784 PB 430 NONCONTACT GROWTH ANALYSIS USING 3-D IMAGE PROCESSING

in HortScience
View More View Less
  • 1 Department of Horticulture, Michigan State University, East Lansing, MI 48824-1325

A computer vision system for noncontact growth analysis was developed. Front and side images of a plant were captured simultaneously using a mirror system and CCD camera and were magnetically stored on a magneto-optical disk. Images acquired at night were obtained by irradiating plants with incandescent light filtered to wavelengths of 850 nm and greater. Images were automatically captured and saved every 12 minutes. After images were collected, outlines of plant shape were extracted from stored images, a three-dimensional center line of the plant was extracted from the outline, and the elongation rate was computed. The outline extraction algorithm was modified to improve spatial resolution of images, and the thinning algorithm created a representative line of the plant by calculating a center line of the stem so the three-dimensional length could be calculated. Results of growth analysis on Verbena bonariensis L. plants grown under three photoperiods (8, 12, and 16 hours) and three day/night air-temperature combinations (15/25, 20/20, and 25/15) will be presented.

If the inline PDF is not rendering correctly, you can download the PDF file here.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 70 12 1
PDF Downloads 47 12 2