The overall objective of the categorization of all pixels in a (SAR) image into semantically meaningful classes is one of the most conducted processing steps in image processing for geosciences. Thematic classification allocates pixels to classes based on functions of their spectral (or backscatter) properties. In this lesson parametric and non-parametric classification concepts are explained.
Speckle is a common phenomenon in SAR data, which has its origin in the radar image acquisition principles. In almost all applications in geosciences speckle is considered as noise. Therefore, many approaches for the reduction of speckle exist. This lessons gives insight into the origin of speckle and an overview of concepts and approaches to distinguish between speckle and relevant information content.
Image texture is an important feature in image interpretation in optical as well as radar remote sensing. This lesson gives insights into the concept of texture, ways to measure and interpret it. Furthermore, some application examples of texture in radar remote sensing are given for the demonstration of its usability as measure to quantify land cover information.
The module “Digital Image Processing Basics” contains a set of slides that might come in handy for teaching and learning the very basics of the subject. The set contains visual representations of the following topics:
- The Histogram. Basics and typical distributions
- Elements of (SAR) image interpretation.
- File Storage Orders. Band Sequential, Band Interleaved by Line, Band Interleaved by Pixel.
- RGB composites. Definitions and SAR methods.
- What means resolution? Different perspectives on digital images.