GraphRiverClassifier

GraphRiverClassifier offers a tool to detect surface water and sediment in LandSat images, utilizing the high-quality, manually annotated RiverPIXELS Dataset by LANL scientists. This project aims to enhance environmental monitoring and research with its precise identification capabilities for water bodies and sediments globally.

This project provides a demo utilizing ChatGPT and Google Earth Engine, designed for easy use on Google Colab. It allows users to effortlessly obtain and analyze Landsat satellite images for water and sediment pixels of a specified location.

Ucayali River

Murray River

Quick Start

Prerequisites

We recommend using Google Colab for running this demo, as it provides easy access to Google Earth Engine. Before you start, make sure that:

1. Your Google Colab session is connected to your Google Drive.
2. Download data to the folder RiverPIXELS from Shared Google Drive
3. You have created a Google Earth Engine project under the same Google account.
4. Please guarantee your Google Colab support GPU.

Setting Up

Follow these steps to set up and run the demo:

1. Download this project and place it in your Google Drive under the path MyDrive/Global_Earth_Classifier.
2. Open the project in Google Colab.
3. Modify the config.yaml file in the project folder:
   • Change the GEE_PROJECT variable to your Google Earth Engine project name.
   • If you want to use the ChatGPT API for automatic selection of Longitude and Latitude values, replace the OPEN_AI_API_KEY with your own API key.

Running the Demo

After setting up, you can use this Google Colab example to run the demo.

Important Hyperparameters

• --lon_lat (Type: str, Default: 'auto'): Specifies the longitude and latitude values in the format "lon, lat". Set to "auto" to automatically use ChatGPT for determining these values based on the provided location.

• --location (Type: str, Default: 'Ucayali River'): Used when --lon_lat is set to "auto". It specifies the location's name to fetch its longitude and latitude values automatically.

• --lon_range (Type: float, Default: 0.25): Defines the range of longitude to create a rectangle area for analysis. It determines how wide the area around the specified longitude will be considered.

• --lat_range (Type: float, Default: 0.25): Similar to --lon_range, this parameter sets the range of latitude for the rectangular area of interest, specifying the height of the area around the given latitude.

• --gee_save_name (Type: str, Default: 'Landsat5_Image'): Determines the save name of the Google Earth Engine (GEE) image. This name is used for saving and referencing the processed Landsat image within your project.

License

MIT License

Copyright (c) 2024 Bohan Chen

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.