From image segmentation to anti-textons

We apply the 'patchwork engine' (PE; van Tender and Ejima, 2000 Neural Netw orks forthcoming) to encode spaces between textons in an attempt to find a suitable feature representation of anti-textons [Williams and Julesz, 1991, in Neural Networks for Perception volume 1. Human and Machine Perception E d. H Wechsler (San Diego, CA: Academic Press); 1992, Proceedings of the Nat ional Academy of Sciences of the USA 89 6531-6534]. With computed anti-text ons it is possible to show that tessellation and distribution of anti-texto ns can differ from that of textons depending on the ratio of texton size to anti-texton size. From this we hypothesise that variability of anti-texton s can enhance texture segregation, and test our hypothesis in two psychophy sical experiments. Texture segregation asymmetry is the topic of the first test. We found that targets on backgrounds with regular anti-textons segreg ate more strongly than on backgrounds with highly variable anti-textons. Th is neatly complements other explanations for texture segregation asymmetry (eg Rubenstein and Sagi, 1990 Journal of the Optical Society of America A 7 1632 - 1643). Second the relative significance of textons and anti-textons in human texture segregation is investigated for a limited set of texture patterns. Subjects consistently judged a combination of texton and antitext on gradients as more conspicuous than texton-only gradients, and judged tex ton-only gradients as being more conspicuous than anti-texton-only gradient s. In the absence of strong texton gradients the regularity versus irregula rity of anti-textons agrees with perceived texture segregation. Using PE ou tputs as anti-texton features thus enabled the conception of various useful tests on texture segregation. The PE is originally intended as a general i mage segmentation method based on symmetry axes. With this paper we therefo re hope to relate anti-textons with visual processing in a wider sense.