A robust feature-based registration method of multimodal image using phase congruency and coherent point drift

Xia, Renbo; Zhao, Jibin; Liu, Yunpeng

doi:10.1117/12.2031615

Cited by 12 publications

(5 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The network architecture consists of five convolution layers followed by three dense layers, eventually predicting the rotation, scale and translation of T . We are using CNN variants of conventional registration approaches [12,13]…”

Section: Coarse Fusion Network (Cfn)mentioning

confidence: 99%

Two-Phase Multimodal Image Fusion Using Convolutional Neural Networks

Kusram

Transue

Choi

2021

2021 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

The fusion of multiple imaging modalities presents an important contribution to machine vision, but remains an ongoing challenge due to the limitations in traditional calibration methods that perform a single, global alignment. For depth and thermal imaging devices, sensor and lens intrinsics (FOV, resolution, etc.) may vary considerably, making perpixel fusion accuracy difficult. In this paper, we present Ac-cuFusion, a two-phase non-linear registration method to fuse multimodal images at a per-pixel level to obtain an efficient and accurate image registration. The two phases: the Coarse Fusion Network (CFN) and Refining Fusion Network (RFN), are designed to learn a robust image-space fusion that provides a non-linear mapping for accurate alignment. By employing the refinement process, we obtain per-pixel displacements to minimize local alignment errors and observe an increase of 18% in average accuracy over global registration.

show abstract

Section: Coarse Fusion Network (Cfn)mentioning

confidence: 99%

Two-Phase Multimodal Image Fusion Using Convolutional Neural Networks

Kusram

Transue

Choi

2021

2021 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

show abstract

“…Feature-based approaches use image features such as landmark points, lines, and surfaces for registration, whereas intensity-based approaches use scalar values in the image pixels or voxels for performing registration [27]. The main steps involved in performing image registration are feature detection [28], feature matching [29], transformation [30] of source image into target image, and the optimization procedure [31]. In feature detection and matching steps, interested locations in the images (closed boundary regions, edges, contours, line intersections, corners, etc) are first detected and then a correspondence is made between the detected features.…”

Section: Integrating Information From Different Modalitiesmentioning

confidence: 99%

Challenges and Solutions in Multimodal Medical Image Subregion Detection and Registration

Alam

Rahman

2019

Journal of Medical Imaging and Radiation Sciences

View full text Add to dashboard Cite

Background: The automatic detection of common subregions and registration in multimodal functional and structural images is challenging. This article gives an overview of multimodal image registration and the developments and technical issues with automatic detection and registration of subregions of interest in multimodal images. Discussion: The available knowledge about subregion detection and registration in multimodal images are described in detail. Besides the provision of compact knowledge on subregion detection and registration, the challenges and proposed solutions are also discussed. Conclusion: This article provides research guidelines for the development of automatic detection and registration of subregions of interest in functional and structural images with high accuracy and efficiency. R ESUM E Contexte : La d etection et la superposition automatiques des sousr egions communes dans les images fonctionnelles et structurelles multimodales restent un probl eme difficile. Les connaissances actuelles sur la d etection et la superposition des sous-r egions int eress ees en imagerie multimodale sont d ecrites de façon d etaill ee. M ethodologie : Outre la fourniture d'information compacte sur la d etection et la superposition des sous-r egions int eress ees, cet examen traite egalement des d efis et propose des solutions. Conclusion : Le but de cette etude et de fournir des orientations de recherche pour le d eveloppement de la d etection et du calage automatis es des sous-r egions en imagerie fonctionnelle et structurelle, avec une pr ecision et une efficacit e elev ees.

show abstract

“…Compared with the typical features-based methods, structural features-based methods can extract more robust common features from different modalities and are less sensitive to the contrast differences. Due to these advantages, they have been successfully applied to multimodal image registration [10,21,[28][29][30][31][32][33]. As a valid structural feature extraction method, phase congruency was proposed by Morrone et al [34], which is the ratio of local energy to the overall path length taken by the local Fourier components in reaching the endpoint.…”

Section: Related Workmentioning

confidence: 99%

“…Wong and Orchard [29] constructed local phase-coherent representations of images and applied their method to multimodal medical image registration successfully. Xia et al [30] combined phase congruency representations of images with scale-invariant feature transform (SIFT) to achieve multimodal medical image registration. Recently, Liu et al [31] proposed mean local phase angle (MLPA) and frequency spread phase congruency (FSPC) by using local frequency information on Log-Gabor wavelet transformation space, which improved the robustness compared with traditional multimodal matching.…”

Section: Related Workmentioning

confidence: 99%

A Novel Affine and Contrast Invariant Descriptor for Infrared and Visible Image Registration

Liu

Zhang

et al. 2018

Remote Sensing

View full text Add to dashboard Cite

Infrared and visible image registration is a very challenging task due to the large geometric changes and the significant contrast differences caused by the inconsistent capture conditions. To address this problem, this paper proposes a novel affine and contrast invariant descriptor called maximally stable phase congruency (MSPC), which integrates the affine invariant region extraction with the structural features of images organically. First, to achieve the contrast invariance and ensure the significance of features, we detect feature points using moment ranking analysis and extract structural features via merging phase congruency images in multiple orientations. Then, coarse neighborhoods centered on the feature points are obtained based on Log-Gabor filter responses over scales and orientations. Subsequently, the affine invariant regions of feature points are determined by using maximally stable extremal regions. Finally, structural descriptors are constructed from those regions and the registration can be implemented according to the correspondence of the descriptors. The proposed method has been tested on various infrared and visible pairs acquired by different platforms. Experimental results demonstrate that our method outperforms several state-of-the-art methods in terms of robustness and precision with different image data and also show its effectiveness in the application of trajectory tracking.

show abstract

A robust feature-based registration method of multimodal image using phase congruency and coherent point drift

Cited by 12 publications

References 25 publications

Two-Phase Multimodal Image Fusion Using Convolutional Neural Networks

Two-Phase Multimodal Image Fusion Using Convolutional Neural Networks

Challenges and Solutions in Multimodal Medical Image Subregion Detection and Registration

A Novel Affine and Contrast Invariant Descriptor for Infrared and Visible Image Registration

Contact Info

Product

Resources

About