A knowledge-based system for the automatic interpretation of aerial images of suburban scenes is described. Design considerations, the system architecture, and the procedure of obtaining a symbolic description of a picture is explained. Emphasis is put on knowledge representation and control strategy, as well as processing methods. Results are demonstrated via a sequence of pictures.

A Knowledge-Based System for the Analysis of Aerial Images

Expert system technology, especially techniques for rule-based knowledge engineering, has been successfully applied to many practical problems. Although there are inherent limitations in the complexity and power that can be achieved with traditional expert system approaches (Davis 1982), particularly when the number of rules is not constrained, there has been little evidence of its application to image interpretation. Typically, vision systems are highly system or application dependent and consequently it has been difficult to transfer them to different task domains. For example, there have been several systems developed for applications with aerial or satellite images (Bajcsy and Tavakoli 1976; Ballard et al. 1978; Faugeras and Price 1981; Nagao and Matsuyama 1980) but these systems have not been applied to other domains such as industrial scenes, medical applications, or road scenes. Only a few research efforts (Brooks 1981; Hanson and Riseman 1978c) have attempted to bridge task domains. We believe that a system employing a general methodology will be required if a large number of industrial applications are to be solved without requiring the construction of a customized system for each application. The bulk of this chapter will use results from outdoor scenes, specifically house scenes. However, we believe that our methodology can be applied in a straightforward manner to complex industrial applications by changing the knowledge base associated with the set of objects to be recognized. In an effort to motivate the reader to provide a sense of the potential, in the next section we have included the output of low-level algorithms applied to images of industrial parts.

A methodology for the development of general knowledge-based vision systems

This study covers the fundamentals of 2D and 3D computer vision and their most widespread industrial applications, such as automated inspection, robot guidance and workpiece acquisition. The level of explanation is that of an introductory text; however, besides the basic material, some specified advanced topics are included in each chapter, together with an extensive bibliography. The book covers the different aspects of computer vision, following a unified line of exposition with the result that the book is not biased towards any particular topic or approach. The text aims to provide readers with a solid background on the techniques used at the successive stages of the vision process, as well as a broad view of the possibilities of applying this new technology in industry.

Computer Vision: Theory and Industrial Applications

A rule-based system to make inferences about document images is introduced. Given a digitized document image, the sys tern controls the analysis of the document, and identifies all the different printed regions in the document image. Logical "blocks" of information on the document image are interpreted and classified by this system which then produces as output an editable description of the entire document. The system uses a goal-directed top down approach, and utilizes a three-level rule hierarchy to implement its control strategy.

/pdf/a-rule-based-system-for-document-understanding-5u3vjxcth5.pdf

A rule-based system for document understanding

The dissertation investigates an approach to the development of visual / spatial computer representations for architectural purposes through the development of the computerized handbook of architectural plans (chap), a knowledge-based computer system capable of recognizing the metric properties of architectural plans. This investigation can be summarized as an introduction of computer vision to the computerization of architectural representations: chap represents an attempt to automate recognition of the most essential among conventional architectural drawings, floor plans. The system accepts as input digitized images of architectural plans and recognizes their spatial primitives (locations) and their spatial articulation on a variety of abstraction levels. The final output of chap is a description of the plan in terms of the grouping formations detected in its spatial articulation. The overall structure of the description is based on an analysis of its conformity to the formal rules of its stylistic context (which in the initial version of chap is classical architecture). Chapter 1 suggests that the poor performance of computerized architectural drawing and design systems is among others evidence of the necessity to computerize visual / spatial architectural representations. A recognition system such as chap offers comprehensive means for the investigation of a methodology for the development and use of such representations. Chapter 2 describes a fundamental task of chap: recognition of the position and shape of locations, the atomic parts of the description of an architectural plan in chap. This operation represents the final and most significant part of the first stage in processing an image input in machine environment. Chapter 3 moves to the next significant problem, recognition of the spatial arrangement of locations in an architectural plan, that is, recognition of grouping relationships that determine the subdivision of a plan into parts. In the absence of systematic and exhaustive typologic studies of classical architecture that would allow us to define a repertory of the location group types possible in classical architectural plans, Chapter 3 follows a bottom-up approach based on grouping relationships derived from elementary architectural knowledge and formalized with assistance from Gestalt theory and its antecedents. The grouping process described in Chapter 3 corresponds both in purpose and in structure to the derivation of a description of an image in computer vision [Marr 1982]. Chapter 4 investigates the well-formedness of the description of a classical architectural plan in an analytical manner: each relevant level (or sublevel) of the classical canon according to Tzonis & Lefaivre [1986] is transformed into a single group of criteria of well-formedness which is investigated independently. The hierarchical structure of the classical canon determines the coordination of these criteria into a sequence of cognitive filters which progressively analyses the correspondence of the descriptions derived as in Chapter 3 to the constraints of the canon. The methodology and techniques presented in the dissertation are primarily considered with respect to chap, a specific recognition system. The resulting specification of chap gives a measure of the use of such a system within the context of a computerized collection of architectural precedents and also presents several extensions to other areas of architecture. Although these extensions are not considered as verifiable claims, Chapter 5 describes some of their implications, including on the role of architectural drawing in computerized design systems, on architectural typologies, and on the nature and structure of generative systems in architecture.

Development of a computerized handbook of architectural plans

We present an interpretation system which utilizes world knowledge in the form of simple object hypothesis rules, and more complex interpretation strategies attached to object and scene schemata, to reduce the ambiguities in image measurements. These rules involve sets of partially redundant features each of which defines an area of feature space which represents a "vote" for an object. Convergent evidence from multiple interpretation strategies is organized by top-down control mechanisms in the context of a partial interpretation. One such strategy extends a kernel interpretation derived through the selection of object exemplars, which represent the most reliable image specific hypotheses of a general object class, resulting in the extension of partial interpretations from islands of reliability.

T. E. Weymouth

Papers

Rule based strategies for image interpretation