tools for processing and analyzing satellite imagery, utilizing advanced machine learning techniques for object detection, image enhancement, and text analytics from images.
## Installation
1. **Download Package**:
To install the Remote Sensing Analysis package, simply run the following command:
```bash
pip install Remote-Sensing-Analysis
```
2. **Download Model Weights**:
The package requires specific model weights to function correctly. Download the model weights from the following Google Drive link:
[Download Model Weights](https://drive.google.com/file/d/1KL3H-Fe1SVoCEFaO4KM4J_FMRF4ocoCz/view?usp=sharing)
After downloading, place the weights under the `pretrained` folder.
## Usage
### Parameters
When initializing the `ImageProcessor`, you can specify the following parameters:
- **model_weights_path**: Path to the model weights file, default is `"pretrained/YOLOv9_DOTA1_100EPOCHS.pt"`.
- **confidence_threshold**: The confidence threshold for object detection. Objects with a confidence level higher than this threshold are considered. Default is `0.1`.
- **output_folder**: The directory where results will be saved. Default is `"results"`.
- **known_phrases**: A list of phrases against which the descriptions of detected objects will be compared. This helps in identifying specific activities or features in images.
### Example Code
Here is how you can use the `ImageProcessor` in your scripts:
```python
from PIL import Image
from Remote_Sensing_Analysis.ImageProcessor import ImageProcessor
def test_image_processing():
processor = ImageProcessor(
model_weights_path="pretrained/YOLOv9_DOTA1_100EPOCHS.pt",
confidence_threshold=0.1,
output_folder="results",
known_phrases=[
"Rocket positioned on the launch pad for final countdown",
"Final checks on the launch systems",
"Lots of Activity in the Image",
"Rocket being fueled"
]
)
path = "path_to_your_test_image.jpg"
im1 = Image.open(path)
# Using .inference method
report, percentage = processor.inference(im1)
# Or using .generate method directly with an image object
report, percentage = processor.generate(im1)
if __name__ == "__main__":
test_image_processing()
```
The `report` object holds a comprehensive report on the image analysis. The `percentage` object indicates the likelihood of rocket preparation activities occurring. For additional information and data, please refer to the `output_folder` directory.
Replace path_to_your_test_image.jpg with the path to the image file you wish to process.
Raw data
{
"_id": null,
"home_page": "http://github.com/aiden200/Remote_Sensing_Analysis",
"name": "Remote-Sensing-Analysis",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.9",
"maintainer_email": null,
"keywords": "Machine-Learning Remote-Sensing Object-detection",
"author": "Aiden Chang",
"author_email": "aidenchang@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/43/19/c6f192c3bdbc0cb50330283f6d8aac15ece9f5a935dc04932c73fb16d7c8/remote_sensing_analysis-0.1.2.tar.gz",
"platform": null,
"description": "# Remote Sensing Analysis\n\nThis package provides tools for processing and analyzing satellite imagery, utilizing advanced machine learning techniques for object detection, image enhancement, and text analytics from images.\n\n## Installation\n\n1. **Download Package**:\n\nTo install the Remote Sensing Analysis package, simply run the following command:\n\n```bash\npip install Remote-Sensing-Analysis\n```\n\n2. **Download Model Weights**:\n The package requires specific model weights to function correctly. Download the model weights from the following Google Drive link:\n [Download Model Weights](https://drive.google.com/file/d/1KL3H-Fe1SVoCEFaO4KM4J_FMRF4ocoCz/view?usp=sharing)\n\n After downloading, place the weights under the `pretrained` folder.\n\n## Usage\n\n### Parameters\n\nWhen initializing the `ImageProcessor`, you can specify the following parameters:\n\n- **model_weights_path**: Path to the model weights file, default is `\"pretrained/YOLOv9_DOTA1_100EPOCHS.pt\"`.\n- **confidence_threshold**: The confidence threshold for object detection. Objects with a confidence level higher than this threshold are considered. Default is `0.1`.\n- **output_folder**: The directory where results will be saved. Default is `\"results\"`.\n- **known_phrases**: A list of phrases against which the descriptions of detected objects will be compared. This helps in identifying specific activities or features in images.\n\n### Example Code\n\nHere is how you can use the `ImageProcessor` in your scripts:\n\n```python\nfrom PIL import Image\nfrom Remote_Sensing_Analysis.ImageProcessor import ImageProcessor\n\ndef test_image_processing():\n processor = ImageProcessor(\n model_weights_path=\"pretrained/YOLOv9_DOTA1_100EPOCHS.pt\",\n confidence_threshold=0.1,\n output_folder=\"results\",\n known_phrases=[\n \"Rocket positioned on the launch pad for final countdown\",\n \"Final checks on the launch systems\",\n \"Lots of Activity in the Image\",\n \"Rocket being fueled\"\n ]\n )\n path = \"path_to_your_test_image.jpg\"\n im1 = Image.open(path)\n # Using .inference method\n report, percentage = processor.inference(im1)\n # Or using .generate method directly with an image object\n report, percentage = processor.generate(im1)\n\nif __name__ == \"__main__\":\n test_image_processing()\n```\n\nThe `report` object holds a comprehensive report on the image analysis. The `percentage` object indicates the likelihood of rocket preparation activities occurring. For additional information and data, please refer to the `output_folder` directory.\n\nReplace path_to_your_test_image.jpg with the path to the image file you wish to process.\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Library that conducts analysis on Satellite Imagery",
"version": "0.1.2",
"project_urls": {
"Homepage": "http://github.com/aiden200/Remote_Sensing_Analysis"
},
"split_keywords": [
"machine-learning",
"remote-sensing",
"object-detection"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "ddb10ff0116222a776946162399ce4f4fc976ba804ba687a0ba8beaa146babec",
"md5": "2bf7d09d4e252f8803a8af8fa1848416",
"sha256": "b54ec4e412fbcfdbe5b3c66be494e203e2bddf89097a299f9ebe18566c6e3987"
},
"downloads": -1,
"filename": "Remote_Sensing_Analysis-0.1.2-py3-none-any.whl",
"has_sig": false,
"md5_digest": "2bf7d09d4e252f8803a8af8fa1848416",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.9",
"size": 9657,
"upload_time": "2024-04-21T06:25:59",
"upload_time_iso_8601": "2024-04-21T06:25:59.250126Z",
"url": "https://files.pythonhosted.org/packages/dd/b1/0ff0116222a776946162399ce4f4fc976ba804ba687a0ba8beaa146babec/Remote_Sensing_Analysis-0.1.2-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "4319c6f192c3bdbc0cb50330283f6d8aac15ece9f5a935dc04932c73fb16d7c8",
"md5": "050918f4403a6ecb2cd943e06aa81256",
"sha256": "4848b85555e0f7bf317ae7feb145f31f367de55379cc7d09d0af099fdc8d582f"
},
"downloads": -1,
"filename": "remote_sensing_analysis-0.1.2.tar.gz",
"has_sig": false,
"md5_digest": "050918f4403a6ecb2cd943e06aa81256",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.9",
"size": 343779,
"upload_time": "2024-04-21T06:26:02",
"upload_time_iso_8601": "2024-04-21T06:26:02.650940Z",
"url": "https://files.pythonhosted.org/packages/43/19/c6f192c3bdbc0cb50330283f6d8aac15ece9f5a935dc04932c73fb16d7c8/remote_sensing_analysis-0.1.2.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-04-21 06:26:02",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "aiden200",
"github_project": "Remote_Sensing_Analysis",
"github_not_found": true,
"lcname": "remote-sensing-analysis"
}
JSON
