scholarly journals Unsupervised Change Detection Using Spectrum-Trend and Shape Similarity Measure

2020 ◽  
Vol 12 (21) ◽  
pp. 3606
Author(s):  
Yi Tian ◽  
Ming Hao ◽  
Hua Zhang
Keyword(s):  
Change Detection ◽  
Satellite Images ◽  
Spectral Information ◽  
Feature Descriptor ◽  
Shape Similarity ◽  
Local Spectrum ◽  
Remote Sensing Images ◽  
Shape Context ◽  
Spot 5 ◽  
Very High

The emergence of very high resolution (VHR) images contributes to big challenges in change detection. It is hard for traditional pixel-level approaches to achieve satisfying performance due to radiometric difference. This work proposes a novel feature descriptor that is based on spectrum-trend and shape context for VHR remote sensing images. The proposed method is mainly composed of two aspects. The spectrum-trend graph is generated first, and then the shape context is applied in order to describe the shape of spectrum-trend. By constructing spectrum-trend graph, spatial and spectral information is integrated effectively. The approach is performed and assessed by QuickBird and SPOT-5 satellite images. The quantitative analysis of comparative experiments proves the effectiveness of the proposed technique in dealing with the radiometric difference and improving the accuracy of change detection. The results indicate that the overall accuracy and robustness are both boosted. Moreover, this work provides a novel viewpoint for discriminating changed and unchanged pixels by comparing the shape similarity of local spectrum-trend.

scholarly journals APPLICATION OF OFF-NADIR SATELLITE IMAGERY IN EARTHQUAKE DAMAGE ASSESSMENT USING OBJECT-BASED HOG FEATURE DESCRIPTOR

2019 ◽  
Vol XLII-3/W8 ◽  
pp. 167-171
Author(s):  
S. Jabari ◽  
M. Krafczek
Keyword(s):  
Change Detection ◽  
Damage Detection ◽  
Disaster Management ◽  
Satellite Images ◽  
Feature Descriptor ◽  
Object Based ◽  
Before And After ◽  
Map Generation ◽  
Spectral Comparison ◽  
Damage Map

<p><strong>Abstract.</strong> One of the most crutial applications of very-high-resolution (VHR) satellite images is disaster management. In disaster management, time is of great importance. Therefore, it is vital to acquire satellite images as quickly as possible and benefit from automatic change detection to speed up the process. Automatic damage map generation is performed by overlaying the co-registered before and after images of the area of interest and, compring them to highlight the affected infrastructures. For speeding up image capture, satellites tilt their imaging sensor and take images from oblique angles. However, this kind of image acquisition causes severe geometric distortion in the images, which hinders image co-registration in automatic change detection. In this study, a Patch-Wise Co-Registration (PWCR) solution is used. In this algorithm, the before and after images are co-registered in a segment-by-segment manner. From the literature, this algorithm is followed by a spectral comparison to detect changes. However, due to the complicated structure of debris in damage detection applications, spectral comparison methods cannot perform well. In this work, we developed an object-based method using Histogram of Oriented Gradient descriptor to detect damges and compared our results to different existing spectral and textural change detection methods. The algorithm is tested on images from the 2010-Heidi earthquake, captured by DigitalGlobe. The achieved highly accurate results demonstrate the potential of using off-nadir remote sensing images for automatic urban damage detection possibly in early response systems as it speeds up the damage map generation by providing flexibility to utilize images taken from different anlges.</p>

scholarly journals Unsupervised Change Detection Using Fast Fuzzy Clustering for Landslide Mapping from Very High-Resolution Images

2018 ◽  
Vol 10 (9) ◽  
pp. 1381 ◽  
Author(s):  
Tao Lei ◽  
Dinghua Xue ◽  
Zhiyong Lv ◽  
Shuying Li ◽  
Yanning Zhang ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Image Segmentation ◽  
High Resolution ◽  
Change Detection ◽  
Level Set ◽  
Execution Time ◽  
Remote Sensing Images ◽  
Fuzzy C Means ◽  
Landslide Mapping ◽  
Very High

Change detection approaches based on image segmentation are often used for landslide mapping (LM) from very high-resolution (VHR) remote sensing images. However, these approaches usually have two limitations. One is that they are sensitive to thresholds used for image segmentation and require too many parameters. The other one is that the computational complexity of these approaches depends on the image size, and thus they require a long execution time for very high-resolution (VHR) remote sensing images. In this paper, an unsupervised change detection using fast fuzzy c-means clustering (CDFFCM) for LM is proposed. The proposed CDFFCM has two contributions. The first is that we employ a Gaussian pyramid-based fast fuzzy c-means (FCM) clustering algorithm to obtain candidate landslide regions that have a better visual effect due to the utilization of image spatial information. The second is that we use the difference of image structure information instead of grayscale difference to obtain more accurate landslide regions. Three comparative approaches, edge-based level-set (ELSE), region-based level-set (RLSE), and change detection-based Markov random field (CDMRF), and the proposed CDFFCM are evaluated in three true landslide cases in the Lantau area of Hong Kong. The experiments show that the proposed CDFFCM is superior to three comparative approaches in terms of higher accuracy, fewer parameters, and shorter execution time.

Sign in / Sign up

Export Citation Format

Share Document