Remote Sensing

"Remote Sensing" is a Python-based project inspired by ENVI (developed by L3Harris Geospatial) but designed to push the boundaries a bit further. Remote sensing involves gathering data about objects or phenomena without physical contact, typically applied to Earth and planetary observation. This repository is my personal exploration of modern remote sensing techniques using Python, fueled by a passion for integrating cutting-edge tech into everyday workflows.

This project is organized into topical folders, each focusing on a specific aspect of remote sensing. Lets dive into what youll find inside!

Folder Overview

01. Import

Kick off your remote sensing journey here. Learn how to import satellite imagery and visualize it using Python. This folder provides a simple starting point with examples like this:

02. Image Enhancement

Enhance your images to make them more suitable for analysis or display. Techniques like Gamma, Logarithm, and Sigmoid functions (via scikit-image) are applied to Landsat-8 imagery. Example using the Gamma method:

03. Mask

Mask satellite images using shapefiles (.shp) to focus on specific regions:

04. Mosaic

Create seamless orthomosaics from image collections, corrected for geometric distortion and color-balanced:

05. HSV

Explore the HSV (Hue, Saturation, Value) color space to analyze image properties numerically:

06. Spectral Profile

Generate spectral profiles to analyze pixel data across an image:

07. Principal Component Analysis (PCA)

Reduce data complexity with PCA, visualized and analyzed using scikit-learn on Landsat-8 imagery:

08. Spectral Indices

Calculate indices (e.g., vegetation, water) using spectral reflectance combinations:

09. Clustering

Apply k-means clustering (e.g., 4 clusters) to group similar pixels in Landsat-8 images:

10. Filters

Enhance or suppress image features using spatial filters:

11. Change Detection

Detect changes over time using Change Vector Analysis (CVA) methods like direction and magnitude:

12. Gray-Level Co-Occurrence Matrix (GLCM)

Analyze texture with GLCM, examining pixel relationships:

13. Support Vector Machine (SVM)

Classify data using SVM for supervised learning:

14. Time Series

Study temporal patterns in satellite data:

15. Convolutional Neural Network (CNN)

Use CNNs for lithology classification on Sentinel-2 data, validated with ROC curves and confusion matrices:

License

This project is licensed under the MIT License.

Contributing

Contributions are welcome! If youd like to collaborate or improve this project, please email me at:
mahdifarmahinifarahani@aol.com

Acknowledgments

• Thanks to the open-source community for the incredible libraries and tools that made this project possible.
  
• A special shoutout to Vcafe (Felestin St.) and Levana Cafe (ASP Towers) for the coffee that fueled countless coding sessions.

Enjoy exploring the repo!
Questions? Feel free to reach out:
mahdifarmahinifarahani@aol.com
ORCID: 0009-0008-3800-8688