An Anisotropic Diffusion Filtering Implementation to Execute in Parallel Distributed Systems

Sobieranski, Antonio Carlos; Coser, L.; Dantas, Mário A. R.; Wangenheim, Aldo von; Comunello, Eros

doi:10.1109/csew.2008.64

Cited by 3 publications

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although originally proposed for gray-scale images [4,5], it has been long extended to color [6,7], multi [8] and hyperspectral [9,10] images. Furthermore, fast numerical approximations and parallelization [11,12] schemes have improved computation times significantly. However, dealing with high resolution images remains challenging.…”

Section: Introductionmentioning

confidence: 99%

Multi scale representation for remotely sensed images using fast anisotropic diffusion filtering

Vanhamel

Alrefaya

Sahli

2010

2010 IEEE International Geoscience and Remote Sensing Symposium

View full text Add to dashboard Cite

Object based image analysis has gained on the traditional per-pixel multi-spectral based approaches. The main pitfall of using anisotropic diffusion for creating a multi scale representation of a remotely sensed image remains the computational burden.Producing the coarser scales in a multi scale representation or, diffusing spatially large images involves significant time and resources. This paper proposes a fast approach for anisotropic diffusion that overcomes spatial size limitations by distributing the diffusion as individual sub-processes over several overlapping sub-images. The overlap areas are synchronized at specific diffusion time ensuring that the fast approximation does not deviate too much from its single process equivalent. The later shall be demonstrated for a medium sized image, which can be diffused using both schemes. In addition, we shall include experimental data for very large images that can not efficiently be processed using a sequential approach.

show abstract