Automated Building Section Evaluation for Defect Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing building evaluation methods are prone to human error and subjective interpretations, and on-site assessments are often incomplete due to practical limitations.

Innovation Solution

An automated system utilizing an image capture apparatus, spatial computing, machine learning models, and blockchain networks for real-time evaluation of structural integrity, enabling remote feedback and report generation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If automated image capture apparatus is used, then operator error is reduced, but device complexity increases

Engineering Contradiction:

Improveevaluation accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent uses image capture apparatus to create visual copies of building sections, which are then analyzed by machine learning models. This replaces manual inspection with automated image-based assessment, reducing operator error while managing system complexity through digital replication of physical inspection processes

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent substitutes mechanical/manual inspection processes with automated image capture and machine learning analysis. The machine learning model processes images to identify defective characteristics, replacing human operators and reducing measurement errors associated with manual evaluation

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Loss of information

If comprehensive building sections are captured, then assessment completeness is improved, but loss of time increases

Engineering Contradiction:

Improveassessment completenessVSAvoidinspection time

Core Design Contradiction:

Loss of informationVSLoss of time

Solution Approach 1:

The patent captures comprehensive images of building sections during the initial inspection phase, storing them for later analysis by machine learning models. This preliminary capture ensures all necessary information is collected upfront, allowing thorough assessment without requiring multiple site visits and reducing total inspection time

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system enables continuous processing of building assessment data through automated machine learning analysis of captured images. Multiple building sections can be evaluated simultaneously and continuously, eliminating idle time between inspection tasks and maintaining productive action throughout the assessment process

Inventive Principle:

Principle #20Continuity of useful action

Data Source

PatentUS12430735B2Methods and systems for evaluating buildings

Publication Date: 2025.09.30 BANK OF AMERICA CORP

AI summary

A method is provided herein for evaluating a physical building, utilizing an automated image capture apparatus, a spatial computing apparatus, a machine learning model, a blockchain network, and a computer processor, the method including capturing and analyzing images of the building, dividing the building into sections, identifying defective characteristic(s) in and assigning a status to each section, sharing the images and status identifiers with a remote user, the remote user reviewing the images and identifiers, and receiving feedback from the remote user. When a section lacks sufficient images to determine whether a defect is present, the section may be flagged, and a supplemental image(s) may be captured and provided by the remote user. When no additional images are needed, the section may be indicated as completed. Otherwise, the section may retain its flagged status, and the aforementioned steps may be repeated as necessary.