Skip to main content

2004 | Buch

Advanced Image Processing Techniques for Remotely Sensed Hyperspectral Data

verfasst von: Professor Dr. Pramod K. Varshney, Dr. Manoj K. Arora

Verlag: Springer Berlin Heidelberg

insite
SUCHEN

Über dieses Buch

Over the last fifty years, a large number of spaceborne and airborne sensors have been employed to gather information regarding the earth's surface and environment. As sensor technology continues to advance, remote sensing data with improved temporal, spectral, and spatial resolution is becoming more readily available. This widespread availability of enormous amounts of data has necessitated the development of efficient data processing techniques for a wide variety of applications. In particular, great strides have been made in the development of digital image processing techniques for remote sensing data. The goal has been efficient handling of vast amounts of data, fusion of data from diverse sensors, classification for image interpretation, and development of user-friendly products that allow rich visualization. This book presents some new algorithms that have been developed for high­ dimensional datasets, such as multispectral and hyperspectral imagery. The contents of the book are based primarily on research carried out by some members and alumni of the Sensor Fusion Laboratory at Syracuse University.

Inhaltsverzeichnis

Frontmatter

Introduction

Introduction
Abstract
From time immemorial, man has had the urge to see the unseen, to peer beneath the earth, and to see distant bodies in the heavens. This primordial curiosity embedded deep in the psyche of humankind, led to the birth of satellites and space programs. Satellite images, due to their synoptic view, map like format, and repetitive coverage are a viable source of gathering extensive information. In recent years, the extraordinary developments in satellite remote sensing have transformed this science from an experimental application into a technology for studying many aspects of earth sciences. These sensing systems provide us with data critical to weather prediction, agricultural forecasting, resource exploration, land cover mapping and environmental monitoring, to name a few. In fact, no segment of society has remained untouched by this technology.
Pramod K. Varshney, Manoj K. Arora

General

Frontmatter
Chapter 1. Hyperspectral Sensors and Applications
Abstract
Since the beginning of remote sensing observation, scientists have created a “toolbox” with which to observe the varying dimensions of the Earth’s dynamic surface. Hyperspectral imaging represents one of the later additions to this toolbox, emerging from the fields of aerial photography, ground spectroscopy and multi-spectral imaging. This new tool provides capacity to characterise and quantify, in considerable detail, the Earth’s diverse environments.
Richard Lucas, Aled Rowlands, Olaf Niemann, Ray Merton
Chapter 2. Overview of Image Processing
Abstract
The current mode of image capture in remote sensing of the earth by aircraft or satellite based sensors is in digital form. The pixels correspond to localized spatial information while the quantization levels in each spectral band correspond to the quantized radiometric measurements. It is most logical to regard each image as a vector array, that is, the pixels are arranged on a rectangular grid but the value of the image at each pixel is a vector whose elements correspond to radiometric levels (also known as intensity values or digital numbers) of the different bands. The image formed for each band is a monochrome image and for this reason, we refer to the quantized values of such an image as gray levels. The images may be acquired in a few bands (i.e. multi-spectral image) or in hundreds of bands (i.e. hyperspectral image). Image processing operations for both multispectral and hyperspectral images can therefore be either scalar image oriented, that is, each band is processed separately as an independent image or vector image oriented, where the operations take into account the vector nature of each pixel. Image processing can take place at several different levels. At its most basic level, the processing enhances an image or highlights specific objects for the analyst to view. At higher levels, processing can take on the form of automatically detecting objects in the image and classifying them.
Raghuveer M. Rao, Manoj K. Arora

Theory

Frontmatter
Chapter 3. Mutual Information: A Similarity Measure for Intensity Based Image Registration
Abstract
Mutual information (MI) was independently proposed in 1995 by two groups of researchers (Maes and Collignon of Catholic University of Leuven (Collignon et al. 1995) and Wells and Viola of MIT (Viola and Wells 1995)) as a similarity measure for intensity based registration of images acquired from different types of sensors. Since its introduction, MI has been used widely for a variety of applications involving image registration. These include medical imaging (Holden et al. 2000; Maes et al. 1997; Studhilme et al. 1997; Wells et al. 1996), remote sensing (Chen et al. 2003ab), and computer vision (Chen and Varshney 2001a). The MI registration criterion states that an image pair is geometrically registered when the mutual information between the two images reaches its maximum. The strength of MI as a similarity measure lies in the fact that no assumptions are made regarding the nature of the relation between the intensity values of the image, as long as such a relationship exists. Thus, the MI criterion is very general and has been used in many image registration problems in a range of applications.
Hua-mei Chen
Chapter 4. Independent Component Analysis
Abstract
Independent component analysis (ICA) is a multivariate data analysis method that, given a linear mixture of statistical independent sources, recovers these components by producing an unmixing matrix. Stemming from a more general problem called blind source separation (BSS), ICA has become increasingly popular in recent years with several excellent books (e. g. Cichocki and Amari 2002; Haykin 2000; Hyvärinen et al. 2001) and a large number of papers being published. Its attractiveness is explained by its relative simplicity as well as from a large number of application areas (Haykin 2000). For example, ICA has been successfully employed in sound separation (Lee 1998), financial forecasting (Back and Weingend 1997), biomedical data processing (Lee 1998), image filtering (Cichocki and Amari 2002), and remote sensing (Tu et al. 2001). While in many of these applications, the problems do not exactly fit the required setting, ICA based algorithms have been shown to be robust enough to produce accurate solutions. In fact, it is this robustness that has fueled the theoretical advances in this area (Haykin 2000).
Stefan A. Robila
Chapter 5. Support Vector Machines
Abstract
Support Vector Machines (SVMs) are a relatively new generation of techniques for classification and regression problems. These are based on Statistical Learning Theory having its origins in Machine Learning, which is defined by Kohavi and Foster (1998) as,
...Machine Learning is the field of scientific study that concentrates on induction algorithms and on other algorithms that can be said to “learn.”
Mahesh Pal, Pakorn Watanachaturaporn
Chapter 6. Markov Random Field Models
Abstract
For decades, Markov random fields (MRF) have been used by statistical physicists to explain various phenomena occurring among neighboring particles because of their ability to describe local interactions between them. In Winkler (1995) and Bremaud (1999), an MRF model is used to explain why neighboring particles are more likely to rotate in the same direction (clockwise or counterclockwise) or why intensity values of adjacent pixels of an image are more likely to be the same than different values. This model is called the Ising model. There are a large number of problems that can be modeled using the Ising model and where an MRF model can be used. Basically, an MRF model is a spatial-domain extension of a temporal Markov chain where an event at the current time instant depends only on events of a few previous time instants. In MRF, the statistical dependence is defined over the neighborhood system, a collection of neighbors, rather than past events as in the Markov chain model. It is obvious that this type of spatial dependence is a common phenomenon in various signal types including images. In general, images are smooth and, therefore, the intensity values of neighboring pixels are highly dependent on each other. Because of its highly theoretical and complex nature, and intensive computational requirements, practical uses of MRF were extremely limited until recently. Due to the dramatic improvements in computer technologies, MRF modeling has become more feasible for numerous applications, for example, image analysis because many image properties, such as texture, seem to fit an MRF model, i.e., intensity values of neighboring pixels of images are known to be highly correlated with each other. The Markovian nature of these textural properties has long been recognized in the image processing community, and has widely been used in a variety of applications (e.g. image compression and image noise removal). However, these applications were limited to empirical studies and were not based on a statistical model such as the MRF model.
Teerasit Kasetkasem

Applications

Frontmatter
Chapter 7. MI Based Registration of Multi-Sensor and Multi-Temporal Images
Abstract
The necessity of accurate registration and geometric rectification arises due to the presence of a number of distortions (errors) in remote sensing images that occur as a result of variations in platform positions, rotation of earth and relief displacements etc. (see Chap. 2). The behavior of most of these distortions is systematic and, thus, can be easily removed at data acquisition centers. It is also generally expedient to procure systematic-corrected images as these are already geometrically rectified to a map projection system such as the Universal Transverse Mercator (UTM). However, systematic corrections are normally performed on the basis of platform ephemeris data obtained from the header information, which may be relatively inaccurate. Therefore, some distortions may still be present in the systematic-corrected data. This can be illustrated with the help of agency supplied Landsat TM images taken at two different times as shown in Fig. 7.1. The images were systematically corrected and geometrically rectified to UTM at the data acquisition center. It can, however, be seen that the two points (marked by “+”) having the same UTM coordinates in both images are located at different positions indicating the presence of non-systematic registration errors.
Hua-mei Chen, Pramod K. Varshney
Chapter 8. Feature Extraction from Hyperspectral Data Using ICA
Abstract
Most of the image processing techniques for multispectral or hyperspectral data have complexity that depends directly on the number of spectral bands in the acquired data (Swain and Davis 1978). Due to the large number of bands involved in the hyperspectral images, it is of interest to find methods that transform the image cube into one with reduced dimensionality while, at the same time, maintaining as much information content as possible. These techniques are known under the general name of feature extraction (Richards and Jia 1999). The term feature is used to refer to the spectral bands or other transforms derived from combinations of bands.
Stefan A. Robila, Pramod K. Varshney
Chapter 9. Hyperspectral Classification Using ICA Based Mixture Model
Abstract
Unsupervised classification of remote sensing images is typically based on a mixture model, where the distribution of the entire data is modeled as a weighted sum of the class-component densities (Duda et al. 2000). When the class-component densities are assumed to be multivariate Gaussian, the mixture model is known as the Gaussian mixture model. The K-means and the ISODATA algorithms that are widely used in remote sensing are based on the Gaussian mixture model. These Gaussian mixture model based classification algorithms often perform unsatisfactorily. This stems from the Gaussian distribution assumption for the class-component densities. Gaussianity is only an assumption, rather than a demonstrable property of natural spectral classes, and has been widely accepted due to its analytical tractability and mathematical simplicity. However, if a class happens to be multimodal, it is no longer appropriate to model the class with a multivariate Gaussian distribution. Therefore, the use of the Gaussian mixture model in such cases may lead to unsatisfactory performance.
Chintan A. Shah
Chapter 10. Support Vector Machines for Classification of Multi- and Hyperspectral Data
Abstract
As discussed in Chap. 5, support vector machines (SVMs) have originated from statistical learning theory for classification and regression problems. Unlike the popular neural network classifiers, SVMs do not minimize the empirical training error (Byun and Lee 2003). Instead, they aim to maximize the margin between two classes of interest by placing a linear separating hyperplane between them. While doing so, the upper bound on the generalization error is minimized. Thus, SVM based classifiers are expected to have more generalization capability than neural networks. Other advantages of SVMs are their ability to adapt their learning characteristic via a kernel function and to adequately classify data on a high-dimensional feature space with a limited number of training data sets thereby overcoming the Hughes Phenomenon. The theoretical background on SVMs has been presented in Chap. 5. This chapter builds on this theoretical knowledge to apply SVMs for classification of multi and hyperspectral remote sensing data.
Manoj K. Arora, Pakorn Watanachaturaporn
Chapter 11. An MRF Model Based Approach for Sub-pixel Mapping from Hyperspectral Data
Abstract
Image classification is a key task in many remote sensing applications. As discussed in Sect. 2.6 of Chap. 2, the objective of classification is to allocate each pixel of a remote sensing image into only one class (i.e. hard or per-pixel classification) or to associate the pixel with many classes (i.e. soft, sub-pixel or fuzzy classification). A number of hard classifiers are in vogue based on approaches such as statistical (Mather 1999), neural networks (Foody 2000a) and decision tree (Hansen et al. 2001).
Teerasit Kasetkasem, Manoj K. Arora, Pramod K. Varshney
Chapter 12. Image Change Detection and Fusion Using MRF Models
Abstract
The basic theory of Markov random fields was presented in Chap. 6. In this chapter, we employ this modeling paradigm for two image processing tasks applicable to remote sensing. The objectives of the two tasks are:
1.
To investigate the use of Markov random field (MRF) models for image change detection applications
 
2.
To develop an image fusion algorithm based on MRF models.
 
Teerasit Kasetkasem, Pramod K. Varshney
Backmatter
Metadaten
Titel
Advanced Image Processing Techniques for Remotely Sensed Hyperspectral Data
verfasst von
Professor Dr. Pramod K. Varshney
Dr. Manoj K. Arora
Copyright-Jahr
2004
Verlag
Springer Berlin Heidelberg
Electronic ISBN
978-3-662-05605-9
Print ISBN
978-3-642-06001-4
DOI
https://doi.org/10.1007/978-3-662-05605-9