Matching Confidence Constrained Bundle Adjustment for Multi-View High-Resolution Satellite Images

Abstract

Bundle adjustment of multi-view satellite images is a powerful tool to align the orientations of all the images in a unified framework. However, the traditional bundle adjustment process faces a problem in detecting mismatches and evaluating low/medium/high-accuracy matches, which limits the final bundle adjustment accuracy, especially when the mismatches are several times more than the correct matches. To achieve more accurate bundle adjustment results, this paper formulates the prior knowledge of matching accuracy as matching confidences and proposes a matching confidence based bundle adjustment method. The core algorithm firstly selects several highest-confidence matches to initially correct orientations of all images, then detects and eliminates the mismatches under the initial orientation guesses and finally formulates both the matching confidences and the forward-backward projection errors as weights in an iterative bundle adjustment process for more accurate orientation results. We compared our proposed method with the famous RANSAC strategy as well as a state-of-the-art bundle adjustment method on the high-resolution multi-view satellite images. The experimental comparisons are evaluated by image checking points and ground control points, which shows that our proposed method is able to obtain more robust and more accurate mismatch detection results than the RANSAC strategy, even though the mismatches are four times more than the correct matches and it can also achieve more accurate orientation results than the state-of-the-art bundle adjustment method.

[Retracted] Design and Implementation of Mental Health Consultation System for Primary and Secondary School Students Based on Credibility Matching Model

“…Matching. Ling et al use kmeans algorithm to cluster users, and then complete recommendation according to all users or proxy users in the cluster where the target users are located [13]. Hsu et al solved the problem of data sparsity by principal component analysis.…”

Geo-positioning Method of Optical Satellite Unconventional Stereo Imagery Based on Optimal Intersection

“…Based on the principles of photogrammetry and corresponding mathematical theories, Wang uses the combination of RFM model, collinear equation model and affine transformation model to achieve the composite positioning of multi-source images for the same area images taken by different sensors or camerasError! Reference source not found.. Xing established and verified the correctness of the bundle adjustment model for optical and radar images [11]. On this basis, they analyzed the influence of the distribution, quantity, and precision of GCPs on the positioning results.…”

Oak Ridge National Laboratory Bundle Adjustment of Maxar WorldView and GeoEye Imagery

“…The absolute accuracy RMSE was 0.13 m in latitude and 0.12 m longitude and 0.28 m in elevation [Loghin et al, 2020]. Ling et al [2020] perform bundle adjustment of six multi-view WV03 images collected over 6 months.…”

“…Previous works maked use of the random sample consensus (RANSAC) algorithm [Anzid et al, 2017, Fischler andBolles, 1987] as a method of binary parsing of matches into inliers and outliers. Ling et al [2020] introduced a matching confidence metric that rejects outliers and translates inlier confidence into weights within the BA. This methodology allows for accurate orientation results from the BA process even when a high percentage of mismatched tie points remain from the matching step [Ling et al, 2020].…”