2021
DOI: 10.1109/access.2021.3056577
|View full text |Cite
|
Sign up to set email alerts
|

Identification of Tobacco Crop Based on Machine Learning for a Precision Agricultural Sprayer

Abstract: Agrochemicals, which are very efficacious in protecting crops, also cause environmental pollution and pose serious threats to farmers' health upon exposure. In order to cut down the environmental and human health risks associated with agrochemical application, there is a need to develop intelligent application equipment that could detect and recognize crops/weeds, and spray precise doses of agrochemical at the right place and right time. This paper presents a machine-learning based crop/weed detection system f… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1

Citation Types

0
19
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
7
1

Relationship

0
8

Authors

Journals

citations
Cited by 38 publications
(19 citation statements)
references
References 46 publications
0
19
0
Order By: Relevance
“…e excavation of digital agriculture has brought convenience to the concept and technology of digital multimedia crop cultivation and has brought new technologies and technical information and management services in new directions to the development of digital multimedia crops, making digital multimedia crop production gradually digitized [1][2][3]. erefore, the basic concepts, core technologies, and development prospects of digital multimedia crop cultivation need to be understood and mastered in detail, which is conducive to the progress and development of the subject, is more beneficial to the theoretical concepts and technical systems, and increases the competitiveness and sustainability of digital multimedia crop cultivation, therefore significantly improving the management and benefits of agriculture [4][5][6].…”
Section: Introductionmentioning
confidence: 99%
“…e excavation of digital agriculture has brought convenience to the concept and technology of digital multimedia crop cultivation and has brought new technologies and technical information and management services in new directions to the development of digital multimedia crops, making digital multimedia crop production gradually digitized [1][2][3]. erefore, the basic concepts, core technologies, and development prospects of digital multimedia crop cultivation need to be understood and mastered in detail, which is conducive to the progress and development of the subject, is more beneficial to the theoretical concepts and technical systems, and increases the competitiveness and sustainability of digital multimedia crop cultivation, therefore significantly improving the management and benefits of agriculture [4][5][6].…”
Section: Introductionmentioning
confidence: 99%
“…The combination of defined and adaptive thresholds in ExG, NExG and ExHSV was used in this study to better determine weed presence where weeds are infrequent. It is highly unlikely that framerate is a limiting factor for OWL, with real-time operation in large-scale systems (forward speed dependent) observed at framerates above 6 FPS for other systems 29 31 , 53 , 63 , with current commercial systems operating between 16 and 17 FPS for forward speeds between 2.67 and 6.67 m s −1 64 , 65 . The performance of algorithms in the field is likely also dependent on the ambient lighting conditions, which in turn influence the blurriness of the video feed.…”
Section: Discussionmentioning
confidence: 99%
“…A number of studies have developed weed detection algorithms based on specific plant features such as colour 18 , 21 , 22 , shape 23 , 24 , texture 25 , 26 or a combination of these features 27 29 . Importantly, non-machine learning algorithms typically have lower computational requirements and perform faster on less powerful processors, such as the Raspberry Pi, improving the likelihood of real-time operation in large-scale cropping systems 30 , 31 . Critically, the computational requirements of the algorithm determine the framerate and hence the real-time capability of the device.…”
Section: Introductionmentioning
confidence: 99%
“…A number of studies have developed weed detection algorithms based on specific plant features such as colour (Burks et al, 2000;Esau et al, 2018;Woebbecke et al, 1995a), shape (Lee et al, 1999;Woebbecke et al, 1995b), texture (Chang et al, 2012;Tian and Reid, 1999) or a combination of these features (Burgos-Artizzu et al, 2011;Golzarian and Frick, 2011;Kavdir, 2004). Importantly, non-machine learning algorithms typically have lower computational requirements and perform faster on less powerful processors, such as the Raspberry Pi, improving the likelihood of real-time operation in large-scale cropping systems (Chechliński et al, 2019;Tufail et al, 2021). Critically, the computational requirements of the algorithm determine the framerate and hence the real-time capability of the device.…”
mentioning
confidence: 99%
“…The combination of defined and adaptive thresholds in ExG, NExG and ExHSV was used in this study to better determine weed presence where weeds are infrequent. It is highly unlikely that framerate is a limiting factor for OWL, with real-time operation in large-scale systems (forward speed dependent) observed at framerates above 6 FPS for other systems (Burgos-Artizzu et al, 2011;Chechliński et al, 2019;Milioto et al, 2018;Olsen et al, 2019;Tufail et al, 2021), with current commercial systems operating at 17 FPS for forward speeds above 6.67 m s -1 (Martin, 2021). The performance of algorithms in the field is likely also dependent on the ambient lighting conditions, which in turn influence the blurriness of the video feed.…”
mentioning
confidence: 99%