Click on “Download PDF” for the PDF version or on the title for the HTML version.


If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Inter-plant Spacing Sensing at Early Growth Stages Using a Time-of-Flight of Light Based 3D Vision Sensor

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org

Citation:  2010 Pittsburgh, Pennsylvania, June 20 - June 23, 2010  1009216.(doi:10.13031/2013.29832)
Authors:   Akash D Nakarmi, Lie Tang
Keywords:   3D vision, spacing sensing, early growth stages, time-of-flight, image processing

Uniform plant spacing is always desired for equal distribution of water and nutrients among plants. Researchers in the past have shown that variations in plant spacing result in significant variation in final crop yields. Planter manufacturers and researchers have been working closely to develop computer vision-based automatic interplant spacing sensing systems. Current systems mostly utilize top-view images using a stereo rig, or a video camera. These systems are highly sensitive to color variations introduced by shadow formations and glares and have difficulties when plant canopies start occluding. We developed an interplant spacing sensing system using a time-of-flight (TOF) based 3D vision sensor. The camera was capable of capturing depth and intensity data with one single shot. The depth images captured from the side were stitched together using distance information from a wheel encoder in conjunction with a feature-based image sequencing process. Multiple layers of image data were used for stem location identification. The use of depth images made the plant identification less sensitive to color variations. A covered vehicle was designed to prevent the sunlight from directly shedding on the plants and to reduce the interference from wind, which in turn made the system usable throughout the day. The vertical camera position was easily adjustable to work with different growth stages of the crops. The use of side-view images made the system capable to detecting inclined plants and therefore, boosted the performance of the system in precisely locating the stem centers, and thereby minimized the measurement errors. Based on the initial trials on corn plants of growth stages V3-V6, the system has achieved 100% plant identification accuracy with a RSME of 0.15 cm for inter-plant spacing measurements.

(Download PDF)    (Export to EndNotes)