agriculture

Article
Winter Wheat Yield Prediction Using an LSTM Model from
MODIS LAI Products
Jian Wang 1 , Haiping Si 1 , Zhao Gao 2 and Lei Shi 1, *

1 College of Information and Management Science, Henan Agricultural University, Zhengzhou 450002, China
2 The First Geodetic Survey Team of the Ministry of Natural Resources, Shaanxi Bureau of Surveying,
Mapping and Geoinformation, Xi’an 710054, China
* Correspondence: [email protected]; Tel.: +86-0371-56990030

Abstract: Yield estimation using remote sensing data is a research priority in modern agriculture. The
rapid and accurate estimation of winter wheat yields over large areas is an important prerequisite
for food security policy formulation and implementation. In most county-level yield estimation
processes, multiple input data are used for yield prediction as much as possible, however, in some
regions, data are more difficult to obtain, so we used the single-leaf area index (LAI) as input data for
the model for yield prediction. In this study, the effects of different time steps as well as the LAI time
series on the estimation results were analyzed for the properties of long short-term memory (LSTM),
and multiple machine learning methods were compared with yield estimation models constructed by
the LSTM networks. The results show that the accuracy of the yield estimation results using LSTM
did not show an increasing trend with the increasing step size and data volume, while the yield
estimation results of the LSTM were generally better than those of conventional machine learning
methods, with the best R2 and RMSE results of 0.87 and 522.3 kg/ha, respectively, in the comparison
between predicted and actual yields. Although the use of LAI as a single input factor may cause
yield uncertainty in some extreme years, it is a reliable and promising method for improving the
yield estimation, which has important implications for crop yield forecasting, agricultural disaster
monitoring, food trade policy, and food security early warning.
Citation: Wang, J.; Si, H.; Gao, Z.; Shi,
L. Winter Wheat Yield Prediction
Keywords: winter wheat; yield estimation; LSTM; LAI; deep learning
Using an LSTM Model from MODIS
LAI Products. Agriculture 2022, 12,
1707. https://doi.org/10.3390/
agriculture12101707
1. Introduction
Academic Editor: Koki Homma
Wheat is an important crop in China, and its yield is directly related to the development
Received: 25 August 2022 of the national economy. Timely, accurate, and wide-ranging monitoring and forecasting of
Accepted: 12 October 2022 wheat yields is of great practical significance for national economic development and food
Published: 17 October 2022 policy formulation [1,2]. Due to its large coverage area and short detection period, satellite
Publisher’s Note: MDPI stays neutral
remote sensing provides a new technical tool for large-scale crop estimation and is rapidly
with regard to jurisdictional claims in
becoming the most widely used technology in crop estimation.
published maps and institutional affil- At present, the methods of crop yield estimation using remote sensing technology can
iations. be broadly classified into three categories according to the characteristics of the models used:
(1) the empirical modeling method; (2) the mechanistic modeling method; and (3) the semi-
empirical (semi-mechanistic) modeling method. The empirical model directly uses spectral
vegetation indices or canopy remote sensing inversion parameters to establish relationships
Copyright: © 2022 by the authors. with crop yields, which are characterized by their simplicity and ease, involving fewer
Licensee MDPI, Basel, Switzerland. crop yield formation mechanisms, and the relationships are generally established using
This article is an open access article conventional machine learning methods, such as support vector machine and random
distributed under the terms and forest, with NDVI or leaf area index (LAI) as input parameters [3–8]. Such relationships
conditions of the Creative Commons are usually localized and difficult to generalize to other agricultural areas. Semi-empirical
Attribution (CC BY) license (https://
models and semi-mechanical models are also known as parametric models, among which
creativecommons.org/licenses/by/
the light energy utilization efficiency model is the most widely used [6,9–11], but some
4.0/).

Agriculture 2022, 12, 1707. https://doi.org/10.3390/agriculture12101707 https://www.mdpi.com/journal/agriculture

Agriculture 2022, 12, 1707 2 of 13

parameters are difficult to quantitatively simulate. Mechanical models fully consider the
mechanism of crop yield formation, but their solution process is complex and requires
more input parameters, and some necessary parameters in the operation are difficult to
obtain at a regional scale; meanwhile, most of these models estimate crop yields at the
field scale [12–15]. Although the models work well for archiving yields, the accuracy may
be reduced when scaled to the national level. Many a priori parameters are required in
regional estimation. Due to the heterogeneity of the ground surface, the accuracy of the
ground parameters is generally low, especially in the case of small farmland in China,
resulting in low regional accuracy. Moreover, the computational process is complicated
and requires many parameters, which can be limited in practical use.
In recent years, deep learning has been successfully applied to several fields, such as
image recognition and language translation [11,16–22]. Compared with traditional machine
learning methods, deep learning techniques often achieve better performance. CNN and
recurrent neural network (RNN) are more widely used models in neural networks and
have also been applied to crop yield estimation and prediction [19,23–29]. LSTM is a special
kind of RNN [30,31], due to its recursive structure and gating mechanism that regulates
the entry and exit of information into and out of cells, as well as its processing of sequential
data. The LSTM has feedback connections and can handle the input sequences of arbitrary
length and is often preferred in the classification, processing, and prediction based on time
series data. Several studies used LSTM for crop yield prediction with impressive results.
LSTM not only captures trends in the data but also describes the dependencies of the time
series data. Tian et al. built an LSTM model by integrating the meteorological data and
two remote sensing indices (vegetation temperature condition index (VTCI) and LAI) to
estimate wheat yield in Guanzhong Plain [32]. Jeong et al. used water-related indices and
the maximum temperature as inputs for rice yield prediction using an LSTM model, which
showed reliable early prediction accuracy [16]. Sun et al. used a CNN-LSTM model to
predict the end-of-season and in-season yields of soybean in the county. The input data for
the model included meteorological data and MODIS surface temperature (LST) [24]. The
LSTM model was shown to have high prediction accuracy for crop yield estimation, but all
of the above methods for estimating crop yield use multiple data as input parameters, and
it is difficult to obtain non-remote sensing data in some regions [33–35]. There are a large
number of mature remote sensing products for LAI data, so this study mainly considered
using LAI remote sensing products as single model input data and what kind of accuracy
could be achieved when using deep learning algorithms for yield estimation.
In this study, LSTM was used to estimate the winter wheat yield at the county scale
based on the relationship between time series LAI products and winter wheat yield. Con-
sidering the simplicity of obtaining LAI data, the model input parameters were only the
leaf area index to verify the accuracy that could be achieved under the influence of a
single factor. Moreover, the time step of the input remote sensing data of the model was
considered so as to determine the accuracy in different time data.

2. Materials and Methods

2.1. Study Area
Henan Province is located in eastern central China (31◦ 230 ~36◦ 220 N, 110◦ 210 ~116◦ 390 E).
Figure 1 is a schematic diagram of the location of the study area. Henan Province is located
between the warm temperate zone and the subtropical zone. The terrain of Henan Province
is high in the west and low in the east, with mountains above 1000 m above sea level in
the west and plains below 100 m in the east. Mountains and hills account for 44.3% of
the total area, and plains account for 55.7%. The total wheat output of Henan Province
ranks first in China, accounting for more than 28% of the country’s total wheat output.
The sowing time of wheat varies from north to south by nearly two weeks, and there is a
large gap in yield in different regions. The topography of Henan Province is complex and
diverse, the topography is low in the east and high in the west, with significant differences;
the surface morphology is complex and diverse. Due to the influence of landforms and
Agriculture 2022, 12, x FOR PEER REVIEW 4 of 13

large amount of quality information associated with each pixel, whether the pixel is la-
Agriculture 2022, 12, 1707 3 of 13
beled as cloudy, clear, or in cloud shadow. In a subsequent study, MODIS LAI selected
high-quality pixels derived using the master algorithm under cloud-free conditions. Fig-
ure 2 shows the time series results of the MODIS LAI for the entire winter wheat growing
monsoons,
region Henan
in Henan, and Province has a wide
the maximum valuevariety
of the of
LAIsoilwas
types and large differences
approximately in climate
2. Considering
resources, resulting in significant differences in crop yields in different regions.
that LAI is at a low level in the early and late stages of wheat growth, and that these According
to the
growth production
stages conditions
do not have of producing
a particularly areas andonthe
strong influence climatic
yield characteristics
formation, of the
we used LAI
wheat growth period, the wheat-growing
from flowering to maturity each year as input data.areas in Henan Province can be divided into
five regions.

Figure 1. The geographic location of the study area and classification results (The red box on the left
Figure 1. The geographic location of the study area and classification results (The red box on the left
is location
is the the location of Henan
of Henan Province.
Province. (a)—false
(a)—false color
color composite
composite withwith the MODIS09
the MODIS09 data;
data; andand (b)—land
(b)—land
classification,
classification, green
green represents
represents the the winter
winter wheat
wheat region).
region).

1. The wheat area in Nanyang Basin, including Nanyang City and Biyang Countyin
Zhumadian, is a typical rainfed and semi-rainfed area due to the relatively poor field
supporting projects and irrigation conditions.
2. Rice stubble and wheat areas in southern Henan, including Xinyang, southern Zhu-
madian, and Nanyang Tongbai. The soil in this area is heavy, and the precipitation
during the wheat growth period is relatively high.
3. In western Henan, southwestern Henan, and northern Henan, dry wheat areas,
including Luoyang, Sanmenxia, Jiyuan, Pingdingshan, Anyang, and other shallow
hilly areas, drought, winter, and spring freezing damage, rust, and yellow dwarf
disease are the main factors affecting wheat yield.
4. The wheat area in north-central Henan Province, including Xuchang, Zhengzhou,
Luoyang, and the irrigated land north of the Yellow River, has good production
conditions and high production levels.
Agriculture 2022, 12, x FOR PEER REVIEW 4 of 13

large amount of quality information associated with each pixel, whether the pixel is la-
beled as cloudy, clear, or in cloud shadow. In a subsequent study, MODIS LAI selected
Agriculture 2022, 12, 1707 4 of 13
high-quality pixels derived using the master algorithm under cloud-free conditions. Fig-
ure 2 shows the time series results of the MODIS LAI for the entire winter wheat growing
region in Henan, and the maximum value of the LAI was approximately 2. Considering
5. LAI
that The is
wheat
at a area
low in the central
level and eastern
in the early partstages
and late of Henan Province
of wheat includes
growth, andthe irrigated
that these
land in the middle- and high-yield wheat areas in the north-central part of Zhumadian,
growth stages do not have a particularly strong influence on yield formation, we used LAI
Luohe, Zhoukou,
from flowering Shangqiu,
to maturity andas
each year Pingdingshan
input data. Mountain.
The topography and climate within each of the above production areas are relatively
consistent; so, we developed a winter wheat yield model for each production area. The
actual yield data for winter wheat in this study were provided by the Henan Provincial
Bureau of Statistics.

2.2. MODIS LAI

LAI is defined as the area of unilateral green leaves per unit of ground area in a
broadleaf canopy and half of the total needle surface area per unit ground area in a
coniferous canopy. The LAI product selected for this study was MCD15A2H. MCD15A2H
is an 8-day composite product with a total of 46 scenes per year and a spatial resolution of
500 m. The inversion algorithm for the Moderate Resolution Imaging Spectroradiometer
(MODIS) LAI product was a look-up table constructed based on a three-dimensional
radiative transfer model. When the main algorithm failed, a backup algorithm using an
empirical relationship between NDVI and the canopy LAI was triggered to estimate the
LAI for each pixel, and the look-up table was used to compare whether the observed and
simulated canopy top BRFs were within a given biologically relevant range. All canopy/soil
patterns and corresponding LAI values that differed between the modeled and observed
BRFs within a given level of uncertainty were considered acceptable solutions. The product
used for this study was version 6, and the final result was the true LAI.
All satellite data are archived in HDF-EOS format, and the MODIS Reprojection Tool
(MRT) software provided by NASA enables the user to read the HDF-EOS format. This
software supports the performing geo-transformations to different coordinate systems
or cartographic projections and writing the output to other file formats (GeoTIFF). All
data are initially projected onto an integer sine wave (ISIN) mapping grid. These data
were corrected to UTM coordinates using MRT and resampled using the nearest neighbor
algorithm. Non-wheat fields were masked using a land cover classification, and then the
corresponding areas were cut out of the remotely sensed imagery using SHP data based
on the extent of each city. The MODIS LAI products include the quality control (QC)
information designed to help users make the best use of these data. Each QC layer has a
large amount of quality information associated with each pixel, whether the pixel is labeled
as cloudy, clear, or in cloud shadow. In a subsequent study, MODIS LAI selected high-
quality pixels derived using the master algorithm under cloud-free conditions. Figure 2
shows the time series results of the MODIS LAI for the entire winter wheat growing region
in Henan, and the maximum value of the LAI was approximately 2. Considering that LAI
is at a 1.
Figure lowThelevel in the early
geographic and
location of late stages
the study of wheat
area growth, and
and classification that(The
results these
redgrowth stages
box on the left
dothe
is not have aofparticularly
location strong
Henan Province. influence
(a)—false oncomposite
color yield formation,
with thewe used LAI
MODIS09 from
data; andflowering
(b)—land
classification,
to maturity eachgreenyear
represents thedata.
as input winter wheat region).

Figure 2. The mean curve of MODIS LAI in Henan Province from 2003 to 2016. The red line is the
error bars plotted using standard deviation; blue dots represent the mean; between the two green
dotted lines is the LAI for one year.
Agriculture 2022, 12, 1707 5 of 13

2.3. Mapping of Wheat Distribution

This study focused on winter wheat, which first required the extraction of winter
wheat distribution areas from remote sensing images. We used the product produced by
Dong [36]. The product was produced by first synthesizing monthly NDVI maxima and
using FROM-GLC as a priori knowledge to obtain crop distribution information; then, we
used the standard seasonal growth curve of winter wheat combined with the time-weighted
dynamic time warping (TWDTW) to determine the area of winter wheat. The accuracy of
the final product was higher than 89.30% and 90.59% for producers and users, respectively.
To maintain consistency with the resolution of MODIS data, the classification results were
finally resampled to 500 m using the mode resampling method. Mode resampling selects
the value with the highest frequency of occurrence among all sampling points, and the
results maintain the real state of the ground surface to some extent. The classification
results are shown in Figure 1b.

2.4. LSTM
The recurrent neural network (RNN) is a type of neural network with short-term
memory capabilities. In a cyclic neural network, a neuron cannot only receive information
from other neurons but also its own information, forming a network structure with loops.
Compared with feedforward neural networks, recurrent neural networks are more in line
with the structure of biological neural networks. Recurrent neural networks have been
widely used in tasks such as speech recognition, language modeling, and natural language
generation. The parameter learning of recurrent neural networks can be learned by the
backpropagation algorithm over time. The backpropagation algorithm with time transmits
the error information step by step in the reverse order of time. When the input sequence
is relatively long, there will be the problem of gradient explosion and disappearance. In
order to solve this problem, people have made many improvements to the cyclic neural
network. The most effective means of improvement is to introduce a gating mechanism,
one of which is called a long short-term memory network (LSTM). LSTM is a variant of a
cyclic neural network, which can effectively solve the problem of gradient explosion or the
disappearance of a simple cyclic neural network.
The ingenuity of LSTM is that, by increasing the input threshold, the forgetting
threshold, and output threshold, the weight of the self-loop is changed. As such, when
the model parameters are fixed, the integration scale at different times can be dynamically
changed, thereby avoiding the problem of gradient disappearance or gradient expansion.
The LSTM network introduces a gating mechanism to control the path of information
transmission. The three “gates” are the input gate it , the forget gate f t , and the output gate
ot . The functions of these three gates are as follows:
(1) The forgetting gate f t controls how much information needs to be forgotten to
control the internal state ct−1 at the last moment;
(2) The enter gate it controls the candidate state cet at the current moment and how
much information needs to be saved;
(3) The output gate ot controls how much information of the internal state ct at the
current moment needs to be output to the external state ht .
When f t = 0 and it = 1, the memory unit clears the historical information and writes
the candidate state vector cet . However, the memory unit ct is still related to the historical
information at the previous moment. When f t = 1 and it = 0, the memory unit will copy the
content of the previous moment without writing new information.
Figure 3 shows the cyclic unit structure of the LSTM network. The calculation process
is: (1) first use the external state ht−1 at the previous moment and the input xt at the current
moment to calculate the three gates and the candidate state cet ; (2) combine the forget gate
f t and the input gate it to update the memory unit ct ; and (3) combine the output gate ot to
transfer the information of the internal state to the external state ht .
Agriculture 2022, 12, x FOR PEER REVIEW 6 of 13

Agriculture 2022, 12, 1707 6 of 13

forget gate 𝑓 and the input gate 𝑖 to update the memory unit 𝑐 ; and (3) combine the
output gate 𝑜 to transfer the information of the internal state to the external state ℎ .

Figure3.3.Cyclic
Figure Cycliccell
cellstructure
structureof
ofthe
theLSTM
LSTMnetwork.
network.

By
Bymeans
meansof
ofLSTM
LSTMcyclic
cyclicunits,
units,the
thewhole
wholenetwork
networkcan canbe
bebuilt
builtup
upwith
with long-distance
long-distance
temporal dependencies. It can be succinctly described
temporal dependencies. It can be succinctly described as as
 𝑐̃  𝑡𝑎𝑛ℎ 
cet tanh  
ot 𝑜 = 𝜎 𝑥 
 𝑖 =  σ 𝜎 w𝑤 xℎt ++b𝑏 , (1)

 it   σ  (1)
h t −1
𝑓
f t σ𝜎
𝑐 =K 𝑓 ⨀𝑐 + 𝑖K ⨀𝑐̃ , (2)
ct = f t c t −1 + i t cet (2)
ℎ o=
t 𝑜 ⨀tanh
tan h((𝑐
ct )). (3)
K
ht = (3)
Inthe
In theLSTM
LSTMnetwork,
network, aa memory
memory unitunit cc can
can capture
capture aakey
keypiece
pieceof
ofinformation
information atat aa
certainmoment
certain momentandandcan
canstore
storethis
thiskey
keyinformation
informationfor foraacertain
certaintime
timeinterval.
interval.The
Thelifetime
lifetime
ofinformation
of information stored
stored in
in memory
memory unit
unit cc isislonger
longerthan
thanthat
thatofofshort-term
short-term memory
memory hhbut but
Agriculture 2022, 12, x FOR PEER REVIEW
muchshorter
much shorterthan
thanthat
thatofoflong-term
long-termmemory.
memory.FigureFigure44shows
showsthetheworkflow
workflowfrom
fromLAI
LAI7 data
of 13
data
processingto
processing toLSTM
LSTMmodel
modelbuilding
buildingand
andyieldyieldestimation.
estimation.

2.5. Model Evaluation Metrics

We chose root mean square error (RMSE), coefficient of determination (R2), and mean
absolute percentage error (MAPE), the three metrics in Equations (4)–(6). To evaluate the
performance of the model, the difference between the predicted and actual statistical val-
ues of the model was calculated. A smaller RMSE indicates a better performance of the
model, and a larger R2 indicates a higher regression accuracy of the model. The lower the
value of MAPE and RMSE, the higher the accuracy of the obtained predictive model.
MAPE measures the error in percentage and specifies the average percentage deviation
between the forecast value and the actual implementation [37]. Usually, the fit of the
model is perfect when the MAPE value is below 10% and when it is in the range from 10%
to 20%, the model fit is good. In the range of 20–30%, the error level is acceptable and
when it exceeds 30%, the model is a poor fit and should be rejected [38].

𝑅𝑀𝑆𝐸 = ∑ (𝑦 − 𝑦 ) , (4)

∑ (𝑦 − 𝑦 )
𝑅 =1− , (5)
∑ (𝑦 − 𝑦 )

100% 𝑦 −𝑦
𝑀𝐴𝑃𝐸 = , (6)
𝑛 𝑦

where 𝑦 represents the actual yield, 𝑦 is the predicted yield, and 𝑛 is the sample size.

Figure 4. Overall
Figure4. Overall structure
structureof
ofthe
theLSTM
LSTMmodel
modelfor
forwheat
wheatyield
yieldestimation.
estimation.

3. Results and Discussion

3.1. Yield Estimation under Different Time Steps and Input Combinations
We used winter wheat LAI time series data from 2003 to 2015 as training samples
Agriculture 2022, 12, 1707 7 of 13

2.5. Model Evaluation Metrics

We chose root mean square error (RMSE), coefficient of determination (R2 ), and mean
absolute percentage error (MAPE), the three metrics in Equations (4)–(6). To evaluate the
performance of the model, the difference between the predicted and actual statistical values
of the model was calculated. A smaller RMSE indicates a better performance of the model,
and a larger R2 indicates a higher regression accuracy of the model. The lower the value
of MAPE and RMSE, the higher the accuracy of the obtained predictive model. MAPE
measures the error in percentage and specifies the average percentage deviation between
the forecast value and the actual implementation [37]. Usually, the fit of the model is perfect
when the MAPE value is below 10% and when it is in the range from 10% to 20%, the model
fit is good. In the range of 20–30%, the error level is acceptable and when it exceeds 30%,
the model is a poor fit and should be rejected [38].
r
1 n
RMSE =
n ∑i=1 (ŷi − yi )2 (4)

2
∑in=1 (ŷi − yi )
R2 = 1 − 2
, (5)
∑in=1 (yi − ŷi )
100% n ŷi − yi
n i∑
MAPE = , (6)
=1
yi
where yi represents the actual yield, ŷi is the predicted yield, and n is the sample size.

3. Results and Discussion

3.1. Yield Estimation under Different Time Steps and Input Combinations
We used winter wheat LAI time series data from 2003 to 2015 as training samples
and modeled them with LSTM to obtain the winter wheat yield prediction results for 2016
and compared the predicted yield with the statistical yield in 2016 to verify the prediction
accuracy of the model. Moreover, considering the characteristics of the LSTM model and
the requirements of the yield prediction task, the input data required for training were
processed as subsequently described, and the winter wheat growth data from March to the
end of May each year were selected so that the input data for one year had 12 LAI values
containing data from the flowering stage to the maturity stage. The fitted data of the model
changed at different input steps, which also had an impact on the accuracy of the prediction
results. To obtain the input step with the highest accuracy, the input steps from 1 to 6 were
compared. The overall RMSE for Henan Province under the six step-size scenarios is shown
in Figure 5. The solid red line in the figure is the 1:1 line, and the estimation results were
mostly evenly distributed on both sides with the solid line as the center, indicating that the
model had a good prediction for the yield. However, when the yield exceeded 7000 kg/ha,
the estimation accuracy of the model tended to decrease, and the yield prediction tended
to be underestimated, which is consistent with the systematic trend of underestimation at
high yields in previous studies, mainly due to the relatively low proportion of data from
high production areas in the sample. We first constructed LSTM models for five winter
wheat production areas and finally obtained the winter wheat yield estimation results for
the whole of Henan Province. A–E in Table 1 show the RMSE, R2 and MAPE of the five
grain-producing regions at different step sizes, and F shows the statistical results for the
whole Henan Province. Figure 6 shows the RMSE histograms of the LSTM model for the
five regions and the whole of Henan Province under different step lengths. When the step
size was short, the predicted yields of the LSTM model for different production areas had
great instability compared with the statistical yields in the field, but when the step size
increased to 3 and 4, the accuracy started to improve and stabilized. However, the accuracy
decreased again when the step size was 6. The LSTM model achieved optimal values of R2
and RMSE at a step size of 4, with results of 88% and 532.16 kg/ha, respectively, while the
value of MAPE was 4.44%, which was at a very high level of fit. Although the problem of
 different step lengths. When the step size was short, the predicted yields of the LSTM
model for different production areas had great instability compared with the statistical
yields in the field, but when the step size increased to 3 and 4, the accuracy started to
improve and stabilized. However, the accuracy decreased again when the step size was 6.
Agriculture 2022, 12, 1707 The LSTM model achieved optimal values of R2 and RMSE at a step size of 4, with results 8 of 13
of 88% and 532.16 kg/ha, respectively, while the value of MAPE was 4.44%, which was at
a very high level of fit. Although the problem of LSTM was effectively solved in terms of
gradient
LSTM was disappearance or explosion
effectively solved in termscompared with
of gradient the general RNN
disappearance networks,compared
or explosion the win-
ter wheat yields were not closely related to yields from many years ago, and
with the general RNN networks, the winter wheat yields were not closely related to yields to some ex-
tent, were only strongly correlated with historical yields from the last three
from many years ago, and to some extent, were only strongly correlated with historical or four years,
and there
yields may
from thebelast
a loss of accuracy
three if theand
or four years, current
thereresults
may beareafitted
loss ofwith data from
accuracy if thea current
longer
period
resultsof time
are ago.
fitted with data from a longer period of time ago.

Figure 5. Comparison of yield estimation results for LSTM models based on different time steps:
Figure 5. Comparison of yield estimation results for LSTM models based on different time steps:
(A) one time
(A) one time step;
step; (B)
(B) two
two time
timesteps;
steps;(C)
(C)three
threetime
timesteps;
steps;(D)
(D)four time
four timesteps; (E)(E)
steps; fivefive
time steps;
time and
steps;
(F) six
and (F) time steps.
six time TheThe
steps. probability distributions
probability of the
distributions statistical
of the and
statistical estimated
and estimatedyields
yieldsareare
plotted on
plotted
thethe
on right and
right toptop
and of of
thethe
Y axis, respectively.
Y axis, respectively.

Table 1.
Table The accuracy
1. The accuracy evaluation
evaluationof
ofthe
the LSTM
LSTMprediction
predictionmodel
modelwas
wascompared
comparedbetween
betweendifferent
different
timesteps
time stepswhen
when1212LAI
LAIdata
datawere
wereinput.
input.

Time Steps RMSE

RMSE (kg/ha) (kg/ha) R2 R2 MAPE
MAPE (%)(%)
Time
Steps A A B BC CD D
E FE A F B A C B DC D
E EF F A A B B CC D
D EE
1 1 1039.93
1039.93 436.63 436.63
350.59 350.59
386.97 386.97 687.70 0.40
687.70 563.38 563.38
0.79 0.40 0.790.92
0.95 0.95 0.92
0.01 0.01
0.870.87 12.33 5.01
12.33 5.01 4.28
4.28 5.19
5.19 7.29
7.29
2 2 468.29468.29
548.72 548.72
831.34 699.47 699.47
831.34 219.07 632.77
219.07 0.86 0.73 0.86
632.77 0.74
0.730.81 0.27 0.27
0.74 0.81 0.830.837.82
7.82 7.72
7.72 10.57 12.40 2.31
10.57 12.40 2.31
3 604.58 589.19 635.35 767.26 583.36 637.88 0.86 0.78 0.87 0.82 0.03 0.85 10.17 9.29 8.79 12.63 5.90
4 3 530.79604.58
461.56 589.19
568.10 635.35
667.68 767.26 583.36 0.83
396.56 532.08 637.88
0.78 0.86 0.780.90
0.87 0.87 0.82
0.38 0.03
0.880.857.56
10.17 7.24
9.29 8.02
8.79 12.63
11.75 5.90
4.44
5 4 393.49530.79
549.11 461.56
635.34 699.27 667.68
568.10 713.08 637.27
396.56 0.90 0.74 0.83
532.08 0.80
0.780.70 0.00 0.38
0.87 0.90 0.780.886.99
7.56 8.70
7.24 9.40 13.01 4.44
8.02 11.75 8.45
6 795.51 414.61 770.62 548.72 674.49 678.36 0.86 0.80 0.78 0.81 0.24 0.80 13.41 6.96 12.04 9.17 7.95
5 393.49 549.11 635.34 699.27 713.08 637.27 0.90 0.74 0.80 0.70 0.00 0.78 6.99 8.70 9.40 13.01 8.45
6 795.51 414.61 770.62 548.72 674.49 678.36 0.86 0.80 0.78 0.81 0.24 0.80 13.41 6.96 12.04 9.17 7.95
The respective highest accuracy estimation results in different regions differed sig-
nificantly in time steps. For the whole of Henan Province, the overall accuracy may not
be optimal if the same step was used for yield estimation. The lowest value of RMSE can
be seen in Figure 6 in the southwestern production area, with an RMSE of approximately
400, and the highest in time steps was in southwestern Henan, with approximately 700.
There was a correlation between the winter wheat yield and time series LAI data, but the
interannual correlation was to some extent not a more accurate result by modeling with
more data, and there are many factors affecting the wheat yield, including variation in
variety, temperature, and topography, which can all cause yield fluctuations, although all
these factors can affect the LAI time series to some extent. However, when using time series
LAI data alone as input parameters, it is not better to use more data, and it is not better to
set a higher time step; these should be quantitatively and individually tested for different
regions and not generalized.
Agriculture 2022, 12, x FOR PEER REVIEW 9 of 13
Agriculture 2022, 12, 1707 9 of 13

Figure 6.
Figure The model
6. The model performance
performance (predicted
(predicted RMSE)
RMSE)using
usingdifferent
differenttime
timesteps
stepsfor
forthe
thewhole
wholegrowing
grow-
season.
ing (A)(A)
season. oneone
timetime
step; (B) (B)
step; twotwo
timetime
steps; (C) three
steps; timetime
(C) three steps; (D) four
steps; timetime
(D) four steps; (E) five
steps; (E) time
five
steps;
time and (F)
steps; andsix
(F)time steps.steps.
six time

3.2. Yield Estimation with Different Input Time Series Data

The respective highest accuracy estimation results in different regions differed sig-
In the
nificantly inprocess of winter
time steps. For thewheatwholeyield estimation,
of Henan Province,it isthe
always
overall desirable
accuracy tomay
predictnot the
be
optimal if the same step was used for yield estimation. The lowest value of RMSE canLAI
yield as early as possible. Therefore, we considered shortening the time range of the be
data in
seen used and the
Figure 6 inused
the data from the plucking
southwestern production stagearea,
to the filling
with an stage
RMSEforofmodeling
approximatelyso that
theirand
400, yield thecould be predicted
highest 16 days
in time steps was before maturity. TheHenan,
in southwestern LAI data withused in this study were
approximately 700.
only six
There wasper year, whichbetween
a correlation halved the theamount of datayield
winter wheat compared
and time to previous
series LAI studies, making
data, but the
it easier to obtain the data, especially the high spatial resolution
interannual correlation was to some extent not a more accurate result by modeling with satellite data. The results
of thedata,
more comparison
and there between the predicted
are many yield andthe
factors affecting thewheat
actualyield,
statistical yield are
including shown in
variation in
Figure 7, from which there was no significant decrease in the prediction
variety, temperature, and topography, which can all cause yield fluctuations, although all accuracy for all of
Henan
these compared
factors to thethe
can affect 12 data,
LAI timeand the accuracy
series to some of individual
extent. However, time steps when increased.
using timeThe
best time step for the overall accuracy occurred at 2, where the R2 and RMSE were 87%
series LAI data alone as input parameters, it is not better to use more data, and it is not
and 522.32 kg/ha, respectively, while MAPE was 5.67%. A–E in Table 2 shows the RMSE,
better to set a higher time step; these should be quantitatively and individually tested for
R2 and MAPE of the five grain-producing regions at different step sizes, and F shows the
different regions and not generalized.
statistical results for the whole Henan Province; Figure 8 shows the histogram of RMSE.
However, it can also be seen that the value of RMSE gradually increased with increasing
3.2. Yield Estimation with Different Input Time Series Data
step size. This may be due to the fact that LAI data were strongly correlated only in the last
two In the process
years, and the of winter
larger thewheat yield estimation,
time difference, the lower it isthe
always desirable
correlation to predict
between LAI andthe
yield as early as possible. Therefore, we considered shortening
yield; moreover, the interannual LAI was not strongly correlated with early and late yield the time range of the LAI
data used and
formation, so inthethe
used data of
process from theestimation
yield plucking stage usingtotimethe series
fillingLAI,
stagea for modeling
higher accuracy so
that
was their
obtained yieldbycould
using be predicted
data from the16 days before
nodulation maturity.
to filling stage.The LAI
It has data
been used that
shown in this
the
study were only six per year, which halved the amount of
accumulation of dry matter in winter wheat is mainly concentrated at the nodulation anddata compared to previous
studies,
gestationmaking
stages,itandeasier
the to
LAI obtain
in thisthe data,was
period especially
closelythe high spatial
correlated with resolution satellite
the yield formation
data. The results of the comparison between the predicted yield
of winter wheat, which is also more consistent with the results of this study. It should alsoand the actual statistical
yield are shown
be noted that theinoverall
FigureRMSE7, fromaccuracy
which there wasbest
was the no significant
except for decrease
the case of instep
the prediction
size 2. The
accuracy
RMSE of for the all of Henan
remaining compared
steps to the 12 data,
was significantly lowerand the accuracy
compared of individual
to the model with 12 time
data;
steps increased.
therefore, using The best time
as many growing stepperiod
for thedata
overall accuracywould
as possible occurred alsoat 2, where
make the R2more
the model and
RMSE
robustwere
when87% usingand 522.32 kg/ha,
time-series LAI forrespectively, while MAPE was 5.67%. A–E in Table 2
yield estimation.
shows the RMSE, R2 and MAPE of the five grain-producing regions at different step sizes,
and F shows the statistical results for the whole Henan Province; Figure 8 shows the his-
togram of RMSE. However, it can also be seen that the value of RMSE gradually increased
with increasing step size. This may be due to the fact that LAI data were strongly corre-
lated only in the last two years, and the larger the time difference, the lower the correlation
between LAI and yield; moreover, the interannual LAI was not strongly correlated with
early and late yield formation, so in the process of yield estimation using time series LAI,
a higher accuracy was obtained by using data from the nodulation to filling stage. It has
been shown that the accumulation of dry matter in winter wheat is mainly concentrated
at the nodulation and gestation stages, and the LAI in this period was closely correlated
with the yield formation of winter wheat, which is also more consistent with the results
 1 1180.61 534.22 798.75 338.77 656.31 727.11 0.17 0.70 0.80 0.96 0.19 0.81 17.63 5.22 10.29 5.67 6.64
2 372.66 552.27 of this study.
534.55 411.21It should
598.78 also be noted
522.32 that 0.86
0.87 0.67 the overall RMSE
0.93 0.18 0.87 accuracy
6.03 7.41was the 6.53
5.61 best except
5.67
3 704.62 362.19 for the case
867.82 of step
739.41 size 2.792.37
901.01 The RMSE of the
0.53 0.85 remaining
0.75 0.67 0.01steps
0.74 was
9.78 significantly lower9.83
4.63 11.11 12.23 com-
4 1010.14 789.28 pared
800.46to 850.57
the model with 12
978.56 data; therefore,
876.48 using
0.18 0.31 0.74 as0.00
0.63 many growing
0.67 period
14.36 12.82 data12.37
11.30 as possible
12.99
5 2022,743.15
Agriculture 12, 1707 591.05 would
928.28 also make1255.65
743.79 the model more0.44
945.34 robust
0.61when using
0.63 0.67 time-series
0.00 0.60 11.69LAI for12.73
10.62 yield12.17
estimation.
17.03
10 of 13
6 1052.22 1572.15 1170.50 1167.80 1954.85 1448.93 0.28 0.02 0.50 0.36 0.02 0.36 20.81 45.16 17.92 21.91 27.96
Table 2. The accuracy evaluation of the LSTM prediction model was compared between different
time steps when 6 LAI data were input.

RMSE (kg/ha) R2 MAPE (%)

Time Steps
A B C D E F A B C D E F A B C D E
1 1180.61 534.22 798.75 338.77 656.31 727.11 0.17 0.70 0.80 0.96 0.19 0.81 17.63 5.22 10.29 5.67 6.64
2 372.66 552.27 534.55 411.21 598.78 522.32 0.87 0.67 0.86 0.93 0.18 0.87 6.03 7.41 5.61 6.53 5.67
3 704.62 362.19 867.82 739.41 901.01 792.37 0.53 0.85 0.75 0.67 0.01 0.74 9.78 4.63 11.11 12.23 9.83
4 1010.14 789.28 800.46 850.57 978.56 876.48 0.18 0.31 0.74 0.63 0.00 0.67 14.36 12.82 11.30 12.37 12.99
5 743.15 591.05 928.28 743.79 1255.65 945.34 0.44 0.61 0.63 0.67 0.00 0.60 11.69 10.62 12.73 12.17 17.03
6 1052.22 1572.15 1170.50 1167.80 1954.85 1448.93 0.28 0.02 0.50 0.36 0.02 0.36 20.81 45.16 17.92 21.91 27.96

Comparisonofofthe
Figure7.7.Comparison
Figure theyield
yieldestimation
estimationresults
resultsfor
forLSTM
LSTMmodels
modelsbased
basedon ondifferent
differenttime
timesteps
steps
(A) one time step; (B) two time steps; (C) three time steps; (D) four time steps; (E) five
(A) one time step; (B) two time steps; (C) three time steps; (D) four time steps; (E) five time steps;time steps; and
(F) (F)
and six six
time steps.
time TheThe
steps. probability
probabilitydistributions of the
distributions statistical
of the and
statistical estimated
and estimated yields areare
yields plotted on
plotted
thethe
on right
rightand top
and topof of
thethe
Y axis, respectively.
Y axis, respectively.

Table 2. The accuracy evaluation of the LSTM prediction model was compared between different
time steps when 6 LAI data were input.

Time RMSE (kg/ha) R2 MAPE (%)

Steps A B C D E F A B C D E F A B C D E
1 1180.61 534.22 798.75 338.77 656.31 727.11 0.17 0.70 0.80 0.96 0.19 0.81 17.63 5.22 10.29 5.67 6.64
2 372.66 552.27 534.55 411.21 598.78 522.32 0.87 0.67 0.86 0.93 0.18 0.87 6.03 7.41 5.61 6.53 5.67
3 704.62 362.19 Figure
867.82 7. Comparison
739.41 901.01 of the yield
792.37 estimation
0.53 0.85 0.75results
0.67 for LSTM
0.01 0.74models
9.78 based
4.63 on11.11
different time steps
12.23 9.83
4 1010.14 789.28 (A) one
800.46 time step;
850.57 978.56(B) two time 0.18
876.48 steps;0.31
(C) three
0.74 time
0.63 steps; (D) four
0.00 0.67 14.36time12.82
steps;11.30
(E) five12.37
time steps;
12.99
5 743.15 591.05 and (F)
928.28 six time
743.79 steps. The
1255.65 probability
945.34 distributions
0.44 0.61 0.63 0.67of the
0.00statistical
0.60 11.69and estimated
10.62 12.73yields are plotted
12.17 17.03
6 1052.22 1572.15 1170.50 1167.80
on the 1954.85
right and top of 1448.93 0.28
the Y axis, 0.02 0.50 0.36 0.02 0.36 20.81 45.16 17.92 21.91 27.96
respectively.

Figure 8. The model performance (predicted RMSE) using different time steps for only one period of
the growing season. (A) one time step; (B) two time steps; (C) three time steps; (D) four time steps;
(E) five time steps; and (F) six time steps.

3.3. Performance Comparison with Machine Learning Methods

Here, four machine learning methods (random forest, support vector regression, par-
tial least squares regression, and XGBoost) were used to construct county-level wheat yield
Agriculture 2022, 12, 1707 11 of 13

models for each agroecological zone. Random forest (RF) is a supervised machine learning
algorithm based on integration learning [39]. Different subsets are randomly drawn from
the provided data and used to build several different decision trees and integrate the results
of one decision tree according to Bagging’s rules. Support vector regression (SVR) is a
regression algorithm that is a variant of SVM in regression analysis [40]. SVR also consid-
ers maximization intervals but considers points within the decision boundary so that as
many sample points as possible lie within the interval. The partial least squares regression
(PLSR) [41] algorithm is a regression modeling method for multiple dependent variables
Y on multiple independent variables X. The algorithm considers extracting as many prin-
cipal components as possible from Y and X in building the regression and maximizing
the correlation between the principal components extracted from X and Y, respectively.
XGBoost [42] is a scalable machine learning system that adds to the objective function of
each iteration regular term to further reduce the risk of overfitting, XGBoost is an all-in-one
machine learning algorithm.
The yield prediction of winter wheat in Henan Province was constructed using the
four methods mentioned above. The prediction model first used winter wheat LAI time
series data from 2003 to 2015, followed by yield prediction for 2016, and the R2 and RMSE
of the prediction results are shown in Table 3. For the whole of Henan Province, the best
performance among the four methods was the SVR with R2, RMSE and MAPE of 0.76,
725.8 kg/ha and 6.33%, respectively, and the worst was PLSR with R2, RMSE and MAPE
of 0.7 and 809.1 kg/ha and 7.74%, respectively. Compared with these machine learning
methods, the prediction results of the LSTM had better accuracy and performance both for
individual wheat growing areas and for the whole of Henan province.

Table 3. Accuracy evaluations comparison among different methods.

RMSE (kg/ha) R2 MAPE (%)

Model
A B C D E F A B C D E F A B C D E
RF 566.2 705.8 946.5 693.3 656.6 774.5 0.66 0.37 0.62 0.73 0.35 0.72 9.31 9.48 9.05 11.24 7.04
SVR 605.0 558.9 980.0 545.2 505.4 725.8 0.60 0.56 0.60 0.83 0.61 0.76 10.26 8.53 10.26 9.31 6.33
PLSR 627.7 680.4 1051.4 558.2 676.8 809.1 0.57 0.37 0.55 0.82 0.31 0.70 11.03 11.18 11.28 9.06 7.74
XGBOOST 579.2 659.4 976.4 737.8 638.1 785.8 0.62 0.45 0.60 0.69 0.38 0.72 8.62 10.40 10.12 11.60 6.98

The prediction accuracy of the four machine learning methods was the lowest in the
southwest Henan region, which is a mountainous and hilly area with complex topography, a
small winter wheat growing area, and large yield variation, and there may be a lack of yield
accuracy in this region using conventional methods to construct the model. Compared with
these, using LSTM model for winter wheat yield prediction had a superior performance.
Compared to algorithms such as SVR and RF, the accuracy of estimation was not sufficient,
because the ability to analyze complex nonlinear relationships between long time series
variables is not as good as LSTM, resulting in poor model performance. These machine
learning methods do not consider the time correlation between winter wheat yields in the
modeling process, and the estimation for each year’s yield is conducted independently,
while LSTM takes into account the time series correlation of yields and also can better
handle the nonlinear relationship, so it has higher accuracy compared to machine learning.
Overall, a better performance capability can be obtained using LSTM models for forecasting
time series data.

4. Conclusions
In this study, considering the complexity of data collection, we used LAI as a single
input variable and five models, including four machine learning models (RF, SVR, PLSR,
and XGBOOST) and one deep learning model (LSTM) to predict the winter wheat yield in
Henan Province in 2016. In general, the LSTM model had superior performance compared
with the machine learning models. Moreover, considering the characteristics of the LSTM,
the time step of the modeled data as well as the growth period data were analyzed, and the
Agriculture 2022, 12, 1707 12 of 13

time step needs to be analyzed for different growth regions, while using only the necessary
growth period data can also obtain a high prediction accuracy; however, for the robustness
of the model, the more growth period data are used accordingly, the better. To date, winter
wheat yield prediction based on remote sensing images has been carried out at the county
level. However, the determination of crop yield remains a challenge because the variability
and uncertainty within the region is unknown. The results of our study on winter wheat
yield prediction at a regional scale using publicly available data, using LAI as an input
variable for determining crop yield, can potentially be applied to crop yield estimation in
regions with sparse observational data and worldwide.

Author Contributions: Conceptualization, J.W.; resources, Z.G.; writing—original draft preparation,

J.W.; writing—review and editing, J.W., L.S. and H.S. All authors have read and agreed to the
published version of the manuscript.
Funding: This research was funded by the National Natural Science Foundation of China (NO.42101362,
31501225); the Natural Science Foundation of Henan Province of China (NO.222300420463).
Institutional Review Board Statement: Not applicable.
Data Availability Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Becker-Reshef, I.; Justice, C.; Sullivan, M.; Vermote, E.; Tucker, C.; Anyamba, A.; Small, J.; Pak, E.; Masuoka, E.; Schmaltz, J.
Monitoring global croplands with coarse resolution earth observations: The Global Agriculture Monitoring (GLAM) project.
Remote Sens. 2010, 2, 1589–1609. [CrossRef]
2. Tilman, D.; Balzer, C.; Hill, J.; Befort, B.L. Global food demand and the sustainable intensification of agriculture. Proc. Natl. Acad.
Sci. USA 2011, 108, 20260–20264. [CrossRef] [PubMed]
3. Becker-Reshef, I.; Vermote, E.; Lindeman, M.; Justice, C. A generalized regression-based model for forecasting winter wheat
yields in Kansas and Ukraine using MODIS data. Remote Sens. Environ. 2010, 114, 1312–1323. [CrossRef]
4. Xie, Y.; Wang, P.; Bai, X.; Khan, J.; Zhang, S.; Li, L.; Wang, L. Assimilation of the leaf area index and vegetation temperature
condition index for winter wheat yield estimation using Landsat imagery and the CERES-Wheat model. Agric. For. Meteorol.
2017, 246, 194–206. [CrossRef]
5. Mkhabela, M.; Bullock, P.; Raj, S.; Wang, S.; Yang, Y. Crop yield forecasting on the Canadian Prairies using MODIS NDVI data.
Agric. For. Meteorol. 2011, 151, 385–393. [CrossRef]
6. Lai, Y.; Pringle, M.; Kopittke, P.M.; Menzies, N.W.; Orton, T.G.; Dang, Y.P. An empirical model for prediction of wheat yield, using
time-integrated Landsat NDVI. Int. J. Appl. Earth Obs. Geoinf. ITC J. 2018, 72, 99–108. [CrossRef]
7. Moriondo, M.; Maselli, F.; Bindi, M. A simple model of regional wheat yield based on NDVI data. Eur. J. Agron. 2007, 26, 266–274.
[CrossRef]
8. Wall, L.; Larocque, D.; Léger, P. The early explanatory power of NDVI in crop yield modelling. Int. J. Remote Sens. 2008, 29,
2211–2225. [CrossRef]
9. Chen, Y.; Zhang, Z.; Tao, F. Improving regional winter wheat yield estimation through assimilation of phenology and leaf area
index from remote sensing data. Eur. J. Agron. 2018, 101, 163–173. [CrossRef]
10. Franch, B.; Vermote, E.; Becker-Reshef, I.; Claverie, M.; Huang, J.; Zhang, J.; Justice, C.; Sobrino, J.A. Improving the timeliness of
winter wheat production forecast in the United States of America, Ukraine and China using MODIS data and NCAR Growing
Degree Day information. Remote Sens. Environ. 2015, 161, 131–148. [CrossRef]
11. Pantazi, X.E.; Moshou, D.; Alexandridis, T.; Whetton, R.L.; Mouazen, A.M. Wheat yield prediction using machine learning and
advanced sensing techniques. Comput. Electron. Agric. 2016, 121, 57–65. [CrossRef]
12. Huang, J.; Ma, H.; Sedano, F.; Lewis, P.; Liang, S.; Wu, Q.; Su, W.; Zhang, X.; Zhu, D. Evaluation of regional estimates of winter
wheat yield by assimilating three remotely sensed reflectance datasets into the coupled WOFOST–PROSAIL model. Eur. J. Agron.
2019, 102, 1–13. [CrossRef]
13. Huang, J.; Sedano, F.; Huang, Y.; Ma, H.; Li, X.; Liang, S.; Tian, L.; Zhang, X.; Fan, J.; Wu, W.; et al. Assimilating a synthetic Kalman
filter leaf area index series into the WOFOST model to improve regional winter wheat yield estimation. Agric. For. Meteorol. 2016,
216, 188–202. [CrossRef]
14. Huang, J.; Tian, L.; Liang, S.; Ma, H.; Becker-Reshef, I.; Huang, Y.; Su, W.; Zhang, X.; Zhu, D.; Wu, W.; et al. Improving winter
wheat yield estimation by assimilation of the leaf area index from Landsat TM and MODIS data into the WOFOST model. Agric.
For. Meteorol. 2015, 204, 106–121. [CrossRef]
Agriculture 2022, 12, 1707 13 of 13

15. Zhuo, W.; Huang, J.; Li, L.; Zhang, X.; Ma, H.; Gao, X.; Huang, H.; Xu, B.; Xiao, X. Assimilating soil moisture retrieved from
Sentinel-1 and Sentinel-2 data into WOFOST model to improve winter wheat yield estimation. Remote Sens. 2019, 11, 1618.
[CrossRef]
16. Jeong, S.; Ko, J.; Yeom, J.-M. Predicting rice yield at pixel scale through synthetic use of crop and deep learning models with
satellite data in South and North Korea. Sci. Total Environ. 2022, 802, 149726. [CrossRef] [PubMed]
17. Bolton, D.K.; Friedl, M.A. Forecasting crop yield using remotely sensed vegetation indices and crop phenology metrics. Agric. For.
Meteorol. 2013, 173, 74–84. [CrossRef]
18. Krizhevsky, A.; Sutskever, I.; Hinton, G.E. Imagenet classification with deep convolutional neural networks. Commun. ACM 2017,
60, 84–90. [CrossRef]
19. Koirala, A.; Walsh, K.B.; Wang, Z.; McCarthy, C. Deep learning–Method overview and review of use for fruit detection and yield
estimation. Comput. Electron. Agric. 2019, 162, 219–234. [CrossRef]
20. Fischer, T.; Krauss, C. Deep learning with long short-term memory networks for financial market predictions. Eur. J. Oper. Res.
2018, 270, 654–669. [CrossRef]
21. LeCun, Y.; Bengio, Y.; Hinton, G. Deep learning. Nature 2015, 521, 436–444. [CrossRef] [PubMed]
22. Schmidhuber, J. Deep learning in neural networks: An overview. Neural Netw. 2015, 61, 85–117. [CrossRef] [PubMed]
23. Khaki, S.; Wang, L.; Archontoulis, S.V. A cnn-rnn framework for crop yield prediction. Front. Plant Sci. 2020, 10, 1750. [CrossRef]
[PubMed]
24. Sun, J.; Di, L.; Sun, Z.; Shen, Y.; Lai, Z. County-level soybean yield prediction using deep CNN-LSTM model. Sensors 2019, 19,
4363. [CrossRef]
25. Khaki, S.; Wang, L. Crop yield prediction using deep neural networks. Front. Plant Sci. 2019, 10, 621. [CrossRef]
26. Maimaitijiang, M.; Sagan, V.; Sidike, P.; Hartling, S.; Esposito, F.; Fritschi, F.B. Soybean yield prediction from UAV using
multimodal data fusion and deep learning. Remote Sens. Environ. 2020, 237, 111599. [CrossRef]
27. Cao, J.; Zhang, Z.; Luo, Y.; Zhang, L.; Zhang, J.; Li, Z.; Tao, F. Wheat yield predictions at a county and field scale with deep
learning, machine learning, and google earth engine. Eur. J. Agron. 2021, 123, 126204. [CrossRef]
28. Nevavuori, P.; Narra, N.; Linna, P.; Lipping, T. Crop yield prediction using multitemporal UAV data and spatio-temporal deep
learning models. Remote Sens. 2020, 12, 4000. [CrossRef]
29. Hara, P.; Piekutowska, M.; Niedbała, G. Selection of independent variables for crop yield prediction using artificial neural network
models with remote sensing data. Land 2021, 10, 609. [CrossRef]
30. Hochreiter, S.; Schmidhuber, J. Long short-term memory. Neural Comput. 1997, 9, 1735–1780. [CrossRef]
31. Graves, A. Generating sequences with recurrent neural networks. arXiv Prepr. 2013, arXiv:1308.0850.
32. Tian, H.; Wang, P.; Tansey, K.; Han, D.; Zhang, J.; Zhang, S.; Li, H. A deep learning framework under attention mechanism for
wheat yield estimation using remotely sensed indices in the Guanzhong Plain, PR China. Int. J. Appl. Earth Obs. Geoinf. ITC J.
2021, 102, 102375. [CrossRef]
33. Liu, Y.; Wang, S.; Wang, X.; Chen, B.; Chen, J.; Wang, J.; Huang, M.; Wang, Z.; Ma, L.; Wang, P.; et al. Exploring the superiority
of solar-induced chlorophyll fluorescence data in predicting wheat yield using machine learning and deep learning methods.
Comput. Electron. Agric. 2022, 192, 106612. [CrossRef]
34. Zhang, L.; Zhang, Z.; Luo, Y.; Cao, J.; Xie, R.; Li, S. Integrating satellite-derived climatic and vegetation indices to predict
smallholder maize yield using deep learning. Agric. For. Meteorol. 2021, 311, 108666. [CrossRef]
35. Cai, Y.; Guan, K.; Lobell, D.; Potgieter, A.B.; Wang, S.; Peng, J.; Xu, T.; Asseng, S.; Zhang, Y.; You, L.; et al. Integrating satellite
and climate data to predict wheat yield in Australia using machine learning approaches. Agric. For. Meteorol. 2019, 274, 144–159.
[CrossRef]
36. Dong, J.; Fu, Y.; Wang, J.; Tian, H.; Fu, S.; Niu, Z.; Han, W.; Zheng, Y.; Huang, J.; Yuan, W. Early-season mapping of winter wheat
in China based on Landsat and Sentinel images. Earth Syst. Sci. Data 2020, 12, 3081–3095. [CrossRef]
37. Sharma, L.K.; Singh, T.N. Regression-based models for the prediction of unconfined compressive strength of artificially structured
soil. Eng. Comput. 2018, 34, 175–186. [CrossRef]
38. Peng, J.; Kim, M.; Kim, Y.; Jo, M.; Kim, B.; Sung, K.; Lv, S. Constructing Italian ryegrass yield prediction model based on climatic
data by locations in South Korea. Grassl. Sci. 2017, 63, 184–195. [CrossRef]
39. Breiman, L. Random forests. Mach. Learn. 2001, 45, 5–32. [CrossRef]
40. Chang, C.-C.; Lin, C.-J. LIBSVM: A library for support vector machines. ACM Trans. Intell. Syst. Technol. 2011, 2, 1–27. [CrossRef]
41. Wegelin, J.A. A Survey of Partial Least Squares (PLS) Methods, with Emphasis on the Two-Block Case; Technical Report; University of
Washington: Seattle, DC, USA, 2000.
42. Chen, T.; Guestrin, C. Xgboost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International
Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA, 13–17 August 2016; pp. 785–794.