Although many systems for optical reading of printed matter have been developed and are now in wide use, comparatively little success has been achieved in the automatic interpretation of optical images of three-dimensional scenes. This paper is addressed to the latter problem and is specifically concerned with automatic recognition of aircraft types from optical images. An experimental system is described in which certain features called moment invariants are extracted from binary television images and are then used for automatic classification. This experimental system has exhibited a significantly lower error rate than human observers in a limited laboratory test involving 132 images of six aircraft types. Preliminary indications are that this performance can be extended to a wider class of objects and that identification can be accomplished in one second or less with a small computer.

A pattern recognition system is described which is capable of identifying human faces from their full profile silhouettes. Each silhouette is preprocessed to remove noise, smooth edges, and extract the front edge. The processed silhouettes are then represented by a 12-dimensional feature vector, the components of which are obtained by a circular autocorrelation function. Using a weighted k-nearest neighbor decision rule it is shown that a recognition accuracy of 90 percent is attainable in a ten-class problem. An adaptive training procedure is also described which is used for setting up the authority files. This training procedure appears to identify those feature vectors representing a class which are either most important, from an information content point of view, or are observed most often. Finally, a comparison is made between the recognition accuracy obtained using circular autocorrelation features and moment invariant features. It is shown that the former outperforms, in this problem, the latter. The system is also compared to human observers with the result that the system performs no worse than the human observers.

This book contains a rather complete collection of algorithms, heuristics, and techniques for use in statistical or geometrical pattern recognition problems. Included are techniques for recognition, feature selection, and general data analysis. The rationale for presenting such an extensive bag of tricks is due to the author's feeling that mathematical pattern recognition is "...a collection of numerical algorithms for solving very particular problems posed in very particular ways." Because of this, the book should make an excellent reference work in mathematical pattern recognition.

In this paper, one-output n-input combinatorial nondegenerate networks of cells are analyzed. The networks consist of one-output two-input cells that are universal in that they realize all 16 functions of the two-input variables. Such networks have a number of interesting properties, among which is the fact that if network N realizes F(x), it also realizes the dual function Fd(x).

In recent years, considerable interest has been given to the study of threshold logic. This interest stems primarily from the possible economic savings associated with the use of threshold elements in realizing arbitrary switching functions. Such realizations, for the most part, depend on successfully finding threshold functions that closely approximate the given function. Closeness of approximation as defined here means that the approximating threshold function has the smallest set of error vectors with respect to the given function.

An algorithm for the conversion of BCD numbers to numbers in the -2 radix system is presented. This algorithm consists of simple arithmetic operations such as division by two and addition of the constants one and five. These operations may easily be performed in a -2 radix computer by shifting a number right for division by two and a table look-up procedure for the additions. The reverse conversion is also briefly discussed