Artificial Intelligence Approach for Tomato Detection and Mass Estimation in Precision Agriculture

Jaesu Lee, Haseeb Nazki, Jeonghyun Baek, Youngsin Hong and Meonghun Lee
Additional contact information
Jaesu Lee: Department of Agricultural Engineering, National Institute of Agricultural Sciences, Jeollabuk-do 55365, Korea
Haseeb Nazki: Department of Computer Science, University of St Andrews, Fife KY16 9AJ, UK
Jeonghyun Baek: Department of Agricultural Engineering, National Institute of Agricultural Sciences, Jeollabuk-do 55365, Korea
Youngsin Hong: Department of Agricultural Engineering, National Institute of Agricultural Sciences, Jeollabuk-do 55365, Korea
Meonghun Lee: Department of Agricultural Engineering, National Institute of Agricultural Sciences, Jeollabuk-do 55365, Korea

Sustainability, 2020, vol. 12, issue 21, 1-15

Abstract: Application of computer vision and robotics in agriculture requires sufficient knowledge and understanding of the physical properties of the object of interest. Yield monitoring is an example where these properties affect the quantified estimation of yield mass. In this study, we propose an image-processing and artificial intelligence-based system using multi-class detection with instance-wise segmentation of fruits in an image that can further estimate dimensions and mass. We analyze a tomato image dataset with mass and dimension values collected using a calibrated vision system and accurate measuring devices. After successful detection and instance-wise segmentation, we extract the real-world dimensions of the fruit. Our characterization results exhibited a significantly high correlation between dimensions and mass, indicating that artificial intelligence algorithms can effectively capture this complex physical relation to estimate the final mass. We also compare different artificial intelligence algorithms to show that the computed mass agrees well with the actual mass. Detection and segmentation results show an average mask intersection over union of 96.05%, mean average precision of 92.28%, detection accuracy of 99.02%, and precision of 99.7%. The mean absolute percentage error for mass estimation was 7.09 for 77 test samples using a bagged ensemble tree regressor. This approach could be applied to other computer vision and robotic applications such as sizing and packaging systems and automated harvesting or to other measuring instruments.

Keywords: artificial intelligence; convolutional neural network; fruit size estimation; image processing; precision agriculture; machine-learning; mass estimation; tomato detection (search for similar items in EconPapers)
JEL-codes: O13 Q Q0 Q2 Q3 Q5 Q56 (search for similar items in EconPapers)
Date: 2020
References: View complete reference list from CitEc
Citations: View citations in EconPapers (3)

Downloads: (external link)
https://www.mdpi.com/2071-1050/12/21/9138/pdf (application/pdf)
https://www.mdpi.com/2071-1050/12/21/9138/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jsusta:v:12:y:2020:i:21:p:9138-:d:439409

Access Statistics for this article

Sustainability is currently edited by Ms. Alexandra Wu

More articles in Sustainability from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().