Scouting of whiteflies in tomato greenhouse environment using deep learning

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F70883521%3A28140%2F21%3A63541478" target="_blank" >RIV/70883521:28140/21:63541478 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/70883521:28140/22:63541478

Výsledek na webu

<a href="http://dx.doi.org/10.1007/978-981-16-3349-2_27" target="_blank" >http://dx.doi.org/10.1007/978-981-16-3349-2_27</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-981-16-3349-2_27" target="_blank" >10.1007/978-981-16-3349-2_27</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Scouting of whiteflies in tomato greenhouse environment using deep learning

Popis výsledku v původním jazyce

This study shows the possibilities of how to replace tedious human labor—scouting of yellow sticky traps (YST) for whiteflies—using artificial cognitive vision, specifically the deep convolutional network (CNN), as a part of the more complex system—BERABOT. The used CNN is the Faster R-CNN trained by deep transfer learning to substitute human scouting when the low whiteflies infection phase was specifically targeted. The training was conducted on pictures taken inside the heated and lighted tomato production greenhouse of “Bezdínek Farm” in Dolni Lutyne, Czechia. Used pictures were collected in a way planned for future fully automated robotic applications in the BERABOT system. The achieved results were compared with the scouting results of a professional phytopathologist. The trained employee’s scouting results against the professional phytopathologist accomplished root-mean-square error (RMSE) equal to 4.23, while the developed CNN model was evaluated to be 5.83. The results presented here open up new frontiers for further CNN model tuning leading to the potential in substituting an employee(s) in the future and make tomato production less expensive and less human labor dependent. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Název v anglickém jazyce

Scouting of whiteflies in tomato greenhouse environment using deep learning

Popis výsledku anglicky

This study shows the possibilities of how to replace tedious human labor—scouting of yellow sticky traps (YST) for whiteflies—using artificial cognitive vision, specifically the deep convolutional network (CNN), as a part of the more complex system—BERABOT. The used CNN is the Faster R-CNN trained by deep transfer learning to substitute human scouting when the low whiteflies infection phase was specifically targeted. The training was conducted on pictures taken inside the heated and lighted tomato production greenhouse of “Bezdínek Farm” in Dolni Lutyne, Czechia. Used pictures were collected in a way planned for future fully automated robotic applications in the BERABOT system. The achieved results were compared with the scouting results of a professional phytopathologist. The trained employee’s scouting results against the professional phytopathologist accomplished root-mean-square error (RMSE) equal to 4.23, while the developed CNN model was evaluated to be 5.83. The results presented here open up new frontiers for further CNN model tuning leading to the potential in substituting an employee(s) in the future and make tomato production less expensive and less human labor dependent. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/FW01010381" target="_blank" >FW01010381: Inteligentní robotická ochrana zdraví ekosystému hydroponického skleníku</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2021

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Smart Innovation, Systems and Technologies

ISBN

978-981163348-5

ISSN

21903018

e-ISSN

—

Počet stran výsledku

13

Strana od-do

323-335

Název nakladatele

Springer Science and Business Media Deutschland GmbH

Místo vydání

Berlín

Místo konání akce

St. Petersburg

Datum konání akce

7. 6. 2021

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—