Fig Plant Segmentation from Aerial Images Using a Deep Convolutional Encoder-Decoder Network

Crop segmentation is an important task in Precision Agriculture, where the use of aerial robots with an on-board camera has contributed to the development of new solution alternatives. We address the problem of fig plant segmentation in top-view RGB (Red-Green-Blue) images of a crop grown under open...

Full description

Bibliographic Details
Main Authors: J. Fuentes-Pacheco, J. Torres-Olivares, E. Roman-Rangel, S. Cervantes, P. Juarez-Lopez, J. Hermosillo-Valadez, J.M. Rendón-Mancha
Format: Article
Language:English
Published: MDPI AG 2019-05-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/11/10/1157
id doaj-a0dde86760214555ac321f4007223d9e
record_format Article
spelling doaj-a0dde86760214555ac321f4007223d9e2020-11-24T21:28:00ZengMDPI AGRemote Sensing2072-42922019-05-011110115710.3390/rs11101157rs11101157Fig Plant Segmentation from Aerial Images Using a Deep Convolutional Encoder-Decoder NetworkJ. Fuentes-Pacheco0J. Torres-Olivares1E. Roman-Rangel2S. Cervantes3P. Juarez-Lopez4J. Hermosillo-Valadez5J.M. Rendón-Mancha6CONACyT-Centro de Investigación en Ciencias, Instituto de Investigación en Ciencias Básicas y Aplicadas, Universidad Autónoma del Estado de Morelos, Cuernavaca, Morelos 62209, MexicoMaestría en Ciencias, Centro de Investigación en Ciencias, Instituto de Investigación en Ciencias Básicas y Aplicadas, Universidad Autónoma del Estado de Morelos, Cuernavaca, Morelos 62209, MexicoDigital Systems Department, Instituto Tecnologico Autonomo de Mexico, Mexico City 01080, MexicoDepartment of Computational Science and Engineering, Los Valles University Center of University of Guadalajara, Ameca, Jalisco 46600, MexicoFacultad de Ciencias Agropecuarias, Universidad Autónoma del Estado de Morelos, Cuernavaca, Morelos 62209, MexicoCentro de Investigación en Ciencias, Instituto de Investigación en Ciencias Básicas y Aplicadas, Universidad Autónoma del Estado de Morelos, Cuernavaca, Morelos 62209, MexicoCentro de Investigación en Ciencias, Instituto de Investigación en Ciencias Básicas y Aplicadas, Universidad Autónoma del Estado de Morelos, Cuernavaca, Morelos 62209, MexicoCrop segmentation is an important task in Precision Agriculture, where the use of aerial robots with an on-board camera has contributed to the development of new solution alternatives. We address the problem of fig plant segmentation in top-view RGB (Red-Green-Blue) images of a crop grown under open-field difficult circumstances of complex lighting conditions and non-ideal crop maintenance practices defined by local farmers. We present a Convolutional Neural Network (CNN) with an encoder-decoder architecture that classifies each pixel as crop or non-crop using only raw colour images as input. Our approach achieves a mean accuracy of 93.85% despite the complexity of the background and a highly variable visual appearance of the leaves. We make available our CNN code to the research community, as well as the aerial image data set and a hand-made ground truth segmentation with pixel precision to facilitate the comparison among different algorithms.https://www.mdpi.com/2072-4292/11/10/1157convolutional neural networkcrop segmentation<i>Ficus carica</i>unmanned aerial vehicles
collection DOAJ
language English
format Article
sources DOAJ
author J. Fuentes-Pacheco
J. Torres-Olivares
E. Roman-Rangel
S. Cervantes
P. Juarez-Lopez
J. Hermosillo-Valadez
J.M. Rendón-Mancha
spellingShingle J. Fuentes-Pacheco
J. Torres-Olivares
E. Roman-Rangel
S. Cervantes
P. Juarez-Lopez
J. Hermosillo-Valadez
J.M. Rendón-Mancha
Fig Plant Segmentation from Aerial Images Using a Deep Convolutional Encoder-Decoder Network
Remote Sensing
convolutional neural network
crop segmentation
<i>Ficus carica</i>
unmanned aerial vehicles
author_facet J. Fuentes-Pacheco
J. Torres-Olivares
E. Roman-Rangel
S. Cervantes
P. Juarez-Lopez
J. Hermosillo-Valadez
J.M. Rendón-Mancha
author_sort J. Fuentes-Pacheco
title Fig Plant Segmentation from Aerial Images Using a Deep Convolutional Encoder-Decoder Network
title_short Fig Plant Segmentation from Aerial Images Using a Deep Convolutional Encoder-Decoder Network
title_full Fig Plant Segmentation from Aerial Images Using a Deep Convolutional Encoder-Decoder Network
title_fullStr Fig Plant Segmentation from Aerial Images Using a Deep Convolutional Encoder-Decoder Network
title_full_unstemmed Fig Plant Segmentation from Aerial Images Using a Deep Convolutional Encoder-Decoder Network
title_sort fig plant segmentation from aerial images using a deep convolutional encoder-decoder network
publisher MDPI AG
series Remote Sensing
issn 2072-4292
publishDate 2019-05-01
description Crop segmentation is an important task in Precision Agriculture, where the use of aerial robots with an on-board camera has contributed to the development of new solution alternatives. We address the problem of fig plant segmentation in top-view RGB (Red-Green-Blue) images of a crop grown under open-field difficult circumstances of complex lighting conditions and non-ideal crop maintenance practices defined by local farmers. We present a Convolutional Neural Network (CNN) with an encoder-decoder architecture that classifies each pixel as crop or non-crop using only raw colour images as input. Our approach achieves a mean accuracy of 93.85% despite the complexity of the background and a highly variable visual appearance of the leaves. We make available our CNN code to the research community, as well as the aerial image data set and a hand-made ground truth segmentation with pixel precision to facilitate the comparison among different algorithms.
topic convolutional neural network
crop segmentation
<i>Ficus carica</i>
unmanned aerial vehicles
url https://www.mdpi.com/2072-4292/11/10/1157
work_keys_str_mv AT jfuentespacheco figplantsegmentationfromaerialimagesusingadeepconvolutionalencoderdecodernetwork
AT jtorresolivares figplantsegmentationfromaerialimagesusingadeepconvolutionalencoderdecodernetwork
AT eromanrangel figplantsegmentationfromaerialimagesusingadeepconvolutionalencoderdecodernetwork
AT scervantes figplantsegmentationfromaerialimagesusingadeepconvolutionalencoderdecodernetwork
AT pjuarezlopez figplantsegmentationfromaerialimagesusingadeepconvolutionalencoderdecodernetwork
AT jhermosillovaladez figplantsegmentationfromaerialimagesusingadeepconvolutionalencoderdecodernetwork
AT jmrendonmancha figplantsegmentationfromaerialimagesusingadeepconvolutionalencoderdecodernetwork
_version_ 1725972145368989696