Content uploaded by Lars Caspersen
Author content
All content in this area was uploaded by Lars Caspersen on Sep 24, 2019
Content may be subject to copyright.
9TH BRAZIL-GERMANY SYMPOSIUM ON SUSTAINABLE DEVELOPMENT
UNIVERSITY OF HOHENHEIM, STUTTGART, GERMANY
SEPTEMBER 15-17, 2019
Combining socio-economic and remote sensing data for food
insecurity prediction using neural networks
Caspersen, Lars1; da Rocha Lima Filho, Roberto Ivo2; Heisenberg, Gernot1,*; Wöhrle,
Sven1
1Technische Hochschule Köln, Köln, Germany
2Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil
*Corresponding author. E-mail: gernot.heisenberg@th-koeln.de
Keywords: food insecurity, socio-economic data, remote sensing data, convolutional
neural network, recurrent neural network
ABSTRACT
Aid organizations and governments are applying great effort in resolving the
negative impacts of food insecurity induced crisis like famines or mass migration. One of
the most limiting resources these actors face is the lack of preparation time for
consistent and sustainable planning for emergency relief like setting refugee camps or
securing supply with food and energy. Hence, increasing the lead time for preparation is
an essential step and will result in saving many lives. The aim of this research is to
increase the lead time by developing a machine learning based mathematical prediction
model that is able to compute the probability for food insecure areas by learning from
historical data.
For performing such computations, our prediction model is developed and trained on
historic open access data for the Horn of Africa (2009-2018). We used precipitation and
vegetation data derived by remote sensing, as well as socio-economic, medical, armed
conflict and disaster data. To overcome spatial inconsistencies in the input data and to
meet the requirements of spatially homogenous input for neural networks, all data has
been converted to geo-referenced raster maps. Disaster and armed conflict data has
been fitted to districts while local food market prices have been interpolated. The IPC
(Integrated Phase Classifier) has been used as the food security label.
In order to find a prediction model, new generation deep learning methods have
been used [1]. While most approaches generally focus on a single type of neural
network, we have decided to combine Convolutional Neural Networks (CNNs) and
Recurrent Neural Networks (RNNs) as proposed by Interdonato et al. [2]. This allows us
for combining both of their strengths such as the spatial autocorrelation of the CNNs, and
the ability to address for temporal dependencies in remote sensing data by RNNs.
Several analyses were applied on the collected data such as multicollinearity, cluster and
principal component analyses. Evaluation of our method was performed using cross-
validation (70/30 data split) and machine learning metrics (F1-Score, OA, MCC).
Our preliminary results have encouraged us to further investigate the detection of
food insecure areas using open access data.
[1] Zhang, L., Du, B., (2016). Deep learning for remote sensing data: a technical tutorial on the state of the
art. IEEE Geosci. Remote Sens. Mag. 4, 22–40, DOI: 10.1109/MGRS.2016.2540798.
[2] Interdonato et al. (2019), DuPLO: A DUal view Point deep Learning architecture for time series
classificatiOn, ISPRS Journal of Photogrammetry and Remote Sensing 149, 91-104,
DOI: 10.1016/j.isprsjprs.2019.01.011.