Masters Theses

Orcid ID

Date of Award


Degree Type


Degree Name

Master of Science


Biosystems Engineering

Major Professor

Hao Gan

Committee Members

Lori A. Duncan, Annette L. Wszelaki


It is estimated that nearly 75% of major crops have some level of reliance on pollination. Humans are reliant on fruit and vegetable crops for many vital nutrients. With the intensification of agricultural production in response to human demand, native pollinator species are not able to provide sufficient pollination services, and managed bee colonies are in decline due to colony collapse disorder, among other issues. Previous work addresses a few of these issues by designing pollination systems for greenhouse operations or other controlled production systems but fails to address the larger need for development in other agricultural settings with less environmental control. In response to this crisis, this research aims to act as a vital first step towards the development of a more robust autonomous pollination system for agricultural crop production. The main objective of this research is to develop a flower detection and mapping system for a field crop setting. This research presents a method to detect and localize tomato flowers within a three-dimensional (3D) region. Tomato plants were grown in a raised-bed garden where images were collected of the overhead view of the plants. Images were then stitched together using a photogrammetry technique, accomplished by the Pix4Dmapper software, to form an orthomosaic and 3D representation of the raised-bed garden from a high spatial resolution aerial view. Various machine learning architectures were trained to detect tomato flowers from overhead images and were then tested on the orthomosaic images produced by the Pix4D software. The coordinates of the detected flowers in the orthomosaic were then compared to the 3D model representation to find approximate 3D coordinates for each of the flowers relative to a predefined origin. This research serves as a first step in autonomous pollination by presenting a way for machine vision and machine learning to be used to identify the presence and location of flowers on tomato crops. Future work will aim to expand flower detection to other crops varieties in varying field conditions.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."