Showing papers on "Dimensionality reduction published in 1974"

A Projection Pursuit Algorithm for Exploratory Data Analysis

Abstract: An algorithm for the analysis of multivariate data is presented and is discussed in terms of specific examples. The algorithm seeks to find one-and two-dimensional linear projections of multivariate data that are relatively highly revealing.

Image Restoration, Modelling, and Reduction of Dimensionality

Abstract: Recursive restoration of two-dimensional noisy images gives dimensionality problems leading to large storage and computation time requirements on a digital computer This paper shows a two-dimensional second-order Markov process representation can be used for fast recursive restoration of images with small storage requirements Advantages of this method over existing techniques are illustrated by means of examples

Nonlinear Intrinsic Dimensionality Computations

Abstract: In pattern recognition, the raw data and dimensionality of the measurement space is usually very large. Therefore, some form of dimensionality reduction has been commonly considered as a practical preprocessing method for feature selection. Based on a method that increases the variance while maintaining local structure, a technique is developed to determine intrinsic dimensionality. A cost function is introduced to guide the maintenance of the rank order and therefore local structure. Two criteria of using the cost function to increase the variance have been introduced. Several methods of defining the local regions are suggested. A program is implemented and tested to find the intrinsic dimensionality of a variety of experimental data.

A generalized approach to linear methods of feature extraction

Abstract: : An approach to feature extraction based on functions of the class correlation matrices is described. If linear functions of the correlation matrices are chosen, the present method extends the methods of feature extraction proposed by Fukunaga and Koontz. If certain types of non-linear functions are employed, the method reduces to the orthogonal subspace method of Watanabe and Pakvasa. Optimization of selected features through selection of appropriate functions is discussed briefly. Preliminary results of classification of radar signatures using the feature extraction methods described here are presented.

A Binary Feature Extraction Technique

Abstract: Numerous schemes are available for feature selection in a pattern recognition problem, but the feature extraction process is largely intuitive. A sequential feature extraction scheme is proposed for binary features. A decision function, which is linear and near optimal, is developed concurrently with each feature. Performance bounds are developed for several design strategies. Experimental results are given to illustrate the use of the scheme and the effectiveness of the performance bounds.

Multispectral combination and display of ERTS-1 data

Abstract: A significant problem in the use of ERTS-1 data is the extraction of information pertinent to each application and the presentation of that information in a form most suitable to users. When the information is to be displayed for visual study by an observer, the problem can be reduced to two steps: (1) Dimensionality reduction, an objective procedure which attempts to preserve most of the ERTS-1 information in a smaller number of components. (2) Display of the reduced number of components for optimum visibility by an observer. A specific dimensionality reduction technique has been applied to ERTS-1 data for several geographical areas in California and distinct types of earth resources. In the display of the reduced number of components, consideration has to be given to properties of the human visual system and the statistics of the data to be displayed. Previous work on digital image enhancement was applied to this problem to generate color composites which contain and display most of the information provided by the ERTS-1 sensors. Results of this approach were interesting, both in terms of the small mean-square caused by the dimensionality reduction, as well as for the examples of enhanced images that have been obtained.