Image thresholding for landslide detection by genetic programming
This paper describes an approach to image threshold that combines various multiscale and morphological features, including texture, shape and edge filtering, by using genetic programming, to detect the presence of landslides and their active sectors in change detected multitemporal aerial images.
Bibliographic Reference: An oral report given at: The 1st International Workshop on the Analysis of Multitemporal Remote Sensing Images. Held in: Trento (IT), 13-14 September 2001
Record Number: 200113924 / Last updated on: 2001-10-22
Original language: en
Available languages: en