Abstract
Local patterns and other patch based features have been an integral part of various computer vision applications as they encode local structural and statistical information. In this paper, we propose an image coding technique that utilizes Zeckendorf representation of pixel intensities and basic mathematical operators such as intersection, set difference, maximum, summation etc. for summarization of image regions. The algorithm produces a Z-coded image that tells about the homogeneity or the contrast in image regions with all codes in a range of 0 to 255.
Author supplied keywords
Cite
CITATION STYLE
Vigneron, V., Syed, T. Q., Duarte, L. T., Lang, E., Behlim, S. I., & Tomé, A. M. (2017). Z-images. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10255 LNCS, pp. 177–184). Springer Verlag. https://doi.org/10.1007/978-3-319-58838-4_20
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
