Title :
Perceptual equivalence of environmental saliency in naturally complex scene
Author_Institution :
Fac. of Inf. Sci. & Technol., Osaka Inst. of Technol., Hirakata, Japan
Abstract :
A new framework is presented for saliency based analysis of naturally complex scenes. In this framework, the probability distributions of scale shift and chromatic diversity are extracted as a reference to stable maneuver through the bottom up process based on the image features underlying the scene. The scale-chromatic distributions are organized via fractal coding and articulated to generate the environmental saliency: vehicle specific description of ground-object structure. The scale and chromatic aspects of the environmental saliency are demonstrated to be perceptually equivalent through experimental studies.
Keywords :
feature extraction; fractals; natural scenes; statistical distributions; chromatic diversity; environmental saliency; fractal coding; image features; naturally complex scene; probability distribution; scale-chromatic distributions; stable maneuver; Complexity theory; Encoding; Feature extraction; Fractals; Humans; Pixel; Resource management; Environmental Saliency; Multi-Fractal Articulation; Satellite-Roadway-Vehicle Network; Scale-Chromatic Complexity;
Conference_Titel :
SICE Annual Conference 2010, Proceedings of
Conference_Location :
Taipei
Print_ISBN :
978-1-4244-7642-8