Multispectral Misregistration of Sentinel-2A Images: Analysis and Implications for Potential Applications

Abstract

This study aims at analyzing sub-pixel misregistration between multi-spectral images acquired by the Multi-Spectral Instrument (MSI) aboard Sentinel-2A remote sensing satellite, and exploring its potential for moving target and cloud detection. By virtue of its hardware design, MSI's detectors exhibit a parallax angle that leads to sub-pixel shifts that are corrected with special pre-processing routines. However, these routines do not correct shifts for moving and/or high altitude objects. In this letter, we apply a phase correlation approach to detect sub-pixel shifts between B2 (blue), B3 (green) and B4 (red) Sentinel-2A/MSI images. We show that shifts of more than 1.1 pixels can be observed for moving targets, such as airplanes and clouds, and can be used for cloud detection. We demonstrate that the proposed approach can detect clouds that are not identified in the built-in cloud mask provided within the Sentinel-2A Level-1C (L1C) product.

Improved Co-Registration of Sentinel-2 and Landsat-8 Imagery for Earth Surface Motion Measurements

“…Finally, image correlation of quasi-simultaneously acquired Sentinel-2 bands can also be used to detect fast moving (>4.7 m s −1 ) and high-altitude (e.g., clouds above 500 m) objects due to the time lag and parallax angles among visible Sentinel-2 bands [46]. Given that coregis's image correlation algorithm provides dense displacement fields and can, due to regularization, operate with small window sizes, it could be easily deployed for such applications.…”

A Novel Approach for Mapping Wheat Areas Using High Resolution Sentinel-2 Images

“…As reported in Sentinel-2 data quality report in 2018 [ 54 ], Sentinel-2A images before 15 June 2016 stem registration errors, due to three main contributors: (a) dynamic vibrations residuals mainly related to on-board oscillations; (b) static LOS calibration residuals; and (c) correlation noise and outliers. Thus, several previous studies [ 55 , 56 , 57 ] have shown a mis-registration between multi-temporal Sentinel-2A images from the same and different orbits for images acquired in 2016. To tackle the issue, we tried to visually investigate that matter by using the “chessboard” approach proposed by Shakun et al [ 56 ] as well as the qualitative visual registration assessment described in [ 58 ].…”

“…Thus, several previous studies [ 55 , 56 , 57 ] have shown a mis-registration between multi-temporal Sentinel-2A images from the same and different orbits for images acquired in 2016. To tackle the issue, we tried to visually investigate that matter by using the “chessboard” approach proposed by Shakun et al [ 56 ] as well as the qualitative visual registration assessment described in [ 58 ]. In addition, an open source software based on Yan et al [ 58 ] was used to quantify the occurred shifting in the x and y directions.…”

Aircraft and Ship Velocity Determination in Sentinel-2 Multispectral Images

“…S2 carries the Multispectral Sensor Imager [1,2,3] that records images in 13 multispectral bands (see Table 1) with different resolutions and time offsets. As we are interested in small object detection and tracking, we focused on analyzing the high resolution images, the 4 bands with 10 m and the 6 bands with 20 m pixel resolution.…”

“…The Sentinel-2 satellites under the Copernicus program [1,2,3] carry multispectral imaging (MSI) instruments that provide excellent and freely available imagery with pixel resolutions down to 10 m. The orbital period is 5 days between the Sentinel-2 (S2) satellites A + B, but as the swaths from different satellite orbits overlap at higher latitudes, the typical revisit period for each satellite is two or three days in Europe and almost daily in the Arctic. S2 MSI has the potential to greatly improve the marine and airspace situational awareness, especially for non-cooperative ships and aircrafts.…”