Advances in Pattern Recognition in Remote Sensing

Fiche éditeur

Remote sensing is an essential tool for Earth observations, where data collected by spaceborne and airborne platforms can be used for many related applications, such as environmental monitoring, resource management, disaster response, and homeland security, etc. With great advances of sensor technology, the resolutions (i.e., spatial, spectral, temporal, radiometric ones) have been greatly improved, resulting in massive amount of data. Efficient data interpretation is the key for success in practical applications. Pattern recognition is a very useful tool in the analysis of remote sensing data, and the vast amount of remote sensing data makes it uniquely suitable for statistical pattern recognition. Meanwhile, remote sensors with different modalities (e.g., multispectral, hyperspectral, optical, infrared, and radar sensors) are usually required to combine together to achieve the optimal outcomes in information mining and scene understanding, which also challenges the traditional pattern recognition techniques that are often tested with a single type of data modality.
The papers collected in this issue range from theoretical developments to various applications of established pattern recognition techniques using optical (hyperspectral or multispectral), thermal, and synthetic aperture radar (SAR) images.