eduzhai > Applied Sciences > Engineering >

Development and Parallelization of an Improved 2D Moving Window Standard Deviation Python Routine for Image Segmentation Purposes

  • Peter
  • (0) Download
  • 20210302
  • Save

... pages left unread,continue reading

Document pages: 11 pages

Abstract: Two additional features are particularly useful in pixelwise satellitedata segmentation using neural networks: one results from local windowaveraging around each pixel (MWA) and another uses a standard deviationestimator (MWSD) instead of the average. While the former’s complexity hasalready been solved to a satisfying minimum, the latter did not. This articleproposes a new algorithm that can substitute a naive MWSD, by making thecomplexity of the computational process fallfrom O(N2n2) to O(N2n), where N is a square input array side, and n is the movingwindow’s side length. The Numba pythoncompiler was used to make python a competitive high-performance computing language in our optimizations. Ourresults show efficiency benchmars

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×