What is RAVe?

How AI-powered raster-to-vector conversion is modernizing geospatial production

What you'll learn

• What RAVe is and how it works
• Why raster-to-vector conversion is a critical step in modern geospatial production
• How AI is helping modernize chart production while improving speed and consistency

This article is part of the Outcomes Unlocked series exploring how artificial intelligence is solving complex government and defense challenges.

What is RAVe?

RAVe (Raster Automation to Vector) is an AI-enabled capability developed by Leidos that converts raster maps and charts into structured geospatial vector data.

Instead of manually tracing thousands of individual chart features, RAVe automatically identifies those features and converts them into georeferenced vector data that can be reviewed and validated before operational use.

The capability helps organizations modernize one of the most time-consuming stages of geospatial production.

What is raster-to-vector conversion?

Most legacy maps and nautical charts exist as raster images and raster data is made up of pixels.

While people can interpret those images visually, computers cannot easily recognize individual geographic features within them.

Vector data represents those same features as structured digital objects.

Instead of seeing a coastline as pixels, a vector dataset understands it as a coastline. Instead of viewing a navigation channel as an image, it becomes searchable, editable, measurable geospatial data.

Raster-to-vector conversion is the process of making that transformation.

Why does raster-to-vector conversion matter?

Government organizations increasingly rely on structured geospatial data to support navigation, intelligence, mapping, planning, and mission operations.

Converting legacy raster products into vector data enables organizations to:

• Update products more efficiently
• Search and analyze geographic features
• Integrate data into operational systems
• Improve production consistency
• Support modern digital workflows

Without this conversion, much of that information remains locked inside static imagery.

Why has chart production traditionally been so slow?

Traditional chart production requires analysts to manually identify and digitize thousands of individual features.

Those features may include:

• Coastlines
• Navigation channels
• Depth contours
• Soundings
• Restricted areas
• Submerged hazards

Each feature must be manually interpreted, traced, attributed, and quality checked before the finished product can be delivered.

As chart libraries continue to expand, this manual process becomes increasingly difficult to scale.

How does RAVe modernize chart production?

RAVe applies artificial intelligence to automate much of the raster-to-vector conversion process.

Instead of manually digitizing every feature, analysts receive AI-generated vector data that is ready for review and refinement.

This significantly reduces repetitive production work while enabling experts to focus on validation and quality assurance.

For some workflows demonstrated during development, production time decreased from more than 24 hours to seconds or minutes.

How accurate is RAVe?

For operational users, speed is only valuable if the resulting data can be trusted.

During development, RAVe achieved better than "four nines" (99.99%) first-time-right accuracy while reducing touch labor by approximately 50%.

Automation also improves production consistency by reducing repetitive manual tasks that historically required multiple layers of quality control.

Where can RAVe be applied?

Although initially developed to improve nautical chart production, the underlying capability has broader geospatial applications.

Raster-to-vector conversion can support:

• Maritime navigation
• Aeronautical chart production
• Geographic mapping
• Defense intelligence
• Legacy map modernization
• Digital geospatial workflows

Any organization managing large collections of raster-based geospatial products may benefit from AI-assisted conversion.

Why is Leidos developing AI capabilities like RAVe?

Modern defense and intelligence organizations must process growing volumes of geospatial information while maintaining speed, quality, and trust.

Leidos is developing AI-enabled capabilities that automate repetitive production tasks, improve workflow efficiency, and help organizations modernize legacy geospatial processes.

RAVe represents one example of how AI can transform specialized production workflows into scalable, digital-first operations.

RAVe by the Numbers

The following results were achieved using Leidos' RAVe capability.

• 24+ hours → seconds or minutes for some production workflows
• ~50% reduction in touch labor

• 99.99%+ first-time-right accuracy ("four nines")

  ---

Key takeaways

• RAVe automates raster-to-vector conversion for modern geospatial production
• AI-assisted automation improves chart production speed, consistency, and quality
• Modern geospatial workflows depend on structured digital data, not static imagery
   

LEARN MORE ABOUT HOW AI IS TRANSFORMING GEOSPATIAL INTELLIGENCE