Remote Sensing Image Registration with Line Segments and Their Intersections

Lyu, Chengjin; Jiang, Jie

doi:10.3390/rs9050439

Cited by 18 publications

(5 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The registration methods of remote sensing images are roughly divided into gray-level information [5], [6], frequency domain [7], feature [4], [8], [9], and deep learning algorithms [10]. The first three approaches are traditional registration methods, but with low accuracies and robustness for images which have obvious features changes of ground objects [3].…”

Section: Related Work a Registration Methodsmentioning

confidence: 99%

R&D-Net: Integration of Registration-Net and Detection-Net for Identifying Building Changes in High Spatial-Resolution Remote Sensing Images

Chen,

Zheng,

Leng

et al. 2024

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

View full text Add to dashboard Cite

In the detection of building changes in high spatialresolution remote sensing images, both the distribution position and surface characteristics of the same objects are probably different under different imaging phases, which potentially causes high false positives. In order to improve the detection accuracy of building changes, an integration network, named R&D net is proposed in this paper, which comprises a registration network (R-net) followed by a change detection network (D-net). In Rnet, two different phase images are accepted as inputs, corner points and their descriptors are generated to spatially align those images. After that, the spatially aligned images are fed into the Dnet, and building images are detected accordingly. In this paper, a multi-view automatic labeling method is proposed to obtain labeling corner points. A new dataset containing 5,104 image pairs is established. Experimental results demonstrate that the R-net can extract robust invariant features, and then improve registration accuracy under circumstances with obvious changes of surface feature, which is a base of D-net. Uniting pyramid pooling structure with a focal loss function in D-net, both leaky and wrong segmentations can be dramatically improved under complex scenes with many interferences. When compared with baseline methods on different high-resolution remote sensing scenes, the proposed method achieves better performance and more accurate detection results of building changes.

show abstract

Section: Related Work a Registration Methodsmentioning

confidence: 99%

R&D-Net: Integration of Registration-Net and Detection-Net for Identifying Building Changes in High Spatial-Resolution Remote Sensing Images

Chen,

Zheng,

Leng

et al. 2024

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

View full text Add to dashboard Cite

show abstract

“…Despite the presence of a large number of neighboring parallel feature lines and light transformation on the image pair (c), the three approaches can still achieve a good matching result with over 97% accuracy. (2) The N-LPI approach yields false matches, mainly in areas with sparse or no corresponding points coverage. This is because this approach is based on corresponding points in the generation of matching primitives, the construction of geometric descriptors and the filtering of matching candidates with the homography matrix constraint, so the number and distribution of corresponding points have a strong influence on the approach.…”

Section: Comparison To State-of-the-art Matching Methodsmentioning

confidence: 99%

“…Compared to feature point, the feature line in an image can more accurately express the contour features of the object. Therefore, line matching plays a key role in computer vision [1], image registration [2] and 3D reconstruction [3][4][5]. Line matching serves to establish correspondence between corresponding lines from different images using image correlation techniques.…”

Section: Introductionmentioning

confidence: 99%

A New Line Matching Approach for High-Resolution Line Array Remote Sensing Images

2022

View full text Add to dashboard Cite

In this paper, a new line matching approach for high-resolution line array remote sensing images is presented. This approach establishes the correspondence of straight lines on two images by combining multiple constraints. Firstly, three geometric constraints, epipolar, direction and the point-line geometric relationship, are used in turn to reduce the number of matching candidates. After this, two similarity constraints, the double line descriptor and point-line distance, are used to determine the optimal matches. Finally, the co-linearity constraint is used to check the one-to-many and many-to-one correspondences in the results. The proposed approach is tested on eight representative image patches selected from the ZY-3 line array satellite images, and the results are compared with those of two state-of-the-art approaches. Experiments demonstrate the superiority and potential of the proposed approach due to its higher accuracy and greater number of matches in most cases.

show abstract

“…As mentioned in LSD [47], the line segments are validated by the number of level-line aligned pixels in a line-support region. A binary descriptor, derived from the concept of Line Descriptor with Gradually Changing Bands (LDGCB) [48,49], is introduced to evaluate the gradient distribution in the line-support region, which is a rectangle composed of several bands extending on either side of the segment.…”

Section: Energy Formulationmentioning

confidence: 99%

TopoLAP: Topology Recovery for Building Reconstruction by Deducing the Relationships between Linear and Planar Primitives

et al. 2019

View full text Add to dashboard Cite

Limited by the noise, missing data and varying sampling density of the point clouds, planar primitives are prone to be lost during plane segmentation, leading to topology errors when reconstructing complex building models. In this paper, a pipeline to recover the broken topology of planar primitives (TopoLAP) is proposed to reconstruct level of details 3 (LoD3) models. Firstly, planar primitives are segmented from the incomplete point clouds and feature lines are detected both from point clouds and images. Secondly, the structural contours of each plane segment are reconstructed by subset selection from intersections of these feature lines. Subsequently, missing planes are recovered by plane deduction according to the relationships between linear and planar primitives. Finally, the manifold and watertight polyhedral building models are reconstructed based on the optimized PolyFit framework. Experimental results demonstrate that the proposed pipeline can handle partial incomplete point clouds and reconstruct the LoD3 models of complex buildings automatically. A comparative analysis indicates that the proposed method performs better to preserve sharp edges and achieves a higher fitness and correction rate than rooftop-based modeling and the original PolyFit algorithm.

show abstract

Remote Sensing Image Registration with Line Segments and Their Intersections

Cited by 18 publications

References 47 publications

R&D-Net: Integration of Registration-Net and Detection-Net for Identifying Building Changes in High Spatial-Resolution Remote Sensing Images

R&D-Net: Integration of Registration-Net and Detection-Net for Identifying Building Changes in High Spatial-Resolution Remote Sensing Images

A New Line Matching Approach for High-Resolution Line Array Remote Sensing Images

TopoLAP: Topology Recovery for Building Reconstruction by Deducing the Relationships between Linear and Planar Primitives

Contact Info

Product

Resources

About