How Change Detection Works

Summary: Bedrock uses computer vision and AI to compare construction drawings at the pixel level, then categorizes detected changes by type. The process is automatic, requiring no manual alignment or configuration.

Overview

Change detection in Bedrock happens in four stages:

  1. Ingestion: PDFs are processed into normalized images
  2. Matching: Sheets are paired between prior and current revisions
  3. Alignment: Paired sheets are geometrically aligned
  4. Detection: Differences are identified and categorized

Each stage uses purpose-built algorithms optimized for construction drawings.

Stage 1: Ingestion

When you upload a PDF drawing set, Bedrock:

  1. Extracts each page as a high-resolution image
  2. Identifies sheet boundaries and title blocks
  3. Extracts text layers for metadata
  4. Normalizes resolution and color profiles
ParameterSpecification
Processing Resolution300 DPI equivalent
Color HandlingConverted to grayscale for comparison
Text ExtractionOCR + native PDF text layers

Stage 2: Sheet Matching

Bedrock automatically matches corresponding sheets between revisions without requiring manual pairing.

How Matching Works

The system analyzes multiple signals to identify sheet pairs:

SignalWeightDescription
Sheet NumberHighA1.01 in prior matches A1.01 in current
Sheet TitleMedium”First Floor Plan” matches similar titles
Visual SimilarityMediumOverall drawing structure and content
Title Block PositionLowLocation of title block on page

Handling Edge Cases

ScenarioHow It’s Handled
Renumbered sheetsVisual similarity + title matching
New sheetsFlagged as “Added” in current revision
Deleted sheetsFlagged as “Removed” from prior revision
Split sheetsEach new sheet matched to original

Stage 3: Alignment

Matched sheets must be geometrically aligned before comparison. Traditional tools require manual 3-point alignment. Bedrock does this automatically.

Automatic Alignment Process

  1. Feature Detection: Identify stable reference points (grid lines, column markers, title blocks)
  2. Transform Calculation: Compute rotation, scale, and translation
  3. Image Registration: Apply transformation to align drawings
  4. Quality Check: Verify alignment meets accuracy threshold

Alignment Accuracy

Drawing TypeTypical Accuracy
Architectural plansSub-pixel
Structural framingSub-pixel
MEP layouts1-2 pixels
Site plans2-3 pixels

Accuracy varies based on drawing quality and the presence of stable reference geometry.

Stage 4: Change Detection

With sheets aligned, the system identifies what changed between revisions.

Detection Methods

MethodWhat It Finds
Pixel DifferencingRaw visual differences between images
Contour AnalysisChanges to lines and shapes
Text ComparisonModified, added, or deleted text
Symbol DetectionChanges to common drawing symbols

Change Categories

Detected changes are automatically categorized:

CategoryDescriptionExamples
AdditionNew content in current revisionNew walls, equipment, notes
DeletionContent removed from prior revisionRemoved doors, deleted dimensions
ModificationChanged existing contentMoved walls, updated dimensions

Filtering

Not all differences are meaningful changes. Bedrock filters out:

  • Title block updates (dates, revision numbers)
  • Watermarks and stamps
  • Minor PDF rendering variations
  • Compression artifacts

Technical Specifications

SpecificationValue
Detection SensitivityConfigurable (default: medium)
Minimum Change Size0.1” at print scale
Maximum Sheet Size36” x 48” (ARCH E)
Processing MemoryOptimized for large drawings

Limitations

Change detection has known limitations:

LimitationImpactMitigation
Scanned drawingsLower alignment accuracyUse native PDFs when possible
Rotated sheetsMay require manual verificationSystem flags suspected rotation
Color-coded changesMay miss color-only changesGrayscale processing
3D viewsLess reliable than 2D plansFocus on plan views

FAQ

How accurate is change detection?

Graphic element detection exceeds 99% accuracy on native PDFs. Text detection varies based on PDF text layer quality. Scanned drawings have lower accuracy due to image noise.

Does detection work on all drawing types?

Yes, but accuracy is highest on 2D plan views (floor plans, elevations, sections). 3D perspectives and renderings are less reliable due to shading complexity.

Can I adjust detection sensitivity?

Yes. Higher sensitivity catches smaller changes but may increase noise. Lower sensitivity reduces noise but may miss subtle changes. Most users find the default setting optimal.

What causes false positives?

Common causes: PDF regeneration differences, font substitution, compression artifacts, scan quality variations. Filtering algorithms reduce but don’t eliminate false positives.

Key Takeaways

  • Change detection happens in four stages: ingestion, matching, alignment, detection
  • Sheet matching uses multiple signals (number, title, visual similarity)
  • Alignment is automatic with sub-pixel accuracy on most drawings
  • Changes are categorized as additions, deletions, or modifications
  • Filtering removes noise (title blocks, watermarks, rendering differences)
  • Accuracy is highest on native PDFs; scanned drawings have limitations

Last updated: 2026-02-04