Scholarlink

This chapter aims to compare the performance of various edge detection algorithms on satellite imagery. The algorithms compared are Roberts, Sobel, Prewitt, Laplacian of Gaussian (LOG), and Canny. The algorithms are tested on two types of satellite imagery: RGB and multispectral imagery. The image dataset we used was taken from the Sentinel-2 satellite and consists of RGB images. The image processing was done using the Spark framework on Amazon Web Services (AWS). The results demonstrate that the Canny edge detection algorithm is the most successful for RGB imagery. The Canny algorithm also produces the most efficient image in size, followed by the Sobel and Prewitt algorithms. The LOG algorithm is the most resource-intensive algorithm regarding execution time, but it performs better than Roberts, Sobel, and Prewitt on RGB images.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Comparing Different Edge Detection Algorithms on Satellite Images

  • 章节

  • 发表日期：

    2026年

查看全文
  • Georgios Sakellariou,
  • Constandinos X. Mavromoustakis,
  • George Mastorakis,
  • Periklis Chatzimisios,
  • Athina Bourdena,
  • Evangelos Markakis

摘要

This chapter aims to compare the performance of various edge detection algorithms on satellite imagery. The algorithms compared are Roberts, Sobel, Prewitt, Laplacian of Gaussian (LOG), and Canny. The algorithms are tested on two types of satellite imagery: RGB and multispectral imagery. The image dataset we used was taken from the Sentinel-2 satellite and consists of RGB images. The image processing was done using the Spark framework on Amazon Web Services (AWS). The results demonstrate that the Canny edge detection algorithm is the most successful for RGB imagery. The Canny algorithm also produces the most efficient image in size, followed by the Sobel and Prewitt algorithms. The LOG algorithm is the most resource-intensive algorithm regarding execution time, but it performs better than Roberts, Sobel, and Prewitt on RGB images.