Improved Multiple Matching Method for Observing Glacier Motion with Repeat Image Feature Tracking

Area of Science:

• Glaciology
• Remote Sensing
• Geophysics

Background:

• Repeat Image Feature Tracking (RIFT) uses normalized cross correlation (NCC) for glacier surface motion.
• The Multiple-Image Multiple-Chip (MIMC) algorithm improved matching rates via redundant matching but was slow and failed in high shear areas.

Purpose of the Study:

• To enhance the MIMC algorithm for increased speed and matching success rate in glacier surface motion tracking.
• To address limitations of the original MIMC algorithm, particularly in challenging flow conditions.

Main Methods:

• Introduced Quadramatching: additional redundant measurements by swapping image order and matching direction.
• Implemented dynamic linear constraint (DLC): used a priori ice velocity to confine NCC search space, reducing computation.
• Developed pseudosmoothing: a novel post-processing algorithm for determining probable displacement.

Main Results:

• Quadramatching and DLC significantly increased both speed and the success rate of matches.
• Pseudosmoothing further refined displacement determination.
• The combined upgrades demonstrated complementary and multiplicative improvements in MIMC performance.

Conclusions:

• The updated MIMC algorithm offers substantial improvements in glacier surface motion measurement.
• These advancements enable more efficient and accurate analysis of glacier dynamics, even in complex flow regimes.