Publication Date


Document Type


First Advisor

Bharti, Pratool

Degree Name

M.S. (Master of Science)

Legacy Department

Department of Computer Science


Climatic conditions like temperature, drought, and heavy metals disturb plant cell structures and, ultimately, plant growth that significantly affects crop production. Due to increasing climate change, maize crop yields are projected to decline by 24% by the end of century. With the increase in food demands and decrease in agricultural land and water resources, the space for effective farming is left much desired. Though limited to a few crops at this moment, Indoor Vertical Farming is one technique that requires much less land space, water, soil, and sunlight when compared to traditional farming. Vertical farming allows artificial control of temperature, light, humidity, water, and gases that make food production indoors possible. Apart from providing an optimal growing environment, tracking crop growth efficiently and automatically is where Indoor Vertical Farming thrives, which will be the focus of this thesis.Growth tracking mechanisms in traditional agricultural practices are typically labor-intensive, hazardous, and may involve destructive practices in certain crop types, and becomes incrementally difficult with micro-greens such as basil and parsley. However, with the technological advancements in Computer Vision and AI techniques, automatically tracking crop growth in much greater detail is much more achievable. This study focuses on designing AI and vision enabled system to track the growth of sweet basil plants, a popular crop cultivated globally. We developed an algorithm leveraging the object segmentation model (Mask R-CNN) to detect and segment individual plants’ leaves from the images to compute their area to track growth. We further employed a series of techniques to overcome unforeseen challenges related to continuous growth tracking of the crop.


71 pages




Northern Illinois University

Rights Statement

In Copyright

Rights Statement 2

NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.

Media Type