Remote Sensing Digital Image Analysis

In a field like remote sensing image analysis, which changes so quickly, revising a long-standing textbook is a challenge. It is important to incorporate contemporary techniques, while not discarding procedures from the past which, while apparently superseded by newer methods, nevertheless retain value and are often simpler to use. Also, some processing operations from the past can become important again as data types and volumes change. Streaming methods for clustering are an example with the trend now to very large data sets. As with the previous edition, judgements have had to be made about what to leave out, what to retain and what to add. Those judgements have been made against the intended purpose of the book. From the beginning, it has been designed as a teaching text for the senior undergraduate and postgraduate student, and as a fundamental treatment for those engaged in the application of digital image analysis in remote sensing projects or in remote sensing image processing research. The presentation level is for the mathematical non-specialist. Because most operational users of remote sensing come from the earth sciences communities, the text is pitched at a level commensurate with their background. That is important because the recognised authorities in digital image analysis and machine learning tend to be from engineering, computer science and mathematics. Although familiarity with a certain level of mathematics and statistics cannot be avoided, the treatment here progresses through analyses carefully,with many hand-worked examples, so that any lack of depth in mathematical background should not take away from understanding the important aspects of image analysis and interpretation.