Research on Forest Plot Survey System Based on Visual Odometer
CSTR:
Author:
Affiliation:

Clc Number:

Fund Project:

  • Article
  • |
  • Figures
  • |
  • Metrics
  • |
  • Reference
  • |
  • Related
  • |
  • Cited by
  • |
  • Materials
  • |
  • Comments
    Abstract:

    The visual odometer technology was used to restore the posture of the continuous photographic sequences. The sampleplot survey system was constructed based on the images of the restored posture. The system can estimate the position and DBH of the sample trees in the sampleplot by recovering the image posture scale, defining the plot coordinate system, and sample tree marking. Totally 12 circular sample plots with radius of 7.5m were continuously photographed to obtain the image sequence. And then the obtained image sequence was processed with the constructed sample plot survey system to measure the position and DBH of the sample trees. The results showed that the deviations of estimated values of sample tree position of the plots in the xaxis and yaxis directions were 0.04m and -0.03m, respectively, and the root mean square error (RMSE) was 0.21m and 0.17m, respectively. The BIAS and RMSE of the estimated DBH were 0.09cm (0.51%) and 0.88cm (5.03%), respectively. The results showed that the visual odometer technology had a great potential to restore the image posture and to use the nonpoint cloud method to estimate the tree position and the breast diameter from the picture by plot survey method.

    Reference
    Related
    Cited by
Get Citation
Share
Article Metrics
  • Abstract:
  • PDF:
  • HTML:
  • Cited by:
History
  • Received:June 21,2019
  • Revised:
  • Adopted:
  • Online: October 10,2019
  • Published: October 10,2019
Article QR Code