Abstract
Image registration is a core technology of many different image processing areas and is widely used in the remote sensing community. The accuracy of image registration largely determines the effect of subsequent applications. In recent years, phase correlation-based image registration has drawn much attention because of its high accuracy and efficiency as well as its robustness to gray difference and even slight changes in content. Many researchers have reported that the phase correlation method can acquire a sub-pixel accuracy of 1 / 10 or even 1 / 100 . However, its performance is acquired only in the case of translation, which limits the scope of the application of the method. However, there are few reports on the estimation of scales and angles based on the phase correlation method. To take advantage of the high accuracy property and other merits of phase correlation-based image registration and extend it to estimate the similarity transform, we proposed a novel algorithm, the Multilayer Polar Fourier Transform (MPFT), which uses a fast and accurate polar Fourier transform with different scaling factors to calculate the log-polar Fourier transform. The structure of the polar grids of MPFT is more similar to the one of the log-polar grid. In particular, for rotation estimation only, the polar grid of MPFT is the calculation grid. To validate its effectiveness and high accuracy in estimating angles and scales, both qualitative and quantitative experiments were carried out. The quantitative experiments included a numerical simulation as well as synthetic and real data experiments. The experimental results showed that the proposed method, MPFT, performs better than the existing phase correlation-based similarity transform estimation methods, the Pseudo-polar Fourier Transform (PPFT) and the Multilayer Fractional Fourier Transform method (MLFFT), and the classical feature-based registration method, Scale-Invariant Feature Transform (SIFT), and its variant, ms-SIFT.
Highlights
Image registration is a fundamental and challenging task that is used in many different research areas such as remote sensing, medical imaging, computer vision and video processing [1,2,3,4]
The adaptive binning scale-invariant feature transform (AB-Scale-Invariant Feature Transform (SIFT)) descriptor was proposed to enhance the robustness to Remote Sens. 2018, 10, 1719 local geometric distortions [21] and the histogram of orientated phase congruency (HOPC) descriptor was developed to address the nonlinear radiometric differences [22]
The basic principles of phase correlation-based image registration are the Fourier shift theorem and the log-polar transform, which states that the displacement of two images in spatial domain can be formulated as a phase difference in the frequency domain
Summary
Image registration is a fundamental and challenging task that is used in many different research areas such as remote sensing, medical imaging, computer vision and video processing [1,2,3,4]. Among the feature-based methods, the most classical algorithm is the scale-invariant feature transform (SIFT) [11]. In the remote sensing community, the direct application of SIFT faces many problems such as the low positional accuracy of feature points, the poor distribution of detected feature points in the spatial and scale spaces, the sensitivity of feature descriptors to non-linear radiation distortion, and so on. To address the poor distribution of detected features, novel score criteria for the feature points, a feature detection method based on phase congruency, and the uniform partitioning strategy were utilized to generate an adequate number of high-quality, uniformly-distributed point features in the spatial and scale spaces, such as UR-SIFT [23], UC-SIFT [24], and MMPC-Lap [25]. A support-line descriptor based on multiple adaptive binning gradient histograms was developed to filter out the outliers after the initial matching [26] to produce more high-precision correspondences
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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
