- You are here:
-
-
All publications
- A Machine Vision Based In-Season Weed Detection System for Spot Application of Herbicides
All publications from your selection
A Machine Vision Based In-Season Weed Detection System for Spot Application of Herbicides
Authors: Ur Rehman, Tanzeel, Zaman, Qamar Uz, Schumann, Arnold, Farooque, Aitazaz Ahsan
Identifier: CSBE17158
Download file:
Published in: CSBE-SCGAB Technical Conferences » AGM Winnipeg 2017 (with CIGR VI Technical Symposium)
Download RAW file:
DOI:
Description: Low-bush blueberry (Vaccinium angustifolium ait.), often known as wild blueberry, is a perennial deciduous shrub that is native to Atlantic Canada, Quebec and northeastern United States. The improved management practices alleviated the overall crop productivity but the problem of uniform herbicide application still pertains, resulting into the increased cost of production and posing a sever threat to the local climatology. One of the major weed often found in wild blueberry fields is Goldenrod that expands due to rhizomes, subsequently resulting into dense weed patches in the different areas of fields. This patched growth of Goldenrod enables us to take an advantage of the spot or site specific application of herbicide, thereby reducing the overall cost of production and reducing the environmental hazards associated with agro-chemicals. A color co-occurrence matrix (CCM) based textural analysis algorithm was developed with an aim of providing electronic control of agrochemical application rates. The quantitative textural features from CCM were used to classify the goldenrod weed from wild blueberry fruit crop by developing the quadratic and linear discriminant relationships. A non-linear multifactor backpropagation artificial neural network (BP-ANN) based classifying model was also trained. The developed algorithm and the models were tested during the field evaluations by controlling the individual spraying nozzle through a multi-channel variable rate controller circuitry according to the requirement. The results of field evaluation indicated that CCM in combination with BP-ANN have ability to control the individual nozzle as per requirement and can help to minimize the cost of production.
Keywords: Variable rate sprayer, Color co-occurrence matrices, texture analysis, Back propagation neural network
Résumé:
Mots-clés:
Citation:
Volume:
Issue:
Pages -
Contributor:
Date: 2017-08-07
Technical field:
Conference name: CSBE/SCGAB 2017 Annual Conference, Canad Inns Polo Park, Winnipeg, Manitoba, 6-10 August 2017.
Session name: Poster Session
Other information:
Type: Text.Article
Format: PDF
Publication type: Technical conference
Source:
Relation:
Coverage: Canada
Language 1: en
Language 2:
Rights: Canadian Society for Bioengineering
Notes:
Other files:
DOI:
Description: Low-bush blueberry (Vaccinium angustifolium ait.), often known as wild blueberry, is a perennial deciduous shrub that is native to Atlantic Canada, Quebec and northeastern United States. The improved management practices alleviated the overall crop productivity but the problem of uniform herbicide application still pertains, resulting into the increased cost of production and posing a sever threat to the local climatology. One of the major weed often found in wild blueberry fields is Goldenrod that expands due to rhizomes, subsequently resulting into dense weed patches in the different areas of fields. This patched growth of Goldenrod enables us to take an advantage of the spot or site specific application of herbicide, thereby reducing the overall cost of production and reducing the environmental hazards associated with agro-chemicals. A color co-occurrence matrix (CCM) based textural analysis algorithm was developed with an aim of providing electronic control of agrochemical application rates. The quantitative textural features from CCM were used to classify the goldenrod weed from wild blueberry fruit crop by developing the quadratic and linear discriminant relationships. A non-linear multifactor backpropagation artificial neural network (BP-ANN) based classifying model was also trained. The developed algorithm and the models were tested during the field evaluations by controlling the individual spraying nozzle through a multi-channel variable rate controller circuitry according to the requirement. The results of field evaluation indicated that CCM in combination with BP-ANN have ability to control the individual nozzle as per requirement and can help to minimize the cost of production.
Keywords: Variable rate sprayer, Color co-occurrence matrices, texture analysis, Back propagation neural network
Résumé:
Mots-clés:
Citation:
Volume:
Issue:
Pages -
Contributor:
Date: 2017-08-07
Technical field:
Conference name: CSBE/SCGAB 2017 Annual Conference, Canad Inns Polo Park, Winnipeg, Manitoba, 6-10 August 2017.
Session name: Poster Session
Other information:
Type: Text.Article
Format: PDF
Publication type: Technical conference
Source:
Relation:
Coverage: Canada
Language 1: en
Language 2:
Rights: Canadian Society for Bioengineering
Notes:
Other files:
©
2025
CSBE-SCGAB