FieldAnalysisService

About

The main idea of this project is to implement a service which can download satellite imagery, calculate NDVI and provide field NDVI image via REST API.

How to start

To start this application you need to start docker compose by command:

docker-compose up

Supported features

The service provides an API you can interact with. You should preserve following API calls order:

• add field
• trigger download satellite data task
• trigger ndvi calculation task
• get ndvi image

Implementation details

It is A FastAPI project, which consists of

• Backend application backend\:
  • Satellite data provider backend\app\satellite_data_providers - this directory includes files \satellite_data_client.py and \satellite_data_extractor.py which consists of functions that get data from Copernicus open access hub api by footprint and extracts file paths to Red and NIR images
  • Analytics backend\app\analytics - this directory includes file \ndvi_counter.py which consists of functions which calculates NDVI, creates NDVI image and invokes function to save NDVI image
  • File system backend\app\fs - this directory includes \file_system_storage.py and \image_saver.py which consist of functions that provide path to saved satellite data and NDVi images and save NDVI image
  • File unzipper backend\app\utils.py - this file includes function that unzips satellite data, delete zipped archive and saves unzipped file
• Database backend\app\database - we use PostgreSQL to store field id, field GeoJSON, path to NDVI image, status of calculating NDVI image
• Celery tasks backend\app\celery_tasks.py - celery tasks are responsible for getting unzipped satellite data and calculating NDVI and saving NDVI image

Examples of use

Add field to database and get field ID

curl -X 'POST' \
  'http://127.0.0.1:8000/field/' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -d '{
  "type": "string",
  "features": [
    {
      "type": "string",
      "geometry": {
        "coordinates": [
          "string",
          "string"
        ],
        "type": "string"
      },
      "properties": {},
      "id": "string",
      "bbox": [
        "string",
        "string",
        "string",
        "string"
      ]
    }
  ],
  "bbox": [
    "string",
    "string",
    "string",
    "string"
  ]
}'

You should pass GeoJSON in the following format

{
  "type": "string",
  "features": [
    {
      "type": "string",
      "geometry": {
        "coordinates": [
          "string",
          "string"
        ],
        "type": "string"
      },
      "properties": {},
      "id": "string",
      "bbox": [
        "string",
        "string",
        "string",
        "string"
      ]
    }
  ],
  "bbox": [
    "string",
    "string",
    "string",
    "string"
  ]
}

Get satellite image of the field

curl -X 'POST' \
  'http://127.0.0.1:8000/field/image' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -d '{
  "field_id": int
}'

You should pass JSON in the following format

{
    "field_id": int
}

Calculate NDVI and save NDVI image to server file system

curl -X 'POST' \
  'http://127.0.0.1:8000/field/ndvi' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -d '{
  "field_id": int
}'

You should pass JSON in the following format

{
    "field_id": int
}

Get NDVI image

curl -X 'GET' \
  'http://127.0.0.1:8000/field/ndvi/image?field_id=int' \
  -H 'accept: application/json'

Delete field from database and file system

curl -X 'DELETE' \
  'http://127.0.0.1:8000/field/?field_id=int' \
  -H 'accept: application/json'