- You are here:
-
-
All publications
- Potato weeds detection system using deep learning models
All publications from your selection
Potato weeds detection system using deep learning models
Authors: Muhammad Naveed Tahir,
Qamar Zaman,
Shoaib Rashid Saleem,
Faiza Khan,
Muhammad Aqib,
Noureen Zafar
Identifier: CSBE23131
Download file:
Published in: CSBE-SCGAB Technical Conferences » AGM Lethbridge 2023
Download RAW file:
DOI:
Description: Artificial Intelligence is being used in the digital revolution to change agriculture which leads to an increase in yield, lower costs, and reduced environmental impact through improved management and decision-making processes. AI-based spot-specific technologies are developed that not only save crops, the environment, and human health from the adverse effects of agrochemicals but also save resources in terms of time and money. For these spraying technologies, deep learning-based object recognition and classification approaches are being used nowadays. In this work, two deep learning-based models were used that are based on a Convolutional Neural Network (CNN) named You Only Look Once (YOLO). A dataset of five potato weeds were acquired to train these models. The crop was grown in the Potohar region, and the fields were irrigated using three different irrigation setups including controlled, rain gun, and drip irrigation systems. Images were collected under different climatic conditions such as cloudy, partly cloudy, sunny, and at contrasting times of the day weekly. For this purpose, YOLOv4 and tiny-YOLOv4 models were used which gave 69.45% and 49.53% accuracies, respectively. The model with high detection accuracy is deployed into a spot-specific spraying system to apply weedicide on potato crops.
Keywords: Deep Learning, weeds, Potato, CNN, Yolo
Résumé:
Mots-clés:
Citation:
Volume:
Issue:
Pages -
Contributor:
Date: 2023-07-23
Technical field:
Conference name: CSBE/SCGAB 2023 Annual Conference, Lethbridge, Alberta, 23-26 July 2023.
Session name:
Other information:
Type: Presentation
Format:
Publication type: Text.Abstract
Source:
Relation:
Coverage: Asia
Language 1: en
Language 2:
Rights: Canadian Society for Bioengineering
Notes:
Other files:
DOI:
Description: Artificial Intelligence is being used in the digital revolution to change agriculture which leads to an increase in yield, lower costs, and reduced environmental impact through improved management and decision-making processes. AI-based spot-specific technologies are developed that not only save crops, the environment, and human health from the adverse effects of agrochemicals but also save resources in terms of time and money. For these spraying technologies, deep learning-based object recognition and classification approaches are being used nowadays. In this work, two deep learning-based models were used that are based on a Convolutional Neural Network (CNN) named You Only Look Once (YOLO). A dataset of five potato weeds were acquired to train these models. The crop was grown in the Potohar region, and the fields were irrigated using three different irrigation setups including controlled, rain gun, and drip irrigation systems. Images were collected under different climatic conditions such as cloudy, partly cloudy, sunny, and at contrasting times of the day weekly. For this purpose, YOLOv4 and tiny-YOLOv4 models were used which gave 69.45% and 49.53% accuracies, respectively. The model with high detection accuracy is deployed into a spot-specific spraying system to apply weedicide on potato crops.
Keywords: Deep Learning, weeds, Potato, CNN, Yolo
Résumé:
Mots-clés:
Citation:
Volume:
Issue:
Pages -
Contributor:
Date: 2023-07-23
Technical field:
Conference name: CSBE/SCGAB 2023 Annual Conference, Lethbridge, Alberta, 23-26 July 2023.
Session name:
Other information:
Type: Presentation
Format:
Publication type: Text.Abstract
Source:
Relation:
Coverage: Asia
Language 1: en
Language 2:
Rights: Canadian Society for Bioengineering
Notes:
Other files:
©
2026
CSBE-SCGAB