Showing papers by "David A. Landgrebe published in 1991"

A survey of decision tree classifier methodology

Abstract: A survey is presented of current methods for decision tree classifier (DTC) designs and the various existing issues. After considering potential advantages of DTCs over single-state classifiers, the subjects of tree structure design, feature selection at each internal node, and decision and search strategies are discussed. The relation between decision trees and neutral networks (NN) is also discussed. >

Hierarchical classifier design in high-dimensional numerous class cases

Abstract: In applying pattern recognition methods in remote sensing problems, an inherent limitation is that there is almost always only a small number of training samples with which to design the classifier. A hybrid decision tree classifier design procedure that produces efficient and accurate classifiers for this situation is proposed. In doing so, several key questions are addressed, among them the question of the feature extraction techniques to be used and the mathematical relationship between sample size, dimensionality, and risk value. Empirical tests comparing the hybrid design classifier with a conventional single layered one are presented. They suggest that the hybrid design produces higher accuracy with fewer features. The need for fewer features is an important advantage, because it reflects favorably on both the size of the training set needed and the amount of computation time that will be needed in analysis. >

An analytical model of Earth-observational remote sensing systems

Abstract: The authors present a system model for the remote sensing process and some results that yield insight into the process. Key results include interrelations between the atmosphere, sensor noise, sensor view angle and scattered path radiance and their influence on classification accuracy of the ground cover type. Also included are results indicating the tradeoffs in ground cell size and surface spatial correlation and their effect on classification accuracy. >

An analytical model of Earth-observational remote sensing systems

Abstract: The authors present a system model for the remote sensing process and some results that yield insight into the process. Key results include interrelations between the atmosphere, sensor noise, sensor view angle and scattered path radiance and their influence on classification accuracy of the ground cover type. Also included are results indicating the tradeoffs in ground cell size and surface spatial correlation and their effect on classification accuracy. >

Fast likelihood classification

Abstract: A multistage classification that reduces the processing time substantially is proposed. This classification algorithm consists of several stages, and in each stage likelihood values of classes are calculated and compared. If a class has a likelihood value less than a threshold, the class is truncated at that stage as an unlikely class, thus reducing the number of classes for which likelihood values are to be calculated at the next stage. Thus a host of classes can be truncated by using a small portion of the total features at early stages, resulting in substantial reduction of computing time. Several truncation criteria are developed, and the relationship between thresholds and the error caused by the truncation is investigated. Experiments show that the proposed algorithm reduces the processing time by the factor of 3-7, depending on the number of classes and features, while maintaining essentially the same accuracies. >

Parameter trade-offs for imaging spectroscopy systems (remote sensing)

Abstract: To help better understand the problems of specifying data acquisition parameters and extracting desired information from the voluminous data, research has been focused on understanding the remote sensing process as a system and investigating the interrelated effects of various parameters. A system model is used to explore system parameter trade-offs for a model sensor based on the High-Resolution Imaging Spectrometer (HIRIS). Radiometric performance was studied, along with the effect on classification accuracy of several system parameters. The atmosphere and sensor have significant effects on the mean received signal and noise performance. The effect of random uncorrelated errors in the radiometric calibration of the detector array is discussed. Accurate pixel-to-pixel relative radiometric calibration and the use of the image motion compensation (IMC) option are seen to improve classification accuracy. The selection of feature sets based on combining spectral bands was studied under a variety of observational conditions. >

Feature Selection Based on Decision Boundaries

Abstract: > In this paper, a novel approach to feature selection for classification is proposed based directly on the decision boundaries. It is shown how discriminantly redundant features and discriminantly informative features are related to decision boundaries. A novel characteristic of the proposed method arises by noting that only a portion of the decision boundary is effective in discriminating between classes. Next a procedure to extract discriminantly informative features based on a decision boundary is proposed. The proposed feature selection algorithm has several desirable properties: (1) it predicts the minimum number of features necessary to achieve the same classification accuracy as in the original space (2) it finds the necessary feature vectors. Experiments show that the performance of the proposed algorithm compares favorable with that of previous algorithms. Key word: Decision boundary, decision boundary feature matrix, discriminantly redundant, discriminantly informative, intrinsic discriminant dimension. I. FEATURE SELECTION AND' SUBSPACE An observation consisting of N features in pattern recognition can be viewed as a point in the Ndimensional Euclidean space EN. Since an Ndimensional vector space can be spanned by a basis consisting of N independent vectors {a, ,a2,..,aN}, an observation can be expressed by a linear combination of these ai. Then feature selection is equivalent to finding an effective subspace, W , and the new features can be found by projecting an observation into the subspace. It is desired that dim(wj < N. Now consider briefly Bayes' decision rule, which will be used later in the proposed feature selection algorithm. Let X be an observation in the Ndimensional Euclidean space EN under hypothesis HI: X E w, i=l,2. Decisions will be made according to the following rule. h(X) < t X E 01 h(X) > t X E o2

Spatio-Temporal Contextual Classification Based on Markov Random Field Model

Abstract: A contextural classifier based on a Markov random field model, which can utilize both spatial and temporal contexts, is investigated. Spatial and temporal neighbors are defined, and the class assignment of each pixel is assumed to be dependent only on the measurement vectors of itself and those of its spatial and temporal neighbors according to the Markov random field property. Only interpixel class dependency context is used in the classification. The joint prior probability of the classes of each pixel and its spatial and temporal neighbors are modeled by a Gibbs random field. The classification is performed in a recursive manner. Experiments with multi-temporal Thematic Mapper data show promising results.

Spatio-temporal contextual classification based on Markov random field model. [for thematic mapping]

Abstract: A contextural classifier based on a Markov random field model, which can utilize both spatial and temporal contexts, is investigated. Spatial and temporal neighbors are defined, and the class assignment of each pixel is assumed to be dependent only on the measurement vectors of itself and those of its spatial and temporal neighbors according to the Markov random field property. Only interpixel class dependency context is used in the classification. The joint prior probability of the classes of each pixel and its spatial and temporal neighbors are modeled by a Gibbs random field. The classification is performed in a recursive manner. Experiments with multi-temporal Thematic Mapper data show promising results.