Plug transplant production technology has been widely used in the plant production industry and is continuing
to expand in scale. In order to achieve optimal control of the environment for a population of transplants growing in multi–cell
plug trays, frequent measurement of the population.s structural and functional properties is required. A binocular stereovision
system incorporating two digital cameras and two computers was developed, based on a pixel stereological algorithm, to
analyze the aggregate growth variables of a plug tray of transplants. Three–dimensional (3–D) color images of the transplant
population were reconstructed from pairs of digital two–dimensional (2–D) color images taken under a constant light
environment. Sweet potato (Ipomoea batatas (L.) Lam.) was used as a test plant. Average height, leaf area, projected leaf area,
and mass volume of the transplant population were calculated at a pixel level by extracting four user–selected colors. Image
analysis took 2 to 3 min for each transplant population measurement. The aggregate growth variables of the transplant
population calculated from image analysis correlated closely with corresponding average height, leaf area, and fresh and
dry masses determined from destructive measurements. The binocular stereovision system using the pixel stereological
algorithm successfully estimated the values of the transplant growth variables. This image analysis method has the potential
to automatically provide necessary aggregate transplant physical properties for identifying growth responses to various
environmental conditions. This technique will in turn lead to the selection of optimum environmental conditions for transplant
growth.