Image Understanding Tasks Research Articles

KIMS—A knowledge-based computer vision system for production line inspection

Abstract the aim of this paper is to discuss a generic expert system prototyped for image analysis and interpretation tasks. The system christened KIMS is a knowledge-based image management system which oversees the entire process of image processing, segmentation, feature extraction, knowledge representation along with an expandable capability for image understanding task. We discuss the architecture of KIMS, its modeling environment, and demonstrate its usefulness with photo-to-image inspection as applied to a manufacturing line.

Low-level image analysis tasks on fine-grained tree-structured SIMD machines

This paper examines the applicability of fine-grained tree-structured SIMD machines, which are amenable to highly efficient VLSI implementation, to several low-level image understanding tasks. Algorithms are presented for histogramming, thresholding, image correlation, connected component labeling, and computing Euler number. A particular massively parallel machine called NON-VON is used for purposes of explication and performance evaluation. Only NON-VON tree-structured communication capabilities and its SIMD mode of execution are considered in this paper. Novel algorithmic techniques are described, such as vertical pipelining, subproblem partitioning, associative matching, and data duplication, that effectively exploit the massive parallelism available in fine-grained SIMD tree machines while avoiding communication bottlenecks. Simulation results are presented and compared with results obtained or forecast for other highly parallel machines. The relative advantages and limitations of the class of machines under consideration are outlined; except for some types of image correlation, the fine-grained SIMD tree is exceptionally fast.

On the range of applicability of an artificial intelligence machine

Considerable interest has recently been expressed in the construction of parallel machines capable of significant performance and cost/performance improvements in various artificial intelligence applications. In this paper, we consider the capabilities of a particular massively parallel machine called NON-VON, an initial prototype of which is currently operational at Columbia University, for the efficient execution of a rather wide range of AI tasks. The paper provides a brief overview of the general NON-VON architecture, and summarizes certain performance projections, derived through detailed analysis and simulation, in the areas of rule-based inferencing, computer vision, and knowledge-base management. In particular, summaries are presented of projections derived for the execution of OPS5 production systems, the performance of a number of low- and intermediate-level image understanding tasks, and the execution of certain “difficult” relational algebraic operations having relevance to the manipulation of knowledge bases. The results summarized in this paper, most of which are based on benchmarks proposed by other researchers, suggest that NON-VON could provide a performance improvement of as much as several orders of magnitude on such tasks by comparison with a conventional sequential machine of comparable hardware cost.

On the application of massively parallel SIMD tree machines to certain intermediate-level vision tasks

In this paper, we examine the implementation of two middle-level image understanding tasks on fine-grained tree-structured SIMD machines, which have highly efficient VLSI implementations. We first present one such massively parallel machine called NON-VON, and summarize the cost/performance trade-offs of such machines for vision taks. We follow with a more detailed description of the NON-VON architecture (a prototype of which has been operational since January 1985), and of the high-level parallel language in which our algorithms have been written and simulated. The heart of the paper consists of the description and analysis of algorithms for a representative Hough transform, and of an algorithm for the interpretation of moving light displays. Novel algorithmic techniques are motivated and described, and simulation timings are presented and discussed. We conclude that it is possible to exploit the available massive parallelism while avoiding many of the communication bottlenecks common at this level of image understanding, by carefully and inexpensively duplicating data and/or control information, and by delaying or avoiding the reporting of intermediate results.