Land Cover Mapping

The IDRISI software includes a comprehensive suite of image processing tools, making it an excellent choice for land cover mapping applications with remotely-sensed data.

Image processing tools provide for image restoration, enhancement, classification and transformation. IDRISI has the most extensive set of classification tools on the market, including both supervised and unsupervised multi-spectral and hyperspectral classifiers, with special techniques for soft classification analysis. A host of machine learning tools are also provided, including artificial neural network classifiers and classification tree analysis.

A special routine for segmentation classification, which uses segments as landscape objects instead of pixels, provides an alternative to pixel-based classification approaches.

Application areas include:

  • Remotely-sensed image analysis
  • Inventory and baseline land resource mapping
  • Landuse and land change analysis
  • Agricultural monitoring
  • Natural resource monitoring

Learn more about IDRISI

IDRISI Selva Brochure (PDF)
Land Change Modeler for ArcGIS Brochure (PDF)
Classification Tree Analysis Focus Paper (PDF)
Segmentation-Based Classification Focus Paper (PDF)

Analytical Examples:

 
Segmentation IDRISI

Image segmentation in IDRISI creates an image of object-oriented segments that have spectral similarity across many input bands. These image segments can be used as input into the powerful set of existing classification routines in IDRISI.

Back to Top

 
Neural Network Classification IDRISI

A variety of machine learning classifiers are available within IDRISI. Neural network classifiers include a multi-layer perceptron, self-organizing map, radial basis function, and fuzzy ARTMAP. Each allows complete control over all parameters.

Back to Top

Clark University