<title>Quantitative characterization of image clutter: problem, progress, and promises</title>

Abstract

In this paper we present an overview of research studies aimed at deriving quantitative measures for image clutter. Image clutter is identified as a perceptual effect and therefore quantitative measures for clutter should be derived on the basis of models for perception. Previous studies in the field have produced limited success since they concentrated on the signal level interpretation in clutter characterization. In this study, we promote examination of the clutter problem at a higher level, where preattentive cues form the basis for their definition. Specifically, we derive clutter measures based upon preattentive texture features. Results of an extensive experimental study highlight the limitations of signal level clutter measures and the utility of texture based measures.

A novel method for quantifying target tracking difficulty of the infrared image sequence

“…A monotonic relationship is expected between a good image metric and the algorithm performance [2,11,33]. Therefore, to validate the proposed metrics, real infrared image sequences are used as samples for analyzing the relationships between our metrics and actual performance of tracking algorithms.…”

Image Sequence Measures for Automatic Target Tracking

“…Generally speaking, image quality evaluation results should be associated with the ATR algorithm performance and there should be a monotone relationship between a good image measure and the ATR algorithm performance [46,48]. Therefore, to validate the image sequence metrics proposed in this paper, real image sequences are used as samples for analyzing the relationship between our metrics and actual performance of tracking algorithms.…”

“…An image measure that considers both the clutter and the size of target is derived in [46]. It is named texture-based image clutter (TIC ), and defined as…”

Multispectral Image Processing: The Nature Factor

“…Animals rely for their survival on how well they blend into the terrain, yet we have only begun to be able to model the camouflage process in cluttered environments for image processing tasks. 10,30 Part of the problem relates to how we perceive objects based on texture and color variations and how these processes can be applied to imagery in spectral bands outside of the visible. 2,4,25 Finally, neural nets and fuzzy logic are beginning to be used as the image processing speed available in personal computers has attained the hundreds of MHz central processing unit clock speed.…”