Title
Deep Convolutional Neural Networks for Weed Detection in Agricultural Crops Using Optical Aerial Images
Date Issued
01 March 2020
Access level
metadata only access
Resource Type
conference paper
Author(s)
Pontifícia Universidade Católica do Rio de Janeiro
Publisher(s)
Institute of Electrical and Electronics Engineers Inc.
Abstract
The presence of weeds in agricultural crops has been one of the problems of greatest interest in recent years as they consume natural resources and negatively affect the agricultural process. For this purpose, a model has been implemented to segment weed in aerial images. The proposed model relies on DeepLabv3 architecture trained upon patches extracted from high-resolution aerial imagery. The dataset employed consisted in 5 high-resolution images that describes a sugar beet agricultural field in Germany. SegNet and U-Net architectures were selected for comparison purposes. Our results demonstrate that balancing of data, together with a greater spatial context leads better results with DeepLabv3 achieving up to 0.89 and 0.81 in terms of AUC and F1-score, respectively.
Start page
133
End page
137
Language
English
OCDE Knowledge area
Agricultura
Ciencias de la computación
Subjects
Scopus EID
2-s2.0-85091635881
Resource of which it is part
2020 IEEE Latin American GRSS and ISPRS Remote Sensing Conference, LAGIRS 2020 - Proceedings
ISBN of the container
978-172814350-7
Conference
2020 IEEE Latin American GRSS and ISPRS Remote Sensing Conference, LAGIRS 2020 - Proceedings
Sponsor(s)
The authors would like to acknowledge Intel Semiconductors Brasil for recognizing ICA Lab at PUC-Rio University as Intel AI Innovation Center and for supplying a Sky Lake, 96 Intel(R) Xeon(R) Platinum 8169 processors.
Sources of information:
Directorio de Producción Científica
Scopus