摘要
为解决遥感影像规模量过大,影响目标物变化检测精度的问题,设计了基于时空信息特征的大规模遥感影像目标物变化检测算法。构建变分去噪模型,去除遥感影像中的噪声,选取低通高斯核对去噪后的大规模遥感影像进行振幅谱卷积运算,构造遥感影像的显著图。从遥感影像显著图中,提取遥感影像的纹理特征、时间特征、光谱特征及统计特征,构建高维的叠加特征矩阵,作为遥感影像的时空信息特征。设置大规模遥感影像的时空信息特征作为长短期记忆网络的输入,长短期记忆网络通过输入门、输出门及遗忘门的运算,输出遥感影像目标物变化检测结果。实验结果表明,该算法有效检测遥感影像目标物变化,遥感影像规模量大时,仍具有较高的检测精度。
To solve the problem of the large scale of remote sensing images affecting the accuracy of target object change detection,a largescale remote sensing image target object change detection algorithm based on spatiotemporal information features was designed.A variational denoising model is constructed to remove noise from remote sensing images,and the amplitude spectrum convolution operation is carried out on the large scale remote sensing images after the denoising is selected by low-pass Gaussian check.From the salient images of remote sensing images,texture features,time features,spectral features,and statistical features of remote sensing images are extracted,and a high-dimensional superimposed feature matrix is constructed as the spatiotemporal information features of remote sensing images.The spatiotemporal characteristics of large-scale remote sensing images are set as the input of the long short-term memory network.The long short-term memory network outputs the change detection results of remote sensing image target objects through the operation of the input gate,the output gate,and the forget gate.The experimental results show that the algorithm is effective in detecting the changes of remote sensing image target objects,and still has high detection accuracy when the remote sensing image is large in scale.
作者
郭志强
吉顺文
丁国章
GUO Zhiqiang;JI Shunwen;DING Guozhang(Zhejiang Institute of Water Resources and Estuaries(Zhejiang Institute of Ocean Planning and Design),Hangzhou 310000,China)
出处
《测绘与空间地理信息》
2025年第10期119-122,131,共5页
Geomatics & Spatial Information Technology
关键词
时空信息特征
大规模
遥感影像
目标物变化
检测算法
遗忘门
spatiotemporal information characteristics
on a large scale
remote sensing images
the target object changes
detection algorithm
forget gate
