Abstract
The stem diameter of Miscanthus giganteus (MxG) is an important parameter in the measurement of stand volume and yield. To measure the diameters of MxG stems automatically, a vision-based diameter sensor was developed, consisting of a camera and a laser sheet that was slanted downward at a 15° angle. The laser sheet projected Illuminated Line Segments (ILSs) onto the MxG stems, creating detectable features in images that enabled the depth from the camera to the stems and, subsequently, the stem diameters to be measured.To evaluate the method in a sample of 1364 MxG stems, originating from 100 randomly selected images, the depths and diameters of 150 stems were measured manually and plotted versus their automatically measured counterparts. Straight lines with intercept were fitted to both the depth and diameter data. The depth measurements, ranging from 368 to 1486mm, yielded a slope of 1.002, an intercept of -1.2mm and a coefficient of determination of 0.998. The diameter data, ranging from 3.1 to 14.6mm, yielded a slope of 0.845, an intercept of 1.4mm and a coefficient of determination of 0.926.
Original language | English (US) |
---|---|
Pages (from-to) | 298-304 |
Number of pages | 7 |
Journal | Biosystems Engineering |
Volume | 111 |
Issue number | 3 |
DOIs | |
State | Published - Mar 2012 |
ASJC Scopus subject areas
- Food Science
- Agronomy and Crop Science
- Control and Systems Engineering
- Soil Science