Home » Technology » Capability

Texture Classification

Deep Vision's data abstraction technology provides real-time texture classification from imagery.

Deep Vision's novel concept of data abstraction quickly transforms abundant sensor data into a form that is efficiently analysed and easily classified.

The abstractions created from the raw data are used to form a texture sensitive distribution which, when decomposed, isolates every unique texture within the image.

Deep Vision's textural classification technology characterises both homogeneous and inhomogeneous (mixed) textures. By utilising a knowledge base, these textures can be classified.

Additionally, Deep Vision's data abstraction technology operates with a throughput of 100+ frames per second.

Deep Vision's textural classification technology distinguishes between, and identifies, textures in complex environments. This makes it an important tool for local mapping and UV route planning.
Exploitation Value
 

  • Surface classification for UAV landing
  • Mapping and mission planning
  • Local atmospheric analysis
  • Environmental analysis

 
Features

  • Real-time textural analysis and classification
  • Multi-textured scenes
  • Classification of inhomogeneous (mixed) textures
  • Operates in complex environments
  • Sensor independent
  • Distinction between permeable and impermeable surfaces (e.g. Clouds and brick).
 Operating Facts
 

  • Operating System: Any (GNU/Linux recommended)
  • Hardware Requirements: None
  • Sensor Modalities: Visual, Thermal, Sonar
  • Timings†: 100+ FPS
  • Runtime Memory Requirements†: 300 KB
  • Storage Requirements‡: 1.1 KB

† Typical. Based on a 640 x 480 data set
‡ Typical. Based on 45 abstractions (avg. 25 symbols each)
 
Input Requirements

  • Archived videos and images
  • Real-time acquisition from visual, thermal, or sonar sensors.