Raspberry Pi as Visual Sensor Nodes in Precision Agriculture

A Study

Radhika Kamath, Mamatha Balachandra, Srikanth Prabhu

Research output: Contribution to journalArticle

Abstract

Wireless sensor network applications in the agricultural sector are gaining popularity with the advancement of the Internet of Things technology. Predominantly, wireless sensor networks are used in agriculture to sense the important agricultural field parameters, such as temperature, humidity, soil moisture level, nitrite content in the soil, groundwater quality, and so on. These sensed parameters will be sent to a remote station, where it will be processed and analyzed to build a decision support system. This paper describes the implementation of a wireless visual sensor network for precision agriculture to monitor paddy crop for weeds using Raspberry Pi. Bluetooth 4.0 was used by visual sensor nodes to send the data to the base station. Base station forwarded the data to the remote station using IEEE 802.11 a/b/g/n standard. The solar cell battery was used to power up the sensor nodes and the base station. At the remote station, images were preprocessed to remove soil background and different shape features were extracted. Random forest and support vector machine classifiers were used to classify the paddy crop and weed based on the shape features. The results and observations obtained from the experimental setup of the system in a small paddy field are also reported. This system could be expected to enhance the crop production by giving timely advice to the crop producers about the presence of weeds so that steps can be taken to eradicate weeds.

Original languageEnglish
Article number8684829
Pages (from-to)45110-45122
Number of pages13
JournalIEEE Access
Volume7
DOIs
Publication statusPublished - 01-01-2019

Fingerprint

Sensor nodes
Agriculture
Crops
Base stations
Wireless sensor networks
Soils
Bluetooth
Soil moisture
Nitrites
Decision support systems
Sensor networks
Support vector machines
Groundwater
Atmospheric humidity
Solar cells
Classifiers
Temperature

All Science Journal Classification (ASJC) codes

  • Computer Science(all)
  • Materials Science(all)
  • Engineering(all)

Cite this

Kamath, Radhika ; Balachandra, Mamatha ; Prabhu, Srikanth. / Raspberry Pi as Visual Sensor Nodes in Precision Agriculture : A Study. In: IEEE Access. 2019 ; Vol. 7. pp. 45110-45122.
@article{9f1a884310ea419e8f99dcfa6cc88144,
title = "Raspberry Pi as Visual Sensor Nodes in Precision Agriculture: A Study",
abstract = "Wireless sensor network applications in the agricultural sector are gaining popularity with the advancement of the Internet of Things technology. Predominantly, wireless sensor networks are used in agriculture to sense the important agricultural field parameters, such as temperature, humidity, soil moisture level, nitrite content in the soil, groundwater quality, and so on. These sensed parameters will be sent to a remote station, where it will be processed and analyzed to build a decision support system. This paper describes the implementation of a wireless visual sensor network for precision agriculture to monitor paddy crop for weeds using Raspberry Pi. Bluetooth 4.0 was used by visual sensor nodes to send the data to the base station. Base station forwarded the data to the remote station using IEEE 802.11 a/b/g/n standard. The solar cell battery was used to power up the sensor nodes and the base station. At the remote station, images were preprocessed to remove soil background and different shape features were extracted. Random forest and support vector machine classifiers were used to classify the paddy crop and weed based on the shape features. The results and observations obtained from the experimental setup of the system in a small paddy field are also reported. This system could be expected to enhance the crop production by giving timely advice to the crop producers about the presence of weeds so that steps can be taken to eradicate weeds.",
author = "Radhika Kamath and Mamatha Balachandra and Srikanth Prabhu",
year = "2019",
month = "1",
day = "1",
doi = "10.1109/ACCESS.2019.2908846",
language = "English",
volume = "7",
pages = "45110--45122",
journal = "IEEE Access",
issn = "2169-3536",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

Kamath, R, Balachandra, M & Prabhu, S 2019, 'Raspberry Pi as Visual Sensor Nodes in Precision Agriculture: A Study', IEEE Access, vol. 7, 8684829, pp. 45110-45122. https://doi.org/10.1109/ACCESS.2019.2908846

Raspberry Pi as Visual Sensor Nodes in Precision Agriculture : A Study. / Kamath, Radhika; Balachandra, Mamatha; Prabhu, Srikanth.

In: IEEE Access, Vol. 7, 8684829, 01.01.2019, p. 45110-45122.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Raspberry Pi as Visual Sensor Nodes in Precision Agriculture

T2 - A Study

AU - Kamath, Radhika

AU - Balachandra, Mamatha

AU - Prabhu, Srikanth

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Wireless sensor network applications in the agricultural sector are gaining popularity with the advancement of the Internet of Things technology. Predominantly, wireless sensor networks are used in agriculture to sense the important agricultural field parameters, such as temperature, humidity, soil moisture level, nitrite content in the soil, groundwater quality, and so on. These sensed parameters will be sent to a remote station, where it will be processed and analyzed to build a decision support system. This paper describes the implementation of a wireless visual sensor network for precision agriculture to monitor paddy crop for weeds using Raspberry Pi. Bluetooth 4.0 was used by visual sensor nodes to send the data to the base station. Base station forwarded the data to the remote station using IEEE 802.11 a/b/g/n standard. The solar cell battery was used to power up the sensor nodes and the base station. At the remote station, images were preprocessed to remove soil background and different shape features were extracted. Random forest and support vector machine classifiers were used to classify the paddy crop and weed based on the shape features. The results and observations obtained from the experimental setup of the system in a small paddy field are also reported. This system could be expected to enhance the crop production by giving timely advice to the crop producers about the presence of weeds so that steps can be taken to eradicate weeds.

AB - Wireless sensor network applications in the agricultural sector are gaining popularity with the advancement of the Internet of Things technology. Predominantly, wireless sensor networks are used in agriculture to sense the important agricultural field parameters, such as temperature, humidity, soil moisture level, nitrite content in the soil, groundwater quality, and so on. These sensed parameters will be sent to a remote station, where it will be processed and analyzed to build a decision support system. This paper describes the implementation of a wireless visual sensor network for precision agriculture to monitor paddy crop for weeds using Raspberry Pi. Bluetooth 4.0 was used by visual sensor nodes to send the data to the base station. Base station forwarded the data to the remote station using IEEE 802.11 a/b/g/n standard. The solar cell battery was used to power up the sensor nodes and the base station. At the remote station, images were preprocessed to remove soil background and different shape features were extracted. Random forest and support vector machine classifiers were used to classify the paddy crop and weed based on the shape features. The results and observations obtained from the experimental setup of the system in a small paddy field are also reported. This system could be expected to enhance the crop production by giving timely advice to the crop producers about the presence of weeds so that steps can be taken to eradicate weeds.

UR - http://www.scopus.com/inward/record.url?scp=85064634879&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85064634879&partnerID=8YFLogxK

U2 - 10.1109/ACCESS.2019.2908846

DO - 10.1109/ACCESS.2019.2908846

M3 - Article

VL - 7

SP - 45110

EP - 45122

JO - IEEE Access

JF - IEEE Access

SN - 2169-3536

M1 - 8684829

ER -