Performance analysis of edge detection algorithms with THEOS satellite images
Date
2017
ISBN
978-150905209-7
Journal Title
Journal ISSN
Volume Title
Resource Type
Conference paper
Publisher
Institute of Electrical and Electronics Engineers Inc.
Journal Title
Performance analysis of edge detection algorithms with THEOS satellite images
Recommended by
Abstract
The goal of this research is to find a suitable edge detection algorithm with 4 bands, B1, B2, B3, and B4 of different types of satellite image. In this paper, the dataset is derived from raw satellite images, namely THEOS, of the seashore in the Samut Prakan province, a fruit garden in the Chantaburi province and river line in Ayuthaya provinces. Edge detection performance algorithms are Signal-to-Noise Ratio (PSNR), Mean Square Error (MSE), and edge detection processing time, as well as qualitative human visual perception. Our result shows that Canny algorithm and Laplacian of Gaussian algorithm were the best edge detection algorithm for both qualitatively and quantitatively. Moreover, the different kinds of bands in satellite images illustrated the same result. All of the algorithms consumed similar times. � 2017 IEEE.
Description
Citation
2nd Joint International Conference on Digital Arts, Media and Technology 2017: Digital Economy for Sustainable Growth, ICDAMT 2017