A new measure of dependence is formulated, and some of its properties are described. The measure is applied to feature evaluation to show the relation of dependence between features and classes to discrimination between class-conditional probability distributions.

The results of an automated classification of the peripheral blood leukocytes into eight categories are presented. The classification was achieved by means of digital image processing. The categories were: small lymphocytes, medium lymphocytes, large lymphocytes, band neutrophils, segmented neutrophils, eosinophils, basophils, and monocytes. An eight-dimensional multivariate Gaussian classifier was used. The features were extracted from a 50 × 50 point digital image. These features were measures of such visual concepts as nuclear size, nuclear shape, nuclear and cytoplasmic texture, cytoplasm color, and cytoplasm colored texture. The data set consisted of 1041 blood cell images and were divided into a training set of 523 cells and an independent testing set of 518 cells. These cells were digitized directly from the blood smear, which was stained with Wright's stain. Twenty different blood smears were used and were collected over a three-year period from 20 people. The "true" classification of the data set was obtained from four experienced hematology technicians. Their performance was compared to the automated classifier both in terms of an absolute classification of cells and in terms of estimating the percentage composition of the population (or the blood cell differential count). The measure of performance used was the percentage error for each class. The mean percentage error for the eight classes in terms of an absolute classification was 8 and 29 percent for the human observers and the automated classiffier, respectively.

An economical technique for approximating a joint N-dimensional probability density function has been described by Sebestyen and Edie [20]. The algorithm searches for clusters of points and considers each cluster as one hyperellipsoidal cell in an N-dimensional histogram. Among the advantages of this scheme are: 1) the histogram cell descriptors-location, shape, and size-can be determined adaptively from sequentially introduced data samples of known classification and, 2) the number of cells required for a good fit can usually be held to a small number. No assumptions are required about the underlying statistical structure of the data. The algorithm requires three types of "control parameters" which critically affect its performance and are dependent upon the number of dimensions. The three factors control the birth, shape, and growth rate of the cells. Guides were presented in [20] for choosing the control parameter values. These guides functioned well for spaces of 3 dimensions or less, but did not yield usable values for spaces of greater dimensionality. This paper presents heuristics which were developed to automate the selection of the control parameters. The properties of these parameters were studied as a function of dimension. Two of the control parameters were found to be linearly related to dimension. This provides a method for determining their value by extrapolation, thereby avoiding a great deal of computation.

The only guaranteed technique for choosing the best subset of N properties from a set of M is to try all (MN) possible combinations. This is computationally impractical for sets of even moderate size, so heuristic techniques are required. This paper presents seven techniques for choosing good subsets of properties and compares their performance on a nine-class vectorcardiogram classification problem.

A pattern recognition network which computes a weighted sum of nonlinear functions of its inputs is considered. An algorithm for training this multithreshold network is presented. The multithreshold device is used for classifying patterns into more than two categories. Experimental results on the recognition of hand-printed characters are shown. In this case, the nonlinear functions consisted of an orthogonal set of property detectors. The network changed its structure by an evolutionary technique which consisted of periodic replacement of the least useful elements by new ones, randomly chosen.

Compiler programs are used to convert engineering test specifications to a digital code which can be interpreted by the automatic test equipment system. A programming language based upon decision-table techniques allows the test engineer to write his test statements in an extremely convenient fashion and permits him to program any test specification with only a minimum of knowledge about the specific test equipment system and particular programming techniques.