Forest Service Research & Development International Programs
United States Department of Agriculture
An Automated Field-level Wood Identification System
Shelley GardnerUSDA Forest Service
Forest Service Machine Vision
• Botanical identification of wood and wood products is a limiting factor for enforcing laws that combat illegal logging and accurate wood identification in the marketplace
• Human capacity is limited– Time to train– Cost of training– Mobility of individuals
The Problem
Forest Service Machine Vision
Build a machine vision system that can be used as a field deployable tool to identify woody taxa.
Need a technological rather than a human solution
The Project
Forest Service Machine Vision
Patterns in Woods – 3D and 2D
The Project
Forest Service Machine Vision
A simple image A 3D representation
Anatomy of a Digital Image
Forest Service Machine Vision
• Digital imaging can be more sensitive than the human eye
• These systems collect copious amounts of data
• The collected data can be mined to enhance the knowledge of wood anatomical structure
The Approach
Pros
• Requires robust programming
• Needs Big Data
Cons
Forest Service Machine Vision
The ApproachDevelop an open system• Utilize smart phone• Utilize open source software• Have a central database• Repeatable• Easy-to-use• Economical
Forest Service Machine Vision
Summary
Take-Home Message• This is an extremely accurate, field-
deployable, easy to use wood identification system that anyone able to operate a smart phone can use.
Forest Service Machine Vision
Questions?
• USDA Forest Service is looking for cooperators to develop this into a smart phone app
• Contact Shelley Gardner, Illegal Logging Program Coordinator at• [email protected] • 202.644.4554