Improving the accuracy of weed species detection for robotic weed control in complex real-time environments

Olsen, Alex (2020) Improving the accuracy of weed species detection for robotic weed control in complex real-time environments. PhD thesis, James Cook University.

[img]
Preview
PDF (Thesis)
Download (23MB) | Preview
View at Publisher Website: https://doi.org/10.25903/vhbh-w150
 
202


Abstract

Alex Olsen applied deep learning and machine vision to improve the accuracy of weed species detection in real time complex environments. His robotic weed control prototype, AutoWeed, presents a new efficient tool for weed management in crop and pasture and has launched a startup agricultural technology company.

Item ID: 65379
Item Type: Thesis (PhD)
Keywords: deep learning, machine vision, weed species detection, robotic weed control, AutoWeed, weed management, agricultural technology, Lantana camara (lantana), nationally significant weed species
Related URLs:
Copyright Information: © 2020 Alex Olsen.
Additional Information:

Four publications arising from this thesis are stored in ResearchOnline@JCU, at the time of processing. Please see the Related URLs. The publications are:

Chapter 2: Olsen, Alex, Han, Sunghyu, Calvert, Brendan, Ridd, Peter, and Kenny, Owen (2015) In situ leaf classification using histograms of oriented gradients. In: Proceedings of the lnternational Conference on Digital lmage Computing: techniques and applications. pp. 441-448. From: DICTA 2015: lnternational Conference on Digital lmage Computing: techniques and applications, 23-25 Nov 2015, Adelaide, SA, Australia.

Chapters 3 and 4: Olsen, Alex, Konovalov, Dmitry A., Philippa, Bronson, Ridd, Peter, Wood, Jake, Johns, Jamie, Banks, Wesley, Girgenti, Benjamin, Kenny, Owen, Whinney, James, Calvert, Brendan, Rahimi Azghadi, Mostafa, and White, Ronald D. (2019) DeepWeeds: a multiclass weed species image dataset for deep learning. Scientific Reports, 9. 2058.

Chapter 4: Lammie, Corey, Olsen, Alex, Carrick, Tony, and Rahimi Azghadi, Mostafa (2019) Low-power and high-speed deep FPGA inference engines for weed classification at the edge. IEEE Access, 7. pp. 51171-51184.

Chapters 4 and 5: Calvert, Brendan, Olsen, Alex, Philippa, Bronson, and Rahimi Azghadi, Mostafa (2019) Autoweed: detecting Harrisia cactus in the Goondiwindi region for selective spot-spraying. In: https://www.wsq.org.au/downloads/2019-pest-animal-and-weed-symposium-proceedings/. pp. 52-57. From: PAWS 2019: 1st Queensland Pest Animal and Weed Symposium, 20-23 May 2019, Gold Coast, QLD, Australia.

Date Deposited: 21 Dec 2020 00:40
FoR Codes: 07 AGRICULTURAL AND VETERINARY SCIENCES > 0701 Agriculture, Land and Farm Management > 070107 Farming Systems Research @ 35%
09 ENGINEERING > 0906 Electrical and Electronic Engineering > 090602 Control Systems, Robotics and Automation @ 35%
08 INFORMATION AND COMPUTING SCIENCES > 0801 Artificial Intelligence and Image Processing > 080104 Computer Vision @ 30%
SEO Codes: 86 MANUFACTURING > 8614 Machinery and Equipment > 861401 Agricultural Machinery and Equipment @ 100%
Downloads: Total: 202
Last 12 Months: 68
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page