Change detection in SAR images based on superpixel segmentation and image regression
- PDF / 2,795,591 Bytes
- 11 Pages / 595.276 x 790.866 pts Page_size
- 80 Downloads / 208 Views
RESEARCH ARTICLE
Change detection in SAR images based on superpixel segmentation and image regression Rui Zhao 1,2
&
Guo-Hua Peng 1 & Wei-dong Yan 1 & Lu-Lu Pan 1 & Li-Ya Wang 1
Received: 3 December 2019 / Accepted: 25 September 2020 # Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Change detection (CD) is one of the most important application in remote sensing domain. The difference image (DI) generated by traditional change detection methods are sensitive to several factors, such as atmospheric condition changes, illumination variations, sensor calibration, and speckle noise, greatly affecting the detection performance. To avoid the aforementioned problem, in this paper, a novel approach based on superpixel segmentation and image regression is proposed to detect changes between bitemporal synthetic aperture radar (SAR) images. Specifically, the bitemporal images are firstly divided into a number of superpixel pairs under the guidance of segmentation result of a pre-DI. Next, each pixel in pre-event image is reconstructed utilizing its nearest neighbor to reduce the influence of noise. Then, a set of preselected unchanged sample are selected to learn the local regression model and to estimate the post-event image. After that, the final DI can be obtained by measuring the difference between estimated post-event image and the actual one. Finally, the fuzzy c-means (FCM) clustering algorithm is adopted to generate the binary change map. Adequate experiments on four SAR datasets have been tested, and the experimental results compared with the state-of-the-art methods have proved the superiority of the proposed method. Keywords SAR image . Change detection . Image regression . Superpixel segmentation . Spatial neighborhood information
Introduction Image change detection, which aims at identifying of changes occurred in land cover by analysing multitemporal remote sensing images acquired over the same geographical area at two different periods, is of primary interest in the remote sensing community due to the fact that it has found widespread applications in various domains, such as environmental studies (Chavez and Mackinnon 1994; Thomas et al. 2018), damage assessment (Martinis et al. 2011; Moustakides 1986), analysis of urban changes (Ban and Yousif 2012; Hu and Ban 2014; Yousif and Ban 2014), etc.
Communicated by: H. Babaie * Rui Zhao [email protected] 1
Department of Computation Mathematics, School of Science, Northwestern Polytechnical University, Xi’an, China
2
School of Mathematics and Computer Application, Shangluo University, Shangluo, China
In recent years, due to the fact that SAR sensors are capable of acquiring data in all weather and are not affected by different sunlight-illumination conditions, image change detection utilizing SAR images has become an active research topic (Lv et al. 2018). A variety of different techniques related to SAR images change detection have been developed in the literature. Generally, change detection process in SAR images can be divided i
Data Loading...
