Adaptive Imagery Acquisition Tasking for Change Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional image acquisition systems often capture excessive imagery data, with only a small portion showing changes in geographic areas, leading to inefficient resource allocation and increased computational burden in identifying updates.

Innovation Solution

A machine-learning based system that uses a binary classifier model to detect changes in geographic areas by analyzing images at different times, adjusting the frequency and spatial resolution of image acquisition based on detected changes, thereby prioritizing areas with higher levels of change.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If image acquisition systems capture imagery data at high frequency and high spatial resolution for all areas, then the completeness and detail of geographic data are improved, but the resource consumption and computational burden increase significantly

Engineering Contradiction:
Improvespatial resolution of imagery dataVSAvoidresource consumption for data capture and processing
Core Design Contradiction:
Measurement precisionVSLoss of energy

Solution Approach 1:

The system applies different image acquisition strategies to different geographic areas based on their change characteristics. High-resolution frequent capture is applied only to areas with high change levels, while low-resolution or less frequent capture is applied to stable areas, optimizing resource allocation according to local needs

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The image acquisition frequency and resolution are made dynamic rather than static. The system continuously monitors change levels and adjusts acquisition parameters in real-time, increasing frequency for areas undergoing rapid change and decreasing it for stable areas, thereby adapting resource allocation to actual needs

Inventive Principle:
Principle #15Dynamics

2Loss of time

If image acquisition frequency is increased for all areas, then the timeliness of detecting geographic changes is improved, but the inefficiency in resource allocation worsens due to capturing excessive data from stable areas

Engineering Contradiction:
Improvetimeliness of change detectionVSAvoidefficiency of resource allocation
Core Design Contradiction:
Loss of timeVSProductivity

Solution Approach 1:

The system implements location-specific acquisition frequencies based on measured change characteristics of each area. Urban areas with high construction activity receive frequent monitoring while rural stable areas receive less frequent monitoring, ensuring timely detection where needed without wasting resources elsewhere

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system uses detected change levels as feedback to adjust future acquisition frequency. When changes are detected, the system increases monitoring frequency for that area; when no changes are detected over time, it reduces frequency, creating a closed-loop adaptive system that optimizes timeliness versus efficiency

Inventive Principle:
Principle #23Feedback

3Measurement precision

If uniform high-resolution imaging is applied to all geographic areas, then the quality of geographic data is improved, but the cost of data acquisition and processing increases

Engineering Contradiction:
Improvequality of geographic dataVSAvoidvolume of imagery data to be processed
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The system matches image resolution quality to local requirements. Areas with high change levels receive high-resolution imaging to capture detailed changes, while stable areas receive lower resolution imaging since high detail is not necessary for monitoring unchanged features, thereby reducing total data volume while maintaining necessary quality

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system applies high-resolution imaging only partially, specifically to areas where it is necessary for detecting changes. Rather than uniformly applying high resolution everywhere (excessive action), it concentrates high-resolution capture only where change detection requires it, reducing overall data processing burden

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3504661B1Change detection based imagery acquisition tasking system
Publication Date: 2023.08.16 GOOGLE LLC
  • EP3504661B1 patent drawingFigure 1
  • EP3504661B1 patent drawingFigure 2
  • EP3504661B1 patent drawingFigure 3

AI summary

Systems and methods for tasking an image acquisition system are provided. In one embodiment, a method includes obtaining data descriptive of a plurality of images associated with a geographic area. The method includes analyzing at least a subset of the plurality of images to determine an occurrence of one or more changes associated with the geographic area. The method includes determining a level of change associated with the geographic area based, at least in part, on the occurrence of the one or more changes associated with the geographic area. The method includes providing a control command to an image acquisition system to adjust an acquisition of imagery data associated with the geographic area based, at least in part, on the level of change.