The application of machine vision to the selective harvest of green asparagus
Humburg, Daniel Sherman
This item is only available for download by members of the University of Illinois community. Students, faculty, and staff at the U of I may log in with your NetID and password to view the item. If you are trying to access an Illinois-restricted dissertation or thesis, you can request a copy through your library's Inter-Library Loan office or purchase a copy directly from ProQuest.
Permalink
https://hdl.handle.net/2142/23602
Description
Title
The application of machine vision to the selective harvest of green asparagus
Author(s)
Humburg, Daniel Sherman
Issue Date
1991
Doctoral Committee Chair(s)
Reid, John F.
Department of Study
Engineering, Agricultural
Discipline
Engineering, Agricultural
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
Ph.D.
Degree Level
Dissertation
Keyword(s)
Engineering, Agricultural
Language
eng
Abstract
A machine vision system was developed and tested to select and locate harvestable spears of asparagus. An image acquisition vehicle was fabricated to videotape portions of asparagus rows. The difference in reflectance between soil and vegetative material to near infrared light was used to obtain contrasted images of asparagus spears on a soil background. A narrow-band optical bandpass filter was used to enhance the difference in soil and plant reflectance. Gray-level thresholding was used to separate image pixels into object regions and background. Vertical run-length filtering was used to partially eliminate object regions associated with soil reflections and weeds. Vertical runs of object color pixels shorter than a predetermined minimum were filtered from the images. Run-length encoding and connectivity analysis was used to identify all horizontal runs of object pixels belonging to a single object region. The list of runs for a single region was used to calculate spatial parameters of the region. Asparagus spears were longer in pixel length than other object regions and were selected on that criteria. A height calibration procedure provided a set of equations for estimating the height of vertically oriented objects based on their length and position in the image. Spears were located by transforming the image coordinates of the spear bases to ground coordinates. A calibration procedure that assumed the soil surface to be a plane, used to obtain an image-to-ground transformation. Videotape of row segments acquired in the field was analyzed. A guidance rail for the image acquisition vehicle provided a directrix to relate measurements across the row, made by the researchers, to measurements made by the vision system. A series of marker pegs placed along the row provided a reference in the images to the position of the vehicle along the row. The locations of harvestable spears measured in the field were compared to the locations of spears found by the vision system in the laboratory. The vision system correctly identified from 86% to 97% of the harvestable spears in six fifteen meter row segments. The system was able process one image in approximately 10 seconds. (Abstract shortened with permission of author.)
Use this login method if you
don't
have an
@illinois.edu
email address.
(Oops, I do have one)
IDEALS migrated to a new platform on June 23, 2022. If you created
your account prior to this date, you will have to reset your password
using the forgot-password link below.
