Showing papers on "Contextual image classification published in 1985"

Feature Extraction And Pattern Classification In Space - Spatial Frequency Domain

Abstract: A novel feature extraction method, useful for 2-D shape description, is proposed. It is based on an optimal representation of a 1-D signal in space - spatial frequency domain, the Wigner distribution. For shape clasification, one of the many 1-D representations of the 2-D contours is employed. Boundary features, or shape descriptors, are obtained using sigular value decomposition of the Wigner distribution (WD). Properties of WD singular values are presented and shown to encode certain shape features such as the space-bandwidth product, the shape complexity in terms of number of components and their spacing, and the spatial frequency vs. the space dependence. The singular values of the boundary Wigner distri bution possess all the properties required of good shape descriptors. To illustrate the effectiveness of these descriptors in shape classification, a number of examples are presented. The proposed method is useful for robust classification of any 1-D patterns.

A low-cost classifier for multitemporal applications†

Abstract: Remote sensing image classification with the maximum-likelihood decision rule leads to a computational cost that depends quadratically on the number of bands in the data. Moreover, the data has to be modelled beforehand by sets of multivariate normal distributions if acceptable classification accuracies are to be obtained. A new algorithm is presented with a cost linearly proportional to the number of bands. Being based upon a combination of linear classifiers it is not dependent upon a priori parametric modelling of the data. Instead it partitions the measurement space in a piecewise-linear fashion leading to high accuracies at low cost, particularly for multitemporal data.

Correlation Synthetic Discriminant Functions for Object Recognition and Classification in High Clutter

Abstract: Correlation synthetic discriminant functions (SDFs) represent a practical and novel extension of matched spatial filter (MSF) correlators for distortion-invariant multi-object and multi-class pattern recognition. This paper reviews the off-line synthesis of such filters and the advantageous features of correlation shape control that they provide. We then concentrate on extensive tests performed with these filters to assess their performance in the identification of ship images, subjected to 3-D distortions. The pattern recognition problem addressed involves multi-object, multi-class recognition with aspect distortion-invariance in the presence of clutter. An adaptive threshold is shown to allow recognition of objects in the presence of spatially-varying modulation. The noise performance of these filters is also found to be most excellent. Correct classification rates approaching 98% can be obtained with these correlation SDFs.

Techniques For Multispectral Image Classification

Abstract: Over the last twenty years a variety of pattern recognition techniques for classifying terrain and cultural features using multi-spectral imagery have been developed. The purpose of this paper is to review and assess representative methods from major technique classes categorized according to the kinds of pattern models used (statistical, or heuristic), the types of information used (spectral, textural, spatial, and contextual), the manner in which they are applied to the image (i.e., to pixels or regions), and the manner in which they partition the image into classes (e.g., single step or hierarchical). An assessment of the accuracy, computational efficiency, and reliability is performed and trends in the technology are identified.

A Hybrid Digital/Integrated Optical Processor For On-Line Classification

Abstract: A new hybrid electrical/optical system for pattern classification is proposed. The optical circuit consists of c sets of analog electro-optic Bragg modulators, each capable of performing an n-dimensional inner product. This circuit is coupled with an optical threshold comparator to provide a binary representation of the class index. Pattern classification is effected through the use of a linear discriminant. The hybrid system is sufficiently rapid to make on-line classification feasible.

The development of an MSS satellite imagery classification expert system

Abstract: Unsupervised image classification of Landsat MSS imagery entails a significant part of the remote sensing, image analysis effort. Expert systems, a technology developed in the field of artificial intelligence, offers the potential to automate this process, thus greatly increasing the efficiency with which an analyst can perform unsupervised image classification and making the knowledge of the image analyst available to a community of nonexperts. Such a system, under development at the NASA/Ames Research Center, is described and planned enhancements are discussed.

Artificial Intelligence In Image Processing

Abstract: Image processing technology concentrates on the development of data extraction techniques applied toward the statistical classification of visual imagery. In classical image processing systems, an image is [1] preprocessed to remove noise, [2] segmented to produce close object boundaries, [3] analyzed to extract a representative feature vector, and [4] compared to ideal object feature vectors by a classifier to determine the nearest object classification and its associated confidence level. This type of processing attempts to formulate a two-dimensional interpretation of three-dimensional scenes using local statistical analysis, an entirely numerical process. Symbolic information dealing with contextual relationships, object attributes, and physical constraints is ignored in such an approach. This paper describes a number of artificial intelligence techniques which allow symbolic information to be exploited in conjunction with numerical data to improve object classification performance.

A Simple Tunable Interest Operator and Some Applications

Abstract: The extraction and classification of significant points along a contour is fundamental to many image processing tasks. In this paper, we present a simple process for extracting such points with several appealing properties: the operation is developed in terms of contours which are represented discretely; it is completely local and hence suitable for real time operation in vector or parallel processors; and it is tunable to extract significant points at different resolutions of orientation change along a contour. We also describe its use in linear feature extraction and processing restricted cases of environmental motion where the interest operator associates parameterized attributes with extracted image points. Matching features using these attributes allows for significant computational reductions over schemes based upon correlation matching without any loss of robustness, especially for such cases of restricted motion.