DeepNEM: Deep Network Energy-Minimization for Agricultural Field Segmentation
One of the main characteristics of agricultural fields is that the appearance of different crops and their growth status, in an aerial image, is varied, and has a wide range of radiometric values and high level of variability. The extraction of these fields and their monitoring are activities that r...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2020-01-01
|
Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/8986665/ |
id |
doaj-ba6ac74fb99748c5924975e7e708b8c1 |
---|---|
record_format |
Article |
spelling |
doaj-ba6ac74fb99748c5924975e7e708b8c12021-06-03T23:02:32ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352020-01-011372673710.1109/JSTARS.2020.29710618986665DeepNEM: Deep Network Energy-Minimization for Agricultural Field SegmentationMargarita Torre0https://orcid.org/0000-0002-9122-2949Beatriz Remeseiro1https://orcid.org/0000-0001-9265-253XPetia Radeva2https://orcid.org/0000-0003-0047-5172Fernando Martinez3https://orcid.org/0000-0003-3159-2494Department of Computer Science, Universitat Autònoma de Barcelona, Bellaterra, SpainDepartment of Computer Science, Universidad de Oviedo, Gijón, SpainDepartament de Matemàtiques i Informàtica, Universitat de Barcelona, Barcelona, SpainDepartament de Matemàtiques, Universitat Politècnica de Catalunya, Barcelona, SpainOne of the main characteristics of agricultural fields is that the appearance of different crops and their growth status, in an aerial image, is varied, and has a wide range of radiometric values and high level of variability. The extraction of these fields and their monitoring are activities that require a high level of human intervention. In this article, we propose a novel automatic algorithm, named deep network energy-minimization (DeepNEM), to extract agricultural fields in aerial images. The model-guided process selects the most relevant image clues extracted by a deep network, completes them and finally generates regions that represent the agricultural fields under a minimization scheme. DeepNEM has been tested over a broad range of fields in terms of size, shape, and content. Different measures were used to compare the DeepNEM with other methods, and to prove that it represents an improved approach to achieve a high-quality segmentation of agricultural fields. Furthermore, this article also presents a new public dataset composed of 1200 images with their parcels boundaries annotations.https://ieeexplore.ieee.org/document/8986665/Agricultural fieldsimage edge analysisimage segmentationregion extraction |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Margarita Torre Beatriz Remeseiro Petia Radeva Fernando Martinez |
spellingShingle |
Margarita Torre Beatriz Remeseiro Petia Radeva Fernando Martinez DeepNEM: Deep Network Energy-Minimization for Agricultural Field Segmentation IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Agricultural fields image edge analysis image segmentation region extraction |
author_facet |
Margarita Torre Beatriz Remeseiro Petia Radeva Fernando Martinez |
author_sort |
Margarita Torre |
title |
DeepNEM: Deep Network Energy-Minimization for Agricultural Field Segmentation |
title_short |
DeepNEM: Deep Network Energy-Minimization for Agricultural Field Segmentation |
title_full |
DeepNEM: Deep Network Energy-Minimization for Agricultural Field Segmentation |
title_fullStr |
DeepNEM: Deep Network Energy-Minimization for Agricultural Field Segmentation |
title_full_unstemmed |
DeepNEM: Deep Network Energy-Minimization for Agricultural Field Segmentation |
title_sort |
deepnem: deep network energy-minimization for agricultural field segmentation |
publisher |
IEEE |
series |
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
issn |
2151-1535 |
publishDate |
2020-01-01 |
description |
One of the main characteristics of agricultural fields is that the appearance of different crops and their growth status, in an aerial image, is varied, and has a wide range of radiometric values and high level of variability. The extraction of these fields and their monitoring are activities that require a high level of human intervention. In this article, we propose a novel automatic algorithm, named deep network energy-minimization (DeepNEM), to extract agricultural fields in aerial images. The model-guided process selects the most relevant image clues extracted by a deep network, completes them and finally generates regions that represent the agricultural fields under a minimization scheme. DeepNEM has been tested over a broad range of fields in terms of size, shape, and content. Different measures were used to compare the DeepNEM with other methods, and to prove that it represents an improved approach to achieve a high-quality segmentation of agricultural fields. Furthermore, this article also presents a new public dataset composed of 1200 images with their parcels boundaries annotations. |
topic |
Agricultural fields image edge analysis image segmentation region extraction |
url |
https://ieeexplore.ieee.org/document/8986665/ |
work_keys_str_mv |
AT margaritatorre deepnemdeepnetworkenergyminimizationforagriculturalfieldsegmentation AT beatrizremeseiro deepnemdeepnetworkenergyminimizationforagriculturalfieldsegmentation AT petiaradeva deepnemdeepnetworkenergyminimizationforagriculturalfieldsegmentation AT fernandomartinez deepnemdeepnetworkenergyminimizationforagriculturalfieldsegmentation |
_version_ |
1721398820538941440 |