Eliminating the Effect of Image Border with Image Periodic Decomposition for Phase Correlation Based Remote Sensing Image Registration.

Yunyun Dong,Guojin He,Lanfa Liu,Tengfei Long,Weili Jiao

doi:10.3390/s19102329

Abstract

In the remote sensing community, accurate image registration is the prerequisite of the subsequent application of remote sensing images. Phase correlation based image registration has drawn extensive attention due to its high accuracy and high efficiency. However, when the Discrete Fourier Transform (DFT) of an image is computed, the image is implicitly assumed to be periodic. In practical application, it is impossible to meet the periodic condition that opposite borders of an image are alike, and image always shows strong discontinuities across the frame border. The discontinuities cause a severe artifact in the Fourier Transform, namely the known cross structure composed of high energy coefficients along the axes. Here, this phenomenon was referred to as effect of image border. Even worse, the effect of image border corrupted its registration accuracy and success rate. Currently, the main solution is blurring out the border of the image by weighting window function on the reference and sensed image. However, the approach also inevitably filters out non-border information of an image. The existing understanding is that the design of window function should filter as little information as possible, which can improve the registration success rate and accuracy of methods based on phase correlation. In this paper, another approach of eliminating the effect of image border is proposed, namely decomposing the image into two images: one being the periodic image and the other the smooth image. Replacing the original image by the periodic one does not suffer from the effect on the image border when applying Fourier Transform. The smooth image is analogous to an error image, which has little information except at the border. Extensive experiments were carried out and showed that the novel algorithm of eliminating the image border can improve the success rate and accuracy of phase correlation based image registration in some certain cases. Additionally, we obtained a new understanding of the role of window function in eliminating the effect of image border, which is helpful for researchers to select the optimal method of eliminating the effect of image border to improve the registration success rate and accuracy.

Highlights

Image registration is a core technology of image processing, widely applied in the remote sensing community, such as change detection [1,2], image fusion, long-time remote sensing data analysis, surface displacement of landslides [3,4], photogrammetry [5], etc
This paper extends a preliminary version of [33] by adding: (1) a detailed description of various methods of eliminating the effect of image border and the phenomenon of the effect of image border; (2) two groups of experiments about displacement estimation success rate and accuracy to further illustrate the advantages of the proposed method, and a further analysis of elimination of the effect of image border methods; (3) a thorough discussion of elimination of the effect of image border, and a new understanding of the role of window function
We proposed a novel algorithm of eliminating the image border, namely image periodic decomposition technology

Summary

Introduction

Image registration is a core technology of image processing, widely applied in the remote sensing community, such as change detection [1,2], image fusion, long-time remote sensing data analysis, surface displacement of landslides [3,4], photogrammetry [5], etc. For the feature based image registration methods, according to the utilized features, single pixel feature and multipixel feature, they can loosely be grouped into two categories. Its main procedures include feature detection, feature matching, transform model estimation and image warping [7,8,9,10]. Among these procedures, feature detection and feature matching are two core steps, and they are two steps to which researchers devote much effort [11]. Feature descriptors are developed from hand-engineered descriptors (HoG [18], DAISY [19], etc.) to deep learning based descriptors (DDesc [20], L2-Net [21] and HardNet [22])

Objectives

Findings

Discussion

Conclusion