Weed Growth Stage Estimator Using Deep Convolutional Neural Networks

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

Standard

Weed Growth Stage Estimator Using Deep Convolutional Neural Networks. / Teimouri, Nima; Dyrmann, Mads; Nielsen, Per Rydahl; Mathiassen, Solvejg Kopp; Somerville, Gayle J.; Jørgensen, Rasmus Nyholm.

In: Sensors, Vol. 18, No. 5, 1580, 16.05.2018.

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

Harvard

APA

CBE

MLA

Vancouver

Author

Bibtex

@article{6778ad94e4004939aeca4da4312f7604,
title = "Weed Growth Stage Estimator Using Deep Convolutional Neural Networks",
abstract = "This study outlines a new method of automatically estimating weed species and growth stages (from cotyledon until eight leaves are visible) of in situ images covering 18 weed species or families. Images of weeds growing within a variety of crops were gathered across variable environmental conditions with regards to soil types, resolution and light settings. Then, 9649 of these images were used for training the computer, which automatically divided the weeds into nine growth classes. The performance of this proposed convolutional neural network approach was evaluated on a further set of 2516 images, which also varied in term of crop, soil type, image resolution and light conditions. The overall performance of this approach achieved a maximum accuracy of 78{\%} for identifying Polygonum spp. and a minimum accuracy of 46{\%} for blackgrass. In addition, it achieved an average 70{\%} accuracy rate in estimating the number of leaves and 96{\%} accuracy when accepting a deviation of two leaves. These results show that this new method of using deep convolutional neural networks has a relatively high ability to estimate early growth stages across a wide variety of weed species.",
author = "Nima Teimouri and Mads Dyrmann and Nielsen, {Per Rydahl} and Mathiassen, {Solvejg Kopp} and Somerville, {Gayle J.} and J{\o}rgensen, {Rasmus Nyholm}",
year = "2018",
month = "5",
day = "16",
doi = "10.3390/s18051580",
language = "English",
volume = "18",
journal = "Sensors",
issn = "1424-8220",
publisher = "M D P I AG",
number = "5",

}

RIS

TY - JOUR

T1 - Weed Growth Stage Estimator Using Deep Convolutional Neural Networks

AU - Teimouri, Nima

AU - Dyrmann, Mads

AU - Nielsen, Per Rydahl

AU - Mathiassen, Solvejg Kopp

AU - Somerville, Gayle J.

AU - Jørgensen, Rasmus Nyholm

PY - 2018/5/16

Y1 - 2018/5/16

N2 - This study outlines a new method of automatically estimating weed species and growth stages (from cotyledon until eight leaves are visible) of in situ images covering 18 weed species or families. Images of weeds growing within a variety of crops were gathered across variable environmental conditions with regards to soil types, resolution and light settings. Then, 9649 of these images were used for training the computer, which automatically divided the weeds into nine growth classes. The performance of this proposed convolutional neural network approach was evaluated on a further set of 2516 images, which also varied in term of crop, soil type, image resolution and light conditions. The overall performance of this approach achieved a maximum accuracy of 78% for identifying Polygonum spp. and a minimum accuracy of 46% for blackgrass. In addition, it achieved an average 70% accuracy rate in estimating the number of leaves and 96% accuracy when accepting a deviation of two leaves. These results show that this new method of using deep convolutional neural networks has a relatively high ability to estimate early growth stages across a wide variety of weed species.

AB - This study outlines a new method of automatically estimating weed species and growth stages (from cotyledon until eight leaves are visible) of in situ images covering 18 weed species or families. Images of weeds growing within a variety of crops were gathered across variable environmental conditions with regards to soil types, resolution and light settings. Then, 9649 of these images were used for training the computer, which automatically divided the weeds into nine growth classes. The performance of this proposed convolutional neural network approach was evaluated on a further set of 2516 images, which also varied in term of crop, soil type, image resolution and light conditions. The overall performance of this approach achieved a maximum accuracy of 78% for identifying Polygonum spp. and a minimum accuracy of 46% for blackgrass. In addition, it achieved an average 70% accuracy rate in estimating the number of leaves and 96% accuracy when accepting a deviation of two leaves. These results show that this new method of using deep convolutional neural networks has a relatively high ability to estimate early growth stages across a wide variety of weed species.

U2 - 10.3390/s18051580

DO - 10.3390/s18051580

M3 - Journal article

VL - 18

JO - Sensors

JF - Sensors

SN - 1424-8220

IS - 5

M1 - 1580

ER -