BRIDGE INSPECTION PLATFORM · WEB

Bridge Inspection Platform

Build Traceable Links between Inspection Images, Bridge Components and Defects

For bridge-inspection data management and delivery, the platform connects missions, AI-assisted screening, component mapping, manual review and report generation in one project workflow.

Explore Platform Capabilities Consult the Product Team

Project-level Data ManagementComponent-to-defect AssociationAI-assisted ScreeningWord Report Output

Bridge inspection platform digital twin and defect distribution — Linked view of the 3D model, component tree, camera poses and defect information

PRODUCT OVERVIEW

From Discrete Images to Component-level Inspection Results

The platform uses each bridge project as a management unit, organizing the 3D model, components, inspection records, images, suspected defects and reports. AI assists screening, while professionals verify, edit and archive results.

Project-level Organization

Archive models, components, inspection records, images, defects and reports around the same bridge project.

Component-level Association

Associate discrete images and suspected defects with specific components for localization, filtering and reinspection.

Human-AI Collaborative Review

AI assists screening; professionals confirm, revise and approve final results.

Template-based Deliverables

Organize reviewed results by time, component, defect and severity to create project reports.

END-TO-END WORKFLOW

Seven Steps Connect Field Capture, Professional Review and Delivery

The platform does not replace professional bridge assessment. It turns data processing, spatial association and result organization into an executable workflow.

Create the Project

Import the 3D Tiles model, component tree and project information into one data container.

Submit an Inspection

Record date, pilot, drone, camera and target defect types, then upload JPG / PNG imagery.

AI-assisted Screening

Schedule computing resources for image deduplication, suspected-defect detection, size extraction and result merging.

Map Components

Use camera parameters, position and pose to associate images and suspected defects with bridge components.

Filter for Review

Filter data for verification by mission, component, defect type, status and severity.

Manual Review

Verify and revise defect information across 3D, original and historical imagery.

Generate Reports

Compile reviewed results into a Word report with distribution graphics, component tables and illustrated defect details.

PROJECT & INSPECTION

Standardize Every Inspection Submission within a Project

A project links its 3D model, component tree and base data. Each inspection records date, pilot, drone and camera, and selects defect types and uploaded images.

Supports JPG / PNG inspection imagery
Records equipment, personnel and mission information
Displays image-review and AI-analysis progress
Supports stopping or restarting inference

3D DIGITAL TWIN

Locate Components and Their Defects in 3D

The platform loads a 3D Tiles model and combines its component tree, camera poses and suspected defects to build spatial links from bridge structure to field imagery.

Camera poses associated with defects in the bridge model — **Camera Poses & Defect Distribution**Use camera frustums to mark capture direction and view data by component and defect type.

AI-ASSISTED SCREENING

Help Reviewers Start with the Data that Matters

Configure project-specific defect types such as cracks, corrosion and spalling. AI assists image screening while preserving detection and review status.

Multi-condition Filtering

Filter by mission, capture time, component, detection status, review status, defect type and severity.

Batch Processing

Batch download, review and reinference within the project workflow reduce repetitive work.

Threshold Filtering

Filter minor results by defect-size thresholds so reviewers can focus on relevant data.

AI recognition and sizing assist screening. Final defect category, severity and treatment conclusions require professional review against project requirements.

AI DEFECT CATEGORIES

Five Visible Bridge Defect Types Supported by AI Models

The platform can configure project-specific AI models to assist screening five types of visible defects in high-resolution inspection imagery, with suspected findings passed to professionals for review.

Example of cracking on a bridge concrete surface

Cracks

Assists in locating linear, branching or network cracking on concrete surfaces and retains suspected outlines and dimensions for professional review.

Corrosion

Assists in screening visible rust spots, staining and corroded areas on components so reviewers can assess the affected area against the source image.

Example of concrete spalling and damage on a bridge

Spalling / Damage

Assists in identifying surface spalling, edge damage and local material loss in concrete, providing evidence for localization and reinspection.

Exposed Reinforcement

Assists in finding visible reinforcement where the concrete cover has been lost and links suspected findings to the corresponding component and inspection image.

Honeycombing

Assists in screening visible dense voids, pits and rough or uneven areas on concrete surfaces for further professional confirmation.

The images illustrate supported defect categories and do not represent detection results or severity. AI output is treated as suspected-defect evidence; final conclusions require professional review of source imagery, component conditions and project criteria.

HUMAN REVIEW

Three-panel Review Keeps Every Revision in One Workspace

View the component and 3D location on the left, verify the original image and defect outline in the center, and retrieve historical imagery on the right. Reviewers can edit type, outline, size, severity and result status.

Linked 3D component position and 2D source image
Single-image and batch review
Editable defect outlines and attributes
Reviewed results ready for reporting

Three-panel manual review interface of the bridge inspection platform — Joint verification across 3D position, source and historical imagery

REPORT DELIVERY

Organize Reviewed Results into Deliverable Project Reports

Compile defects using project templates and filters into a Word inspection report containing distributions, component details and illustrated records.

Defect Distribution

Show reviewed defect distribution by bridge or span.

Component Details

Organize component and defect information for localization and reinspection.

Illustrated Details

Combine 3D screenshots, defect crops and attributes.

Template Management

Select or maintain report templates for project delivery requirements.

DATA CHAIN

One Clear Data Chain from Input through Processing to Output

Input

3D Tiles Model
Component-tree File
JPG / PNG Imagery
Camera & Pose Data

Processing

Image Deduplication
AI-assisted Detection
Size Extraction
Suspected-defect Merging

Organization

Projects & Inspection Records
Component Mapping
Review Status
Historical-image Association

Output

Defect Distribution
Component Details
Illustrated Details
Word Inspection Report

Functions shown come from current materials and UI examples. Deployment, supported models and file formats, AI defect types, report templates and performance boundaries are subject to project documents and the delivered version.

PRODUCT WORKFLOW

Clear Roles for Planning, Onboard Execution, Controller Monitoring and Result Management

01 / PLAN