You are here: Home / Research Topics / Research Highlights / Individual Highlight

Research Highlights

Individual Highlight

Improved Automated Detection of Surface Defects on Hardwood Logs

Photo of High-resolution point cloud image of a scanned red oak log. Ed Thomas, USDA Forest ServiceHigh-resolution point cloud image of a scanned red oak log. Ed Thomas, USDA Forest ServiceSnapshot : In less than one second, a new parallel computer algorithm processes more than a million surface data points on a hardwood log to find the defects.

Principal Investigators(s) :
Thomas, Ed 
Research Station : Northern Research Station (NRS)
Year : 2014
Highlight ID : 635

Summary

Determining the size and location of surface defects is crucial to evaluating the potential yield and value of hardwood logs. Recently Forest Service scientists developed a surface-defect detection algorithm using the Java computer language. The algorithm was developed around an earlier laser scanning system that had poor resolution along the length of the log (15 scan lines per foot). The scientists and their partners constructed a newer laser scanning system with much higher resolution (192 scan lines per foot) along a log's length. The increased resolution and the slower processing speed of the Java-based algorithm required a new approach. The revised algorithm was designed around the higher resolution data and employs parallel processing technology. The improved processing power permits a more in-depth analysis of the higher resolution scan data, leading to improved detection results.

Forest Service Partners

External Partners

 
  • Liya Thomas, Concord University, Athens, WV 24712