TY - CONF
T1 - Towards image-based measurement of accurate apple size and yield using stereo vision cameras
AU - Mirbod, Omeed
AU - Choi, Daeun
AU - Heinemann, Paul
AU - Marini, Richard
N1 - Funding Information:
This study was supported by the State Horticultural Association of Pennsylvania, The Pennsylvania State University, and the USDA National Institute of Food and Agriculture Multistate Research under Project #PEN04653 and Accession #1016510. The authors would also like to thank Dr. Robert Crassweller, Mr. Don Smith, and Dr. James Schupp for their help on this project.
Publisher Copyright:
© ASABE 2020 Annual International Meeting.
Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.
PY - 2020
Y1 - 2020
N2 - Accurate measurement of fruit size in apple orchards before harvest can have important implications on profits and management practices. Obtaining a distribution of fruit size can be labor intensive for a large orchard and therefore requires an automated system that can quickly and accurately size fruit on every tree. This study proposes an automated imaging system that uses stereo vision for finding the metric surface area of apples. Deep convolutional neural network models were utilized to classify apples as ideal candidates for sizing based on their orientation and visibility in an image. The results produced a correlation of apple size to apple weight of R2=0.69 making the system capable of capturing variability in fruit diameter distribution that ranges by 1cm (or equivalently 60 grams). There was also an improvement in correlation to yield when combining fruit size with fruit count than when utilizing fruit count alone.
AB - Accurate measurement of fruit size in apple orchards before harvest can have important implications on profits and management practices. Obtaining a distribution of fruit size can be labor intensive for a large orchard and therefore requires an automated system that can quickly and accurately size fruit on every tree. This study proposes an automated imaging system that uses stereo vision for finding the metric surface area of apples. Deep convolutional neural network models were utilized to classify apples as ideal candidates for sizing based on their orientation and visibility in an image. The results produced a correlation of apple size to apple weight of R2=0.69 making the system capable of capturing variability in fruit diameter distribution that ranges by 1cm (or equivalently 60 grams). There was also an improvement in correlation to yield when combining fruit size with fruit count than when utilizing fruit count alone.
UR - http://www.scopus.com/inward/record.url?scp=85096609558&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85096609558&partnerID=8YFLogxK
U2 - 10.13031/aim.202001115
DO - 10.13031/aim.202001115
M3 - Paper
AN - SCOPUS:85096609558
T2 - 2020 ASABE Annual International Meeting
Y2 - 13 July 2020 through 15 July 2020
ER -