Modeling the Perception of Visual Complexity in Texture Images and Painting Images
k6232_3.pdf 53.4 MB
k6232_1.pdf 348 KB
k6232_2.pdf 161 KB
support vector machine
This thesis devotes to building a relationship between human visual complexity perception and objective image features.
Visual complexity of images is a worth-well investigation. It has a wide range of applications from computer science (emotion semantic image retrieval, digital watermarking and image analysis, etc.) to arts (the design of product surface, wallpapers, painting appreciation and selection, etc.). Various methods for computing image complexity have been developed depending on different applications, such as information theory, fractal dimension, quad tree method, etc. However, these measures are not sufficiently related to human visual perception of complexity. Intuitively, visual complexity is influenced by various factors perceived by humans from the image, not directly related to simple objective measures like distribution of spatial frequencies.……
Copyright(c) by Author