Theoretical issues in the design of visual inspection systems

Abstract

The objective of this review essay is to both chronicle and analyse literature in the area of visual inspection. Classical as well as contemporary papers are included to describe both the historical development and the state of the art of visual inspection theories and technologies. Human operators, despite well-documented problems, often perform visual inspection. While supervized machine systems obviate some of the problems associated with human inspectors, other problems still exist. In particular, accounting for a supervisor's perception of a machine's performance (as gauged, e.g. by trust) and consequent actions. The difficulties associated with these two alternatives have led to the emergence of a third alternative: collaborative human/machine or hybrid systems, which combine the advantages of both alternatives-in theory. However, in practice, how to best distribute the functions between a human and machine, in a dynamic environment in real time, is problematic. Moreover, a human's perceptions of its counterpart's performance remain an issue. These unresolved problems are subjects for future research. In the interim, the alternatives are critiqued to create a basis for establishing guidelines to select the alternative that is best suited for a given situation.

The Importance of Training for Interactive Trade Space Exploration: A Study of Novice and Expert Users

“…Not captured in the results was the fact that novices tended to misuse the visualizations they selected by not viewing the dimensions of greater importance. In this way, they exhibited the novice tendency of confusing visibility with relevance [14]. With the training video, the novices were better able to visualize the dimensions of higher importance, namely, the objective, through parallel coordinates, or by placing the objective on one of the axis as suggested by the video.…”

“…Viewing data in more than three dimensions also makes it harder to discover relationships, outliers, clusters, and gaps in the data [11]. Pattern detection is important especially for developing user-centered methodologies for interactive trade space exploration because humans are capable of learning from patterns and using this knowledge to improve their performance in a manner unattainable by current algorithms [14]. Additionally, Petre and Green [15] find that both perceptual and interpretive readership skill for graphical representations must be learned.…”

Proposal of a condition index for maintenance of runway beams

“…In addition to these aspects of variability, an inherent variability exists that can hardly be avoided, as explained by Jiang et al (2004). The literature reviewed is not conclusive about the actual effect of variability on the quality of visual inspection.…”

“…Therefore, variability affects the reliability of this system. The main sources of variability were identified by Megaw (1979) and later by Jiang et al (2004) as follows j subjective factors related to the inspector j physical and environmental factors related to the measurement location j task factors related to the complexity of the inspection itself j organisational factors related to training and protocols used to perform the inspection.…”

A Preliminary Study of Novice and Expert Users’ Decision-Making Procedures During Visual Trade Space Exploration

“…Viewing data in more than three dimensions also makes it harder to discover relationships, outliers, clusters, and gaps in the data [8]. Pattern detection is important especially for developing user-centered methodologies such as trade space exploration because humans are capable of learning from patterns and using this knowledge to improve their performance in a manner that no current algorithm can match [11]. Additionally Petre [12] finds that both perceptual and interpretive readership skill for graphical representations must be learned.…”