遥感影像融合的自适应变化检测

提出一种基于影像融合和自适应阈值选择的遥感影像变化检测方法。首先利用经过改进的融合技术对原 始数据的差值影像和比值影像进行处理, 构造融合影像, 在该融合影像的基础上进行自适应迭代运算得到初步变 化阈值范围, 然后通过分析阈值范围两侧影像像元的离散程度, 求解最终的阈值范围, 从而得到更优变化阈值, 提 取变化区域。实验结果表明, 本文方法的检测精度优于传统的变化检测方法, 同时具有一定的稳定性和智能性。

Adaptive change method of remote sensing image fusion

This paper proposes a change detection algorithm for remote sensing images based on image fusion and adaptive threshold selection. An improved image fusion technology has been employed in processing difference image and ratio image of original data in order to construct the fusion images. Based on these images, a coarse range of change threshold has been got from an adaptive iterative operation. Then, after analyzing the discrete levels of the image pixels distributed on both sides of the threshold range, the final threshold range has been achieved. Thus much more optimal change threshold helps to extract the final change region. the experimental results in the paper suggest that the detection accuracy of this method, which has certain stability and intelligence outperform the traditional change detection methods.