Non-Rigid Image Registration With Dynamic Gaussian Component Density and Space Curvature Preservation

Yang, Zhuoqian; Yang, Yang; Yang, Kun; Wei, Ziquan

doi:10.1109/tip.2018.2887204

Cited by 30 publications

(14 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, we randomly select 10 pairs of images from the entire data set to compare the registration accuracy. We follow the same evaluation in [60], [61]: the root mean square error (RMSE), maximum error (MAE) and median error (MEE) are…”

Section: E Results Of Image Registrationmentioning

confidence: 99%

Rotation-Invariant Siamese Network for Low-Altitude Remote-Sensing Image Registration

Liu

Gong

Chen

et al. 2020

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

Self Cite

View full text Add to dashboard Cite

Multiple-view change caused by small unmanned aerial vehicles (UAVs) monitoring the ground, result in image distortion, multi-view transformation and low overlap. Thus, such change has a strong effect on the accuracy of image registration. In this study, we utilize a Siamese network to deal with the complexity registration of low-altitude remote-sensing images. A robust neighbor-guided patch representation is designed to describe feature points based on neighborhood relation reconstruction and patch selection. The network is trained based on rotation-invariant layer to solve the inevitable rotation and nonrigid deformation caused by multi-view images in low-altitude remote-sensing images. With only 3 training images involving 4,500 putative matches, the experiment results demonstrated that the learned network can process the scenarios of yaw rotation, pitch rotation, mixture, and extreme(e.g. mixture, scaling and distortion occur simultaneously) of UAV better than other six state-of-the-art methods.

show abstract

Section: E Results Of Image Registrationmentioning

confidence: 99%

Rotation-Invariant Siamese Network for Low-Altitude Remote-Sensing Image Registration

Liu

Gong

Chen

et al. 2020

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

Self Cite

View full text Add to dashboard Cite

show abstract

“…• Image transformation. The thin plate spline is usually used in computer vision and image analysis [32], [42], [44], [45], particularly in non-rigid transformation model. In order to solve image geometric distortion and non-rigid distortions, we have utilized TPS to simulate a wider transformation.…”

Section: ) Main Processmentioning

confidence: 99%

Spatial Coordinates Correction Based on Multi-Sensor Low-Altitude Remote Sensing Image Registration for Monitoring Forest Dynamics

Lyu

et al. 2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

Tree species diversity plays a significant role in our ecosystem. In order to monitor forest dynamics, hyperspectral remote sensing equipped on a small unmanned aerial vehicle (UAV) is commonly applied, such as individual tree detection and classification. However, low resolution, positioning errors and the imaging perspective of small UAV affected by wind speed/direction, complex terrain, battery capacity, aircraft posture, flying height and other human factors result in relatively large positional errors (i.e., GPS errors) in such hyperspectral images, and the precise forest dynamics monitoring is limited, especially in spatial analysis. In order to reduce the positional errors of hyperspectral images captured from a small UAV and provide a precise forest dynamics monitoring, we present a novel spatial coordinates correction approach by registering low-altitude UAV visible light and hyperspectral images. The proposed method first employs visible light images and ground control points to stitch a geographic coordinate system as our groundtruth. Hyperspectral images (UHI) are then registered onto the stitched visible light image (UVI) via a novel image registration method. Finally, spatial coordinates of the registered hyperspectral images are updated by using the aforementioned groundtruth. Extensive experiments on image registration and spatial coordinates correction demonstrate the favorable performance of our method. Compared against four stateof-the-art registration methods, our method shows the best registration performance, and the positional errors of hyperspectral images are significantly reduced. Such accuracy is considered very high in this research. INDEX TERMS Forest dynamics, small unmanned aerial vehicle, multi-sensor image registration, spatial coordinates correction, hyperspectral images.

show abstract

“…Compared with the SIFT algorithm, the advantages of SURF algorithms are reflected in the following four aspects: detection with high repeatability, distinguishable eigenvectors, strong robustness and higher operational rate. On the other hand, the cores of the well-known ORB algorithm include Features from Accelerated Segment Test (FAST) (Rosten & Drummond, 2006;Yang, Yang, Yang, & Wei, 2019) and Binary Robust Independent Elementary Features (BRIEF) (Calonder, Lepetit, Strecha, & Fua, 2010;Yang, Yang, & Wei, 2018;Zhou, Wang, & Zhang, 2018). The one is used to extract feature points with high computational efficiency, and the other provides a shortcut to find the binary strings directly.…”

Section: Introductionmentioning

confidence: 99%

Improved feature point extraction and mismatch eliminating algorithm

Sun

Zhang

Wang

2020

Systems Science & Control Engineering

View full text Add to dashboard Cite

For the existing image matching algorithms, some inherent shortcomings, such as high mismatch rate and low computational efficiency, give rise to a bad influence on the performance of Visual Simultaneous Localization and Mapping (VSLAM). In this paper, a Grid-Based Motion Statistics for Fast and Random Sample Consensus (GMS-RANSAC) method combining with Multi-Probe Location Sensitive Hash (LSH)-based Adaptive and Generic Corner Detection Based on the Accelerated Segment Test and Oriented FAST and Rotated BRIEF (AGAST-ORB) algorithm is proposed to improve the real time and accuracy of image matching. To this end, the AGAST algorithm and the multi-probe LSH algorithm are firstly integrated into the traditional ORB algorithm to obtain the initial matching set. Specifically, the image feature points are extracted by the AGAST algorithm and then the main direction of feature points is given according to the intensity centroid method to guarantee the rotary invariant of feature points. Based on the extracted feature points, the multi-probe LSH algorithm, benefiting from its high time efficiency, is used to generate the initial matching pairs. In what follows, a GMS-RANSAC algorithm, which is improved by adding a directional similarity constraint model and the traditional RANSAC algorithm, is performed to improve the accuracy of eliminating result further. Finally, the performance test is implemented via a Mikolajczyk standard data set and it is verified that the proposed algorithm has higher matching precision and matching efficiency than traditional image matching algorithms.

show abstract

Non-Rigid Image Registration With Dynamic Gaussian Component Density and Space Curvature Preservation

Cited by 30 publications

References 60 publications

Rotation-Invariant Siamese Network for Low-Altitude Remote-Sensing Image Registration

Rotation-Invariant Siamese Network for Low-Altitude Remote-Sensing Image Registration

Spatial Coordinates Correction Based on Multi-Sensor Low-Altitude Remote Sensing Image Registration for Monitoring Forest Dynamics

Improved feature point extraction and mismatch eliminating algorithm

Contact Info

Product

Resources

About