A
Alex Olsen
Researcher at James Cook University
Publications - 5
Citations - 269
Alex Olsen is an academic researcher from James Cook University. The author has contributed to research in topics: Weed & Deep learning. The author has an hindex of 3, co-authored 5 publications receiving 129 citations.
Papers
More filters
Journal ArticleDOI
DeepWeeds: A Multiclass Weed Species Image Dataset for Deep Learning.
Alex Olsen,Dmitry A. Konovalov,Bronson Philippa,Peter V. Ridd,Jake C. Wood,Jamie Johns,Wesley Banks,Benjamin Girgenti,O.P. Kenny,James Whinney,Brendan Calvert,Mostafa Rahimi Azghadi,Ronald D. White +12 more
TL;DR: The DeepWeeds dataset as mentioned in this paper consists of 17,509 labelled images of eight nationally significant weed species native to eight locations across northern Australia and achieved an average classification accuracy of 95.1% and 95.7%, respectively.
Journal ArticleDOI
Low-Power and High-Speed Deep FPGA Inference Engines for Weed Classification at the Edge
TL;DR: GPU- and FPGA-accelerated deterministically binarized DNNs, tailored toward weed species classification for robotic weed control are introduced, a significant step toward enabling deep inference and learning on IoT edge devices, and smart portable machines such as agricultural robots, which is the target application.
Proceedings ArticleDOI
In Situ Leaf Classification Using Histograms of Oriented Gradients
TL;DR: A novel method for segmenting leaves from a textured background is presented and a scale and rotation invariant enhancement of the HOG feature set for texture based leaf classification is investigated - whose results compare well with a multi-feature probabilistic neural network classifier on a benchmark data set.
Journal ArticleDOI
Robotic Spot Spraying of Harrisia Cactus (Harrisia martinii) in Grazing Pastures of the Australian Rangelands.
TL;DR: In this article, the authors used the MobileNetV2 deep learning architecture to perform real time spot spraying of harrisia cactus with 97.2% average recall accuracy and weed knockdown efficacy of up to 96%.
Journal ArticleDOI
DeepWeeds: A Multiclass Weed Species Image Dataset for Deep Learning
Alex Olsen,Dmitry A. Konovalov,Bronson Philippa,Peter V. Ridd,Jake C. Wood,Jamie Johns,Wesley Banks,Benjamin Girgenti,O.P. Kenny,James Whinney,Brendan Calvert,Mostafa Rahimi Azghadi,Ronald D. White +12 more
TL;DR: The DeepWeeds dataset as mentioned in this paper consists of 17,509 labelled images of eight nationally significant weed species native to eight locations across northern Australia and achieved an average classification accuracy of 95.1% and 95.7%, respectively.