23 September 1999 Peak-tree: a new approach to image simplification
Author Affiliations +
We define a peak of an image I at the grey level g to be a connected region of pixels pij with the grey values g(pij) greater than or equal to g. The peaks form a tree structure, called the peak-tree (Pi) (I). It generalizes the notion of the grey-level histogram of I and partially captures the spatial distribution of grey values in the image. Thereby, it opens up new methods for simplifying an image as illustrated here. We give an efficient O(NV) algorithm for constructing the peak-tree, where N equals #(pixels in I) and V equals #(distinct grey levels in I).
© (1999) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Sukhamay Kundu, Sukhamay Kundu, } "Peak-tree: a new approach to image simplification", Proc. SPIE 3811, Vision Geometry VIII, (23 September 1999); doi: 10.1117/12.364103; https://doi.org/10.1117/12.364103


Learning deep similarity in fundus photography
Proceedings of SPIE (February 23 2017)
Pet fur color and texture classification
Proceedings of SPIE (January 28 2007)
Multistage hierarchy for fast image analysis
Proceedings of SPIE (December 16 1996)

Back to Top