Multi-Resolution Image Damage Detection for Wide-Area Assessment

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing damage assessment methods using low-resolution satellite images are prone to errors due to varying image quality, appraiser skill, and lack of consistency, complicating accurate economic damage estimation after natural disasters.

Innovation Solution

A learning engine is trained using a combination of high-resolution and low-resolution images to align and match features, utilizing high-resolution images as ground truth to improve the accuracy of low-resolution damage detection by establishing a correspondence between the outputs of both resolutions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Area of stationary object

If low-resolution satellite images are used for damage assessment, then coverage area and frequency are improved, but measurement precision and reliability deteriorate

Engineering Contradiction:

Improvecoverage areaVSAvoiddamage detection accuracy

Core Design Contradiction:

Area of stationary objectVSMeasurement precision

Solution Approach 1:

The patent combines multiple low-resolution satellite images captured at different times with high-resolution reference images to create a composite damage assessment. By merging temporal sequences of low-resolution images with periodic high-resolution ground truth data, the system achieves both wide coverage and improved measurement precision through data fusion and comparative analysis.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The system performs preliminary damage detection using low-resolution images to identify potential damage areas, then applies high-resolution image analysis selectively to those identified areas. This preliminary screening approach allows wide coverage while maintaining measurement precision where it matters most for damage assessment.

Inventive Principle:

Principle #10Preliminary action

2Productivity

If low-resolution satellite images are used for damage assessment, then productivity is improved, but measurement precision and consistency worsen

Engineering Contradiction:

Improveassessment speedVSAvoiddamage detection accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system applies high-resolution analysis only partially - specifically to areas where damage is suspected based on low-resolution image changes. Rather than processing all images at high resolution, it performs excessive analysis only where needed, maintaining productivity while improving measurement precision for critical damage detection.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The system uses feedback from low-resolution image comparisons to guide subsequent high-resolution analysis. By detecting changes in low-resolution sequences and using those as feedback to trigger targeted high-resolution verification, the system maintains both productivity and measurement precision through adaptive processing.

Inventive Principle:

Principle #23Feedback

3Adaptability or versatility

If manual appraisal of photographs is used for damage assessment, then adaptability is improved, but reliability and consistency worsen

Engineering Contradiction:

Improveflexibility in assessmentVSAvoidassessment consistency

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The system enables self-service automated damage detection by comparing satellite images against learned patterns of damage. The automated system serves itself by identifying damage without human intervention in routine cases, improving reliability and consistency while maintaining adaptability through machine learning models that can be updated with new damage patterns.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces the mechanical system of manual human appraisal with an automated image processing and machine learning system. This substitution eliminates human subjectivity and inconsistency while maintaining adaptability through programmable algorithms that can be adjusted to recognize different damage types and assessment criteria.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4211643B1Damage detection based on images with varying resolutions

Publication Date: 2026.02.11 CONCIRRUS LTD

Figure 1
Figure 2A
Figure 2B

AI summary

Disclosed are devices, systems, apparatus, methods, products, and other implementations, including a method for detecting damage in a geographical area that includes receiving a first image of the geographical area, with the first image having a first resolution, and detecting damage to at least one object appearing in the first image in response to applying a trained learning engine to the first image. The learning engine is trained to detect damage in geographical areas based on received one or more images in a second set of images having a second resolution higher than the first resolution, and based on one or more images in a first set of images having the first resolution. The one or more images in the first set of images and the one or more images in the second set of images include one or more overlapping portions in which one or more objects appear.