Beginners guide to the Semi-Automatic Classification Plugin (SCP) in QGIS

The Semi-Automatic Classification Plugin (SCP) is an innovative tool designed for users working with Geographic Information Systems (GIS). This plugin enhances the capabilities of QGIS by integrating a new Python library called Remotior Sensus, which allows for efficient pre-processing, band processing, and post-processing of remote sensing images.

Key Features of SCP

The new version offers a user-friendly interface, comprising several essential components:

• SCP Dock: Access all plugin functions, including downloading products, basic tools, and more.
• Training Input Dock: Create and manage Regions of Interest (ROIs).
• Band Sets: Load and manage raster bands for analysis.

Creating Band Sets

Band sets are crucial for managing raster data. Users can create multiple band sets and load raster bands without necessarily adding them to QGIS. This feature facilitates multi-temporal analysis by allowing users to set acquisition dates.

Training Input and Regions of Interest

Once the band sets are established, users can create training input files to store ROIs. The working toolbar allows users to select RGB color composites and create polygons representing different land cover types.

Downloading Remote Sensing Products

The SCP simplifies the process of downloading products from various sources, including Copernicus Sentinel-2 and Harmonized Landsat products. Users can define areas of interest and search for available images directly from the interface.

Basic Tools and Image Processing

The basic tools within SCP enable users to manage training inputs and band sets effectively. Features include:

• Exporting training inputs to various formats.
• Importing signature files and spectral libraries.
• Clipping raster bands based on user-defined coordinates.

Image Conversion and Masking

The image conversion tool allows users to convert images from compatible satellites, such as Sentinel-2 and Landsat, to reflectance. Masking bands is also possible, enabling users to focus on specific classifications.

Classification and Analysis Tools

The classification tool within SCP offers a variety of algorithms, including:

• Maximum Likelihood
• Random Forest
• Spectral Angle Mapping

Additionally, users can perform accuracy assessments and generate classification reports.

Advanced Features

Some advanced tools include:

• Principal Component Analysis: Calculate principal components for the input band set.
• Cross Classification: Compare two classification rasters to analyze differences.
• Reclassification: Edit and reclassify raster values based on user-defined criteria.

Creating Scripts and Automation

With the integration of Remotior Sensus, users can also create Python scripts for automating raster image processing tasks, enhancing workflow efficiency.

Accessing SCP Tools via QGIS Processing Toolbox

A new addition to SCP is the ability to access tools directly from the QGIS processing toolbox, making it easier for users to integrate SCP functions into their models and automated processes.

FAQ

What is the Semi-Automatic Classification Plugin?

The Semi-Automatic Classification Plugin (SCP) is a tool for QGIS that facilitates the processing and classification of remote sensing images.

How can I download remote sensing products using SCP?

You can download products by defining an area of interest and searching for available images directly from the SCP interface.

Can I automate processes using SCP?

Yes, SCP integrates with Python, allowing users to create scripts for automating raster image processing tasks.

Where can I find help and support for SCP?

You can join the Facebook group for sharing information and seeking help regarding the Semi-Automatic Classification Plugin.