IEEE Transactions on Pattern Analysis and Machine Intelligence

Mathematical structures of line drawings of polyhedrons are studied and practical as well as theoretical solutions are obtained for several fundamental problems aroused in scene analysis and in man-machine communication. First, a necessary and sufficient condition for a line drawing to correctly represent a polyhedron is obtained in terms of linear algebra. Next, combinatorial structures are inves... View full abstract»

This paper describes a new method to be used for matching three-dimensional objects with curved surfaces to two-dimensional perspective views. The method requires for each three-dimensional object a stored model consisting of a closed space curve representing some characteristic connected curved edges of the object. The input is a two-dimensional perspective projection of one of the stored models ... View full abstract»

A technique for locating desired structures utilizing user specified information about properties of these structures and their relationships with other more easily extracted objects is described. An edge-based and region-based technique is used for scene segmentation. Experimental results of the processing of aerial pictures are presented. View full abstract»

Computers and brains are modeled by finite and probabilistic automata, respectively. Probabilistic automata are known to be strictly more powerful than finite automata. The observation that the environment affects behavior of both computer and brain is made. Automata are then modeled in an environment. Theorem 1 shows that useful environmental models are those which are infinite sets. A probabilis... View full abstract»

The pictorial information in a stationary time series as depicted by crossings of levels and crossings of random levels and related quantities is studied. It is shown that such graphical features are directly connected with the covariance function and hence with the spectral density. Many of these features can be actually applied in estimation and in the study of extremes. In the Gaussian case, th... View full abstract»

An algorithm is described which extracts primitive regions (i.e., convex, spiral shaped, and biconcave lens) from complex shapes. The interior region bounded by the shape is decomposed by first slicing it into a set of convex subregions and then rotating and dissolving the various boundaries between subregions until a satisfactory decomposition is obtained. The same algorithm also is used to decom... View full abstract»

A set of images is modeled as a stochastic process and Karhunen-Loeve expansion is applied to extract the feature images. Although the size of the correlation matrix for such a stochastic process is very large, we show the way to calculate the eigenvectors when the rank of the correlation matrix is not large. We also propose an iterative algorithm to calculate the eigenvectors which save computati... View full abstract»

This correspondence describes a moving target tracking (MTT) algorithm that performs image registration and motion analysis between pairs of images from a passive sensor. Unlike previously reported moving target indicators that operate at the signal level, the registration and motion analysis in the MTT is totally performed at a symbolic level. The operation of the MTT is demonstrated by simulatio... View full abstract»

The binary n-gram and Viterbi algorithms have been suggested as alternative approaches to contextual postprocessing for text produced by a noisy channel such as an optical character recognizer. This correspondence describes the underlying theory of each approach in unified terminology, and presents new implementation algorithms for each approach. In particular, a storage efficient data structure i... View full abstract»

Probably the most common single discriminant algorithm in use today is the linear algorithm. Unfortunately, this algorithm has been shown to frequently behave poorly in high dimensions relative to other algorithms, even on suitable Gaussian data. This is because the algorithm uses sample estimates of the means and covariance matrix which are of poor quality in high dimensions. It seems reasonable ... View full abstract»

A method of achieving dimensionality reduction is presented. The reduced dimensionality is achieved by utilizing a least squared error technique under the assumption that the goodness criterion is the maximum separation of classes. The criterion is met by first maximizing the spread of the cluster centers, and then minimizing the within class scatter. The derivation of the desired transformation f... View full abstract»

In pattern matching a basic problem is to determine one or more vectors X that maximize an objective function which is a sum of functions of components of X. When this problem is solved by dynamic programming CPU time and storage requirements grow explosively as the amount of intervariable interaction in the objective function increases. This explosion may be reduced by departing from the traditio... View full abstract»

Two new algorithms that are more convenient for computation than existing ones for the slant transform are developed. These algorithms reveal the close relationship between the slant transform and the Walsh-Hadamard transform and demonstrate that the slant transform may be approached by a series of steps which gradually change the transform from a Hadamard or Walsh transform to a slant transform. View full abstract»