Remote Imagery Velocity Estimation With Misregistration Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face challenges in accurately estimating the velocity of moving objects from geospatial imagery due to misregistration between images from different sensors, which leads to errors in velocity estimation.
Innovation Solution
The proposed technology utilizes a satellite system with multiple sensor assemblies, including multispectral and panchromatic sensors, to capture images of the same geographic region at different times, aligning these images to correct for misregistration by using machine learning to identify moving objects and applying a background misregistration value to obtain accurate velocity estimates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If images from different sensors are used to estimate velocity, then more data is available for velocity calculation, but misregistration between images causes errors in velocity estimation
Solution Approach 1:
The patent applies preliminary registration correction to images from different sensors before performing velocity estimation. By pre-aligning the images using reference features and calculating registration transformations, the system eliminates misregistration errors that would otherwise contaminate the velocity measurements, thereby enabling accurate velocity estimation to utilize the full multi-sensor data availability
Solution Approach 2:
The patent introduces reference features (stationary objects visible in multiple sensor images) as intermediaries to establish the registration transformation between different sensor images. These reference features serve as mediators that enable the calculation of relative positioning and orientation between sensors, which is then used to correct the velocity estimates by removing spurious motion components
2Productivity
If multiple sensor assemblies are used to capture images, then velocity estimation can be performed, but misregistration between sensor images introduces errors
Solution Approach 1:
The system performs preliminary registration alignment between images from multiple sensor assemblies before velocity estimation. By calculating the relative positioning and orientation of sensors using stationary reference features and applying transformation corrections, the system ensures that subsequent velocity measurements are not contaminated by misregistration artifacts, thereby maintaining high reliability
Solution Approach 2:
The patent employs an iterative feedback mechanism where velocity estimates are refined by comparing measurements across multiple sensor assemblies. The system uses reference features to establish expected stationary object positions, compares these with actual measured positions, and uses the discrepancies to correct the velocity estimates, thereby improving reliability through continuous validation and adjustment
Data Source
AI summary
A method is provided for estimating velocity of a moving object. The method includes generating a raw velocity for the moving object from a position change between a first image generated by a first sensor and a second image generated by a second sensor of a sensor assembly and selecting a plurality of stationary background features that are captured with the moving object in one or more images. A background misregistration value is generated from apparent movement of the plurality of stationary background features and is used to correct the raw velocity.