Abstract

High-resolution remote sensing image registration has recently been widely applied in many fields such as agriculture, forestry, and urban planning. A high-resolution remote sensing image can depict the texture features of ground details more clearly, but it also brings new challenges to registration such as texture similarity, image storage space, and image information leakage. In this paper, to solve the above problems, we propose a finite-state chaotic compressed sensing (CS) cloud remote sensing image registration method. First, chaotic CS with a finite state not only saves storage space required for an image but also improves the security of the image in the transmission process. Next, we process the image in the cloud platform, improve the speed of image processing, and facilitate real-time data processing. Finally, we propose an improved scale invariant feature transform (SIFT) image registration method based on local and global information (LG-SIFT) that reduces the impact of texture similarity on high-resolution remote sensing images. The experimental results show that the runtime of the original processing method is twice as long as that of the proposed scheme and that the accuracy of registration improves considerably.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call