An Epipolar HS-NCC Flow Algorithm for DSM Generation Using GaoFen-3 Stereo SAR Images

Abstract

Radargrammetry is a widely used methodology to generate the large-scale Digital Surface Model (DSM). Stereo matching is the most challenging step in radargrammetry due to the significant geometric differences and the inherent speckle noise. The speckle noise results in significant grayscale differences of the same feature points, which makes the traditional Horn–Schunck (HS) flow or multi-window zero-mean normalized cross-correlation (ZNCC) methods degrade. Therefore, this paper proposes an algorithm named Epipolar HS-NCC Flow (EHNF) for dense stereo matching, which is an improved HS flow method with normalized cross-correction constraint based on epipolar stereo images. First, the epipolar geometry is applied to resample the image to realize the coarse stereo matching. Subsequently, the EHNF method forms a global energy function to achieve fine stereo matching. The EHNF method constructs a local normalized cross-correlation constraint term to compensate for the grayscale invariance constraint, especially for the SAR stereo images. Additionally, two assessment methods are proposed to calculate the optimal cross-correlation parameter and smoothness parameter according to the refined matched point pairs. Two GaoFen-3 (GF-3) image pairs from ascending and descending orbits and the open Light Detection and Ranging (LiDAR) data are utilized to fully evaluate the proposed method. The results demonstrate that the EHNF algorithm improves the DSM elevation accuracy by 9.6% and 27.0% compared with the HS flow and multi-window ZNCC methods, respectively.

Vibration Detection and Degraded Image Restoration of Space Camera Based on Correlation Imaging of Rolling-Shutter CMOS

“…While the row time of the rolling-shutter CMOS is on the order of microseconds, it is generally believed that the vibration influences of adjacent rows are consistent, allowing for the expansion of a single-row image into a small image suitable for vibration detection. Image registration methods, such as the gray projection algorithm and normalized cross-correlation algorithm [27,28], can be employed to determine the relative offset between images captured at time t i and time t i + nT row of the same scene by comparing rows i − k − 1 to i + k in frame p with those in frame p + 1.…”

“…While the row time of the rolling-shutter CMOS is on the order of microseconds, it is generally believed that the vibration influences of adjacent rows are consistent, allowing for the expansion of a single-row image into a small image suitable for vibration detection. Image registration methods, such as the gray projection algorithm and normalized cross-correlation algorithm [27,28], can be employed to determine the relative offset between images captured at time i t and time + By segmenting an image frame into multiple blocks by modifying i t , the relative offset sequence can be obtained by comparing corresponding blocks between two consecutive frames, and the parameters of the offset curve can be fitted. The sampling interval is determined by the distance between image blocks, whereas the sampling frequency can be adjusted through the modification of this distance.…”

Dense Matching Method for UAV SAR Images Without Epipolar Rectification

“…Some scholars' research is based on SGM to improve, such as using normalized cross correlation (NCC), absolute difference [5], phase-only correlation [6], sum of adaptive NCC [7] and other algorithms for cost calculation, or adding hierarchical pipeline [8] for DIM. There are also algorithms such as improved scale-invariant feature transformation (SIFT) and multiscale coherent point drift [9] to estimate deformation field parameters, or geometric constraints and region matching [10], or improved Horn-Schunck flow algorithms [11] to complete pixel-by-pixel matching between SAR images.…”