Infrastructure Inspection

Bridges, pipelines, power lines, and industrial structures need regular inspection. Inspectors climb scaffolding, rappel down structures, or review drone footage frame by frame. The process is slow, and subjective: two inspectors looking at the same crack may rate its severity differently.

Datature Vi trains a model on your inspection imagery. You label photos showing different types of damage (cracks, corrosion, spalling, deformation) at different severity levels, and the model learns to produce consistent assessments. When paired with drone footage, it can process hundreds of images from a single survey and flag the locations that need attention.

For an interactive overview of this application, visit the infrastructure inspection use case on vi.datature.com.

Common applications

Task

What the model does

Crack detection

Identifies and describes cracks in concrete, steel, or masonry

Corrosion assessment

Flags areas of rust, oxidation, or material degradation

Spalling detection

Identifies concrete spalling and exposed reinforcement

Severity grading

Assigns severity levels (minor, moderate, severe) to visible damage

Report generation

Produces structured inspection findings from image data

Choose your task type

Approach

Best for

Output

Freeform Text (JSON)

Structured inspection reports with damage type, location, and severity

JSON: {"damage_type": "spalling", "severity": "severe", "exposed_rebar": true}

Visual Question Answering (VQA)

Quick pass/fail assessments or specific questions about a structure

Text answer: "Yes, there is moderate cracking along the east face of the pillar."

Phrase Grounding

Locating the damaged area for measurement or follow-up

Bounding box around the damage zone

Annotation examples

Image

Question

Answer

Intact concrete pillar

Describe the condition of this structural element.

The concrete surface appears intact with no visible cracking, spalling, or corrosion. Surface shows normal weathering.

Pillar with spalling

Describe the condition of this structural element.

Severe spalling on the west face. Concrete has broken away exposing corroded reinforcement bars. Affected area is approximately 0.5 meters wide.

Minor surface crack

Describe the condition of this structural element.

A hairline surface crack runs vertically along the lower third of the column. No spalling or rebar exposure. Severity: minor.

Deploy and test

import json
from vi.inference import ViModel

model = ViModel(
    run_id="your-run-id",
    secret_key=".your-secret-key.",
    organization_id="your-organization-id",
)

result, error = model(
    source="drone_image.jpg",
    user_prompt="Assess the structural condition visible in this image.",
    generation_config={"temperature": 0.0, "do_sample": False}
)

if error is None:
    report = json.loads(result.result)
    print(f"Damage type: {report['damage_type']}")
    print(f"Severity: {report['severity']}")

Training tips

Use your own severity scale: define severity levels that match your organization's inspection standards (e.g., NBI condition ratings for bridges, API 570 for piping) and apply them consistently across all annotations.

Include clean/undamaged examples: the model needs to learn what a healthy structure looks like, not only damaged ones.

Capture from multiple distances: drone surveys produce images at different distances. Include close-ups (for detail) and wider shots (for context) in your training data.

Annotate with precise language: "hairline crack on the lower third" is more useful than "crack visible." Consistency and precision in annotations translate directly to consistency in model output.

Next steps

Structured Data Extraction

Return structured JSON inspection reports for asset management systems.

Phrase Grounding

Localize damage regions in images for measurement and documentation.

Chain-of-Thought Reasoning

Multi-step assessments: identify damage, classify type, then grade severity.