Abstract

Abstract. Band registration is one of the most critical steps in the production of multi/hyperspectral images and determines the accuracy of applications directly. Because of the characteristics of imaging devices in some multi/hyperspectral satellites, there may be a time difference between bands during push-broom imaging, which leads to displacements of moving clouds with respect to the ground. And a large number of feature points may gather around cloud contours due to the high contrast and rich texture, resulting in building a transformation more suitable for moving clouds and making ground objects ghosted and blurred. This brings a big challenge for registration methods based on feature extraction and matching. In this paper, we propose a novel coarse-to-fine band registration framework for multi/hyperspectral images containing moving clouds. In the coarse registration stage, a cloud mask is generated by grayscale stretching, morphology and other operations. Based on this mask, a coarse matching of cloud-free regions is performed to eliminate large misalignment between bands. In the refinement stage, low-rank analysis and RASL (Robust Alignment by Sparse and Low-rank decomposition) are used to optimize the rank of coarse results to achieve fine registration between bands. After experiments on a total of 102 images (83 cloudy images and 19 cloud-free images with all 32 bands) from Zhuhai-1 hyperspectral satellite, our method can achieve a registration accuracy of 0.6 pixels in cloudy images, 0.41 pixels in cloud-free images, which is enough for subsequent applications.

Highlights

  • With the continuous development of science and technology, more multi/hyperspectral sensors have been equipped on earth observation satellites, which effectively expands the ability to observe the earth through rich spectral information

  • Due to the characteristics of imaging devices in some multi/hyperspectral satellites, the imaging time of the same ground object may vary in different bands during push-broom imaging, which leads to misalignment between bands

  • Aiming at the problem of poor registration for ground objects in cloudy multi/hyperspectral images, we propose a novel coarse-to-fine band registration framework that eliminates the influence of moving clouds

Read more

Summary

INTRODUCTION

With the continuous development of science and technology, more multi/hyperspectral sensors have been equipped on earth observation satellites, which effectively expands the ability to observe the earth through rich spectral information. Vision-based methods usually estimate a transformation by extracting and matching tie points in different bands and uses it for alignment. Many tie points may gather around cloud contours because of the strong contrast and rich texture when extracting and matching features Transformations based on these tie points are inaccurate and will lead to a poor registration or even failure for ground objects while a good accuracy for clouds as shown in Fig.. Aiming at the problem of poor registration for ground objects in cloudy multi/hyperspectral images, we propose a novel coarse-to-fine band registration framework that eliminates the influence of moving clouds.

Cloud Detection in Remote Sensing Images
THE PROPOSED FRAMEWORK
Coarse Band Registration
Fine Band Registration
Data Introduction
Evaluation Method
Results and Comparisons
Direct Method
CONCLUSION
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