Moving Object Detection for Remote Sensing Video with Satellite Jitter

Moving Object Detection in dynamic scenes with satellite jitter is an important issue in remote sensing video processing. Satellite jitter leads to whole image shaking which cause difficulty for moving object detection. This paper proposed a motion vector clustering method to detect moving feature p...

Full description

Saved in:
Bibliographic Details
Main Authors: Tsai, Chi-An, Lee, Pei-Jun, Bui, Trong-An, Xu, Guo-Cheng, Sheu, Meng-Lieh
Format: Conference Proceeding
Language:English
Subjects:
Clustering methods
Dynamics
Feature extraction
Jitter
moving object detection
moving vectors
Object detection
Remote sensing
satellite jitter
Satellites
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Moving Object Detection in dynamic scenes with satellite jitter is an important issue in remote sensing video processing. Satellite jitter leads to whole image shaking which cause difficulty for moving object detection. This paper proposed a motion vector clustering method to detect moving feature points. Real moving objects whose motion are different to jitter are able to be detected after removing jitter moving feature points. The proposed method increased 69.6% of succeed detected moving objects and reduce 93.6% of fake moving object.
ISSN:2158-4001
DOI:10.1109/ICCE56470.2023.10043483