Abstract
For earthquake disaster assessment using remote sensing (RS), multisource image registration is an important step. However, severe earthquakes will increase the deformation between the remote sensing images acquired before and after the earthquakes on different platforms. Traditional image registration methods can hardly meet the requirements of accuracy and efficiency of image registration of post-earthquake RS images used for disaster assessment. Therefore, an improved image registration method was proposed for the registration of multisource high-resolution remote sensing images. The proposed method used the combination of the Shi_Tomasi corner detection algorithm and scale-invariant feature transform (SIFT) to detect tie points from image patches obtained by an image partition strategy considering geographic information constraints. Then, the random sample consensus (RANSAC) and greedy algorithms were employed to remove outliers and redundant matched tie points. Additionally, a pre-earthquake RS image database was constructed using pre-earthquake high-resolution RS images and used as the references for image registration. The performance of the proposed method was evaluated using three image pairs covering regions affected by severe earthquakes. It was shown that the proposed method provided higher accuracy, less running time, and more tie points with a more even distribution than the classic SIFT method and the SIFT method using the same image partitioning strategy.
Highlights
Disasters caused by severe earthquakes, such as collapsed buildings, road damage, dammed lakes, and secondary geological disasters, pose a threat to people’s lives and property safety worldwide.In recent years, major earthquakes that occurred in China, such as the Wenchuan earthquake on 12 May2018, and the Yushu earthquake on 14 April 2010, attracted considerable attention from the government and society
Considering a local window in the image, Harris corner points are detected based on the determination of the average changes in image intensity that result from shifting the local window by a small amount in various directions
H and RMSEi were estimated iteratively based on the remaining tie points and the outliers were removed until either the ∆xi, ∆yi and RMSEi values were lower than the threshold TR or the number of iterations was higher than a set value denoted as Niter
Summary
Disasters caused by severe earthquakes, such as collapsed buildings, road damage, dammed lakes, and secondary geological disasters, pose a threat to people’s lives and property safety worldwide. It was apt to be affected by image noise and texture changes and was computationally intensive and time-consuming during feature point extraction [26] To solve these problems, several improved versions of the SIFT algorithm were proposed. Many algorithms work well for RS image registration, but when used for the registration of pre-earthquake and post-earthquake RS images, especially for mountain areas with complex terrain, two problems arise, including the lack of tie points and the uneven distribution of matched tie points To solve these two problems, a fast automatic registration method for multisource high-resolution. RS image registration in earthquake damage assessment was proposed in this study In this method, the Shi_Tomasi and SIFT algorithms were combined to obtain tie points using an image partition matching strategy considering geographic information.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have