Show simple item record

dc.contributor.advisorSong, Dezhen
dc.creatorXie, Shuangyu
dc.date.accessioned2022-02-23T18:02:43Z
dc.date.available2023-05-01T06:37:22Z
dc.date.created2021-05
dc.date.issued2021-03-24
dc.date.submittedMay 2021
dc.identifier.urihttps://hdl.handle.net/1969.1/195622
dc.description.abstractTo enable robotic weed control, we develop algorithms to detect nutsedge weed from Bermudagrass turf. Due to the similarity between the weed and the background turf, it is expensive and error-prone to perform manual data labeling. Consequently, directly applying deep learning methods for object detection cannot generate satisfactory results. Building on an instance detection approach, (i.e. Mask R-CNN), we combine synthetic data with raw data to train the network. We propose an algorithm to generate high fidelity synthetic data, adopting different levels of annotations to reduce labeling cost. Moreover, we construct a nutsedge skeleton-based probabilistic map (NSPM) as the neural network input to reduce the reliance on pixel-wise precise labeling. We also modify loss function from cross entropy to Kullback–Leibler divergence which accommodates uncertainty in the labeling process. We have implemented the proposed algorithm and compare it with Faster R-CNN, a typical object detection approach. The results show that our design can effectively reduce the impact of imprecise and insufficient training sample issues and significantly outperforms the counterpart with a false negative rate of 0.4%, a satisfying result for weed control applications.en
dc.format.mimetypeapplication/pdf
dc.language.isoen
dc.subjectWeed detectionen
dc.subjectdeep Learningen
dc.subjectrobotic weed controlen
dc.subjectprecision agricultureen
dc.titleToward Robotic Weed Control: Detection of Nutsedge Weed in Bermudagrass Turf Using Inaccurate and Insufficient Training Dataen
dc.typeThesisen
thesis.degree.departmentComputer Science and Engineeringen
thesis.degree.disciplineComputer Engineeringen
thesis.degree.grantorTexas A&M Universityen
thesis.degree.nameMaster of Scienceen
thesis.degree.levelMastersen
dc.contributor.committeeMemberDamnjanovic, Ivan
dc.contributor.committeeMemberSharon, Guni
dc.type.materialtexten
dc.date.updated2022-02-23T18:02:44Z
local.embargo.terms2023-05-01
local.etdauthor.orcid0000-0002-6094-3874


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record