21 July 2017 Tolerant memoization on local image processing
Author Affiliations +
Proceedings Volume 10420, Ninth International Conference on Digital Image Processing (ICDIP 2017); 104204I (2017) https://doi.org/10.1117/12.2281923
Event: Ninth International Conference on Digital Image Processing (ICDIP 2017), 2017, Hong Kong, China
Abstract
In this paper, we present a new improvement technique in tolerant memoization algorithm for implementing local image processing operations. Window memoization is a method in the field of local image processing operations that uses the benefit of data redundancy in images to minimize the amount of computations performed that leads to achieve speedup. In the proposed method, the benefit of inter-pixel redundancy in images is used to improve the efficiency of tolerant window memoization where the neighboring pixels are correlated. We have developed this new optimized method on software and applied to edge detection to images of 512 × 512 pixels and compared with previous tolerant window memoization method, the results show the average speedup increasing is 34%.
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Mojtaba Farzmahdi, Mojtaba Farzmahdi, Rong Luo, Rong Luo, } "Tolerant memoization on local image processing", Proc. SPIE 10420, Ninth International Conference on Digital Image Processing (ICDIP 2017), 104204I (21 July 2017); doi: 10.1117/12.2281923; https://doi.org/10.1117/12.2281923
PROCEEDINGS
5 PAGES


SHARE
Back to Top