Feature Extraction from High-Resolution Remotely Sensed Imagery using Evolutionary Computation

Scalable Evolutionary Computation for Efficient Information Extraction from Remote Sensed Imagery

“…The processing of remote sensing data can be divided into three stages: pre-processing, processing and post-processing. Pre-processing operations prepare the input data for the actual image processing stage by minimizing the distortions and/or errors in an image that could prevent successful classification, or by extracting the most critical information from an image (Momm & Easson, 2011a;Khorram et al, 2012). Algebraic spectral band combinations (referred to as spectral indices), such as division, addition, subtraction, or multiplication, are examples of the way in which images can be pre-processed in order to enhance information.…”

“…Furthermore, different spectral indices are used for improved change detection and spectral enhancement studies. For instance, infrared band over red band is used for vegetation distribution, green band over red band is used for mapping surface water bodies and wetland delineation, red band over infrared band is used for mapping turbid waters, and red band over green band is used for mineral mapping (Momm & Easson, 2011a;Khorram et a.l, 2012). After pre-processing, satellite images are ready for image classification process that converts the original spectral data, which are variable and may show complex relationships across several image bands, into a simple thematic map for end users (Khorram et al, 2012).…”

“…Remote sensing can aid studies on earth surface processes, homeland security, disaster response, agricultural and environmental resource management, weather forecasting and global change research (Momm & Easson, 2011a). However, as a result of significant advances in technology, many organizations face the challenge developing timely and costeffective techniques that can successfully identify certain types of features from remotely sensed imagery (Momm & Easson, 2011a).…”

“…However, as a result of significant advances in technology, many organizations face the challenge developing timely and costeffective techniques that can successfully identify certain types of features from remotely sensed imagery (Momm & Easson, 2011a). The main challenge lies in analysing a large number of images over a long period of time and converting image data into actionable intelligence by extracting specific features.…”

“…The main challenge lies in analysing a large number of images over a long period of time and converting image data into actionable intelligence by extracting specific features. In addition, the relationship between the different image regions is too complex to be solved by explicit programming (Momm & Easson, 2011a). Evolutionary computation is one of methods that can be used to improve feature extraction.…”