A new productive framework for point‐based matching of oblique aircraft and UAV‐based images

The sweeping camera systems in the surveying and mapping industry are usually efficient in image acquisition, for the photography coverage of a single strip is relatively large. The triangulation angle of correspondence is overly tiny to adopt traditional block adjustment (BA). This study analysed the imaging principle of the Chinese APS7K comprehensive camera system and proposed an aerial triangulation method for the data this system acquired. The proposed method first determines the adjacent matrix from the POS data and trajectory information. The other part of the method is to overcome the weak relative geometry in a single strip by introducing the digital elevation model (DEM) data into the block adjustment scheme. The optimal solution of adjustment is obtained by iteratively solving the problem. We verified the optimisation's effectiveness by checking stitched orthophoto and check points from Google Earth. The results show that mosaic discrepancy is eliminated, the reprojection error is reduced to subpixel level, and positioning accuracy is better than 0.4 meter (ground sample distance is 0.2 meter) after adjustment with ground control points. Finally, the method’s shortcomings and prospects are summarized.

Section: Dem‐assisted Aerial Triangulation Methods For a Sweeping Cam...mentioning

confidence: 99%

Digital Elevation Model‐Assisted Aerial Triangulation Method On An Unmanned Aerial Vehicle Sweeping Camera System

Zhao

Zhou

Dong

et al. 2022

“…The mutual strategy mutually matches the two sets of descriptors so as to eliminate points with a low matching quality. Bas and Ok (2021) propose a similarity measure pipeline which contains NNDR, cross check and an iterative construction of the tentative matches based on AKAZE features. There are a few deep learning methods that can be used as a similarity measure.…”

Section: Related Workmentioning

confidence: 99%

An evaluation of conventional and deep learning‐based image‐matching methods on diverse datasets

Zeng

Zhang

et al. 2023

Image matching plays an important role in photogrammetry, computer vision and remote sensing. Modern deep learning‐based methods have been proposed for image matching; however, whether they will surpass and take the place of the conventional handcrafted methods in the remote sensing field still remains unclear. A comprehensive evaluation on stereo remote sensing images is also lacking. This paper comprehensively evaluates the performance of conventional and deep learning‐based image‐matching methods by dividing the matching process into feature point extraction, description and similarity measure on various datasets, including images captured from close‐range indoor and outdoor scenarios, unmanned aerial vehicles (UAVs) and satellite platforms. Different combinations of the three steps are evaluated. The experimental results reveal that, first, the performance of the different combinations varies between individual datasets, and it is difficult to determine the best combination. Second, by using more comprehensive indicators on all of the datasets, that is, the average rank and absolute rank, the combination of scale‐invariant feature transform (SIFT), ContextDesc and the nearest neighbour distance ratio (NNDR), and also the original SIFT, achieve the best results, and are recommended for use in remote sensing. Third, the deep learning‐based Sub‐SuperPoint extractor obtains a good performance, and is second only to SIFT. The learning based ContextDesc descriptor is as effective as the SIFT descriptor, and the learning based SuperGlue matcher is not as stable as NNDR, but leads to a few top‐performing combinations. Finally, the handcrafted methods are generally faster than the deep learning‐based methods, but the efficiency of the latter is acceptable. We conclude that although a full deep learning‐based method/combination has not yet beaten the conventional methods, there is still much room for improvement with the deep learning‐based methods because large‐scale aerial and satellite training datasets remain to be constructed, and specific methods for remote sensing images remain to be developed. The performance of the different combinations of feature extractor, descriptor and similarity measure varies between individual datasets. The combination of SIFT, ContextDesc and NNDR, and also the original SIFT, achieve the best results when using more comprehensive indicators on all the datasets. For extractor, the learning based Sub‐SuperPoint is second only to SIFT; for descriptor, learning‐based ContextDesc is as effective as the SIFT descriptor; and for matcher, learning‐based SuperGlue is not as stable as NNDR.

“…Good examples can be found in Parallel Tracking and Mapping (PTAM) (Klein & Murray, 2007), ORB‐SLAM (Mur‐Artal et al., 2015) and ORB‐SLAM2 (Mur‐Artal & Tardós, 2017), where the point feature is used for tie‐point matching. An example of the point feature‐based estimate may be seen in the work of Bas and Ok (2021), where A‐KAZE descriptors were employed for the purpose of feature extraction. Additionally, an iterative approach was adopted in order to maximise the number of initial matches.…”

Section: Introductionmentioning

confidence: 99%

PL‐Pose: robust camera localisation based on combined point and line features using control images

Xu,

Niu,

Cui

et al. 2024

Camera localisation is an essential task in the field of computer vision. The objective is to determine the precise position and orientation of one newly introduced camera station based on a collection of control images that are geographically referenced. Traditional feature‐based approaches have been found to face difficulties when confronted with the task of localising images that exhibit significant disparities in viewpoint. Modern deep learning approaches, on the contrary, aim to directly regress camera poses from input image content, being holistic to remedy the problem of viewpoint disparities. This paper posits that although deep networks possess the ability to learn robust and invariant visual features, the incorporation of geometry models can provide rigorous constraints in the process of pose estimation. Following the classic structure‐from‐motion (SfM) pipeline, we propose a PL‐Pose framework to perform camera localisation. First, to improve feature correlations for images with large viewpoint disparities, we perform the combination of point and line features based on a deep learning framework and geometric relation of wireframes. Then, a cost function is constructed using the combined point and line features in order to impose constraints on the bundle adjustment process. Finally, the camera pose parameters and 3D points are estimated through an iterative optimisation process. We verify the accuracy of the PL‐Pose approach through the utilisation of two datasets, that is, the publicly available S3DIS dataset and the self‐collected dataset CUMTB_Campus. The experimental results demonstrate that in both indoor and outdoor scenes, our PL‐Pose method can achieve localisation errors of less than 1 m for 82% of the test points. In contrast, the other four comparison methods yield a best result of merely 72%. Meanwhile, the PL‐Pose method can successfully obtain the camera pose parameters in all the scenes with small or large viewpoint disparities, indicating its good stability and adaptability.