A novel LBP-Mean shift segmentation algorithm for UAV remote sensing images based on LBP textural features and improved Mean shift algorithm
- https://doi.org/10.2991/icmra-15.2015.79How to use a DOI?
- Computer vision; Hyperspectral remote-sensing images; image segmentation; mean shift; LBP texture feature
The proposed method deals with the joint use of the textural features and the image edge by the remote-sensing images.A definition of an adaptive segmentation algorithm is considered.Based on LBP features detecting,the textural information associated with each cell images is extracted as the set of connected cell with an similar flag value to which the cell belongs: The cell’s neighborhood is characterized by the color space and light intensity distribution of the corresponding flag zone.The textural information is the original cell’s value,be it a distinctive characteristic.Using Mean shift algorithm,the edge information are jointly used for the segmentation through a change of distance formula. Experiments on hyperspectral and panchromatic images are presented and show a significant increase in segmentation accuracies for village area:For instance,with the overall accuracy is increased from smash with a conventional algorithm to integrity with the proposed approach. Comparisons with other related software methods show that the method is competitive.
- © 2015, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Surong Xiang AU - Jingwen Xu AU - Junfang Zhao AU - Yong Li AU - Shaoyao Zhang PY - 2015/04 DA - 2015/04 TI - A novel LBP-Mean shift segmentation algorithm for UAV remote sensing images based on LBP textural features and improved Mean shift algorithm BT - Proceedings of the 3rd International Conference on Mechatronics, Robotics and Automation PB - Atlantis Press SP - 400 EP - 404 SN - 2352-538X UR - https://doi.org/10.2991/icmra-15.2015.79 DO - https://doi.org/10.2991/icmra-15.2015.79 ID - Xiang2015/04 ER -