Application of projection pursuit learning to boundary detection and deblurring in images

Projection pursuit learning networks (PPLNs) have been used in many fields of research but have not been widely used in image processing. In this paper we demonstrate how this highly promising technique may be used to connect edges and produce continuous boundaries. We also propose the application of PPLN to deblurring a degraded image when little or no a priori information about the blur is available. The PPLN was successful at developing an inverse blur filter to enhance blurry images. Theory and background information on projection pursuit regression (PPR) and PPLN are also presented.     

Aging and Visual Masking: Sensory and Attentional Factors.

Visual tasks can yield quantitatively similar patterns of performance that reflect different underlying mechanisms in younger and older observers. In 3 experiments, we used the visual masking task of J. T. Enns and V. Di Lollo (1997) to examine 2 of these mechanisms: stimulus contrast and attention. Performance appeared to be equivalent for younger and older observers in some circumstances, although manipulation of contrast and attention suggested that older observers may use focal attention to enhance the perceptual clarity of the target. For older observers, impoverished visual representations may more readily be eliminated by manipulation of attention or by the presence of a mask, indicating that both attention and stimulus quality are important influences on performance. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Simulator driving performance predicts accident reports five years later.

L. Hoffman, J. M. McDowd, P. Atchley, and R. A. Dubinsky (2005) reported that visual and attentional impairment (measured by the Useful Field of View test and DriverScan) and performance in a low-fidelity driving simulator did not predict self-reported accidents in the previous 3 years. The present study applied these data to predict accidents occurring within a subsequent 5-year period (N = 114 older adults, 75% retention rate). Multivariate path models revealed that accidents in which the driver was at least partially at fault were significantly more likely in persons who had shown impaired simulator performance. These results suggest that even low-fidelity driving simulators may be useful in predicting real-world outcomes. (PsycINFO Database Record (c) 2010 APA, all rights reserved)