Journal of Geoscience and Environment Protection, 2022, 10, 265-281

https://www.scirp.org/journal/gep
ISSN Online: 2327-4344
ISSN Print: 2327-4336

Application of Parametric and Non Parametric

Classifiers for Assessing Land Use/Land Cover
Categories in Cocoa Landscape of Juaboso and
Bia West Districts of Ghana

Emmanuel Donkor1*, Edward Matthew Osei Jnr2, Stephen Adu-Bredu3,

Samuel A. Andam-Akorful2, Efiba Vidda Senkyire Kwarteng2, Lily Lisa Yevugah4
1
Resource Management Support Centre of Forestry Commission, Kumasi, Ghana
2
Kwame Nkrumah University of Science and Technology, Kumasi, Ghana
3
Forestry Research Institute of Ghana, Kumasi, Ghana
4
University of Energy and Natural Resources, Sunyani, Ghana

How to cite this paper: Donkor, E., Jnr, E. Abstract

M. O., Adu-Bredu, S., Andam-Akorful, S.
A., Kwarteng, E. V. S., & Yevugah, L. L. Satellite image classification has been used for long time in the field of remote
(2022). Application of Parametric and Non sensing since classification results are used in environmental research, agri-
Parametric Classifiers for Assessing Land culture, climate change and natural resource management. The cocoa land-
Use/Land Cover Categories in Cocoa Land-
scape of Juaboso and Bia West Districts of
scape of Ghana is complex and diverse in nature, composing of mixture of
Ghana. Journal of Geoscience and Environ- closed forest, open forest, settlements, croplands and cocoa farms which
ment Protection, 10, 265-281. make mapping the landscape difficult. The purpose of this research is to as-
https://doi.org/10.4236/gep.2022.1011018 sess and compare the classification performances of three machine learning
Received: October 12, 2022
classifiers: Support Vector Machine (SVM), Random Forest (RF), Artificial
Accepted: November 27, 2022 Neural Network (ANN) and a statistical classification algorithm: Maximum
Published: November 30, 2022 Likelihood (ML) to know which classifier is best suited for mapping the cocoa
landscape of Ghana using Juaboso and Bia West districts of Ghana as study
Copyright © 2022 by author(s) and
area. A representative sampling approach was adopted to collect 1246 sample
Scientific Research Publishing Inc.
This work is licensed under the Creative points for the various Land Use/Land Cover (LULC) types. These sample
Commons Attribution International points were divided at random into 869 which form 70% for classification
License (CC BY 4.0). and 377 which constitute 30% of the total sample points for validation. The
http://creativecommons.org/licenses/by/4.0/
Stacked sentinel-2 image, classification data and validation data storing the
Open Access
identities of the LULC classes were imported in R to run supervised classifi-
cation for each classifier. The classification results show that the highest over-
all accuracy and kappa statistics were produced by the support vector ma-
chine (86.47%, 0.7902); next is the artificial neural network (85.15%, 0.7700),
followed by the random forest (84.08%, 0.7559) and finally the maximum li-
kelihood (78.51%, 0.6668). The final LULC map produced under this study

DOI: 10.4236/gep.2022.1011018 Nov. 30, 2022 265 Journal of Geoscience and Environment Protection
E. Donkor et al.

can be used to monitor cocoa driven deforestation especially in the gazetted

forest and game reserves. This map will also be very useful in the national
forest monitoring framework for the REDD + cocoa landscape project.

Keywords
Support Vector Machine, Random Forest, Artificial Neural Network,
Maximum Likelihood, Image Classification, Cocoa Landscape

1. Introduction
Remote sensing has been an important source of land cover data during the last
three decades (Foody et al., 2004). Improvement in satellite technology has made
it possible to acquire land cover information over wide areas at varying spatial,
spectral and radiometric resolutions (Hopkins et al., 1988).
The maximum likelihood, minimum distance to mean, parallelepiped, Maha-
lanobis distance and the box classifier are some of the traditional statistics-based
classifiers used in remote sensing (Yu et al., 2014).
Machine learning methods, such as Support Vector Machine (SVM), Random
Forest (RF), Decision Trees (DTs), Artificial Neural Network (ANN) and K-nearest
neighbours (K-NN), have become common image classifiers as technology has
evolved. Some research works have been done to compare the machine learning
algorithms with traditional statistical classifiers, and they have been found to
improve classification accuracy (Rogan et al., 2002). Traditional statistical clas-
sifiers are parametric algorithms. The major limitation of parametric classifiers
is their reliance on the data statistical distribution. Also, they have low accuracy
for image classification, whereas non parametric classifiers which are machine
learning methods do not depend on data assigned to any specific statistical dis-
tribution (Caetano, 2009; Mountrakis et al., 2011).
Maximum likelihood (ML), Random forest (RF), Support vector machine (SVM)
and Artificial neural network (ANN) classifiers have been chosen for this study,
because they are extensively used in image classification (Zagajewski et al., 2021;
Saeed et al., 2015).
Maximum likelihood (ML) is one of the simplest but commonly used statis-
tical classification algorithm, in which a pixel with the maximum likelihood is
classified into the corresponding class (Saeed et al., 2015). There are several rea-
sons why the maximum likelihood classifier is so popular; first, the maximum
likelihood decision rule is naturally appealing, because the most likely outcome
among candidates is chosen (Bolstad & Lillesand, 1991). Additionally, covarying
data, this frequently occurrence with satellite image data, can be easily accom-
modated by maximum likelihood classification. Finally, it has been demonstrat-
ed that maximum likelihood classifier, which takes variability into account, per-
forms well across a variety of cover types (Lillesand & Kiefer, 1987). A study was
conducted by comparing maximum likelihood, support vector machines and

DOI: 10.4236/gep.2022.1011018 266 Journal of Geoscience and Environment Protection

 E. Donkor et al.

random forest techniques using RapidEye image for land cover mapping in the
municipality of San Pelayo of Colombian Caribbean. It was found that, support
vector machines produced the highest classification accuracy of 81.32%, followed
by random forest 78.92% and finally maximum likelihood 68.95%. Though max-
imum likelihood produced the least classification accuracy, it was able to cor-
rectly classified infrastructure which was one of the classification classes better
than the other two techniques (Valero et al., 2019) and this could be the ability
of maximum likelihood to consider variability.
Random forest (RF) is one of the most widely used machine learning algo-
rithms (Breiman, 2001). This algorithm is appealing since it is used for both
classification and regression tasks (Woznicki et al., 2019). It is easy to use, effi-
cient and accurate (Meltzer, 2021). Due to its versatility, RF has been applied in
a variety of Earth scientific applications, such as modeling land-use (Araki &
Yamamoto, 2018), land-cover (Nitze & Cawkwell, 2015) and modeling forest
cover (Betts et al., 2017). Rodriguez-Galiano et al. (2012) examined RF to deci-
sion trees and found that RF provided a high accuracy of 92%, outperforming
decision trees of accuracy of 83%. The ensemble architecture of RF, which trains
multiple decision trees on different subsets of the training data, is thought to
account for its improved accuracy.
Support vector machine (SVM) has been shown to outperform other classifi-
ers due to its overall high capacity to simplify complex features (Shao & Lunetta,
2012). Support vector machine was able to achieve high overall accuracy of 88%
in a land cover classification utilizing Landsat-8 and using six land-cover classes
(Goodin et al., 2015). In order to map paddy rice in China in 2015, Mansaray et
al. (2019) examined the effect of training sample size on the overall accuracies of
SVM and RF. It was discovered that SVM and RF achieved overall accuracies of
91.8% and 89.2%, respectively.
Artificial neural network (ANN) has become a popular tool in the analysis of
remotely sensed data (Mas & Flores, 2008). The ability of ANN to learn on its
own and handle complicated issues is one of the reasons it has grown so popular
(Di Franco & Santurro, 2021). Artificial neural network has been used in several
land cover classification studies including using ANN, SVM and ML with IKONOS
image for land cover mapping in Shahriar city of Iran. The classification results
showed that, the overall accuracy and kappa coefficient of ANN (87.75%, 0.820)
was better than that of SVM (85.57%, 0.819) and ML (78.36%, 0.729) (Saeed et
al., 2015). Also comparing classification results of neural network called back
propagation neural (BPN) and extended delta bar delta (EDBD) network with
parallelepiped, minimum distance and maximum likelihood using Landsat 8 to
classify land cover types in Minnesota of United States of America. The classifi-
cation results revealed that the neural network performed best among the clas-
sifiers with overall accuracy and kappa of 95.07%, 0.935 respectively, followed by
maximum likelihood (90.77%, 0.882), minimum distance (84.24%, 0.803), paral-
lelepiped (69.23%, 0.612) (Zhang & Chang, 2015).
Maximum Likelihood (ML) is a supervised classification algorithm which is

DOI: 10.4236/gep.2022.1011018 267 Journal of Geoscience and Environment Protection

E. Donkor et al.

based on the Bayes theorem, assumes the reflectance values for each class in each
band is normally distributed. During the ML classification, a given pixel has a
probability that belongs to a particular class. As a result, the discriminant func-
tion is used to calculate each pixel’s probability, and each pixel is then allotted to
the class with the highest probability (Kulkarni, 2016). ML classifier has shown
to perform effectively across a variety of land cover types as it takes variability
into accounts (Lillesand & Kiefer, 1987).
Random Forest (RF) is an ensemble classifier, which means “union of parts”.
Random Forest uses more decision trees and makes prediction from each deci-
sion tree and selects the best outcome by means of voting (Breiman, 2001). One-
third of the samples, known as the out-of-bag (OOB) samples, are excluded at
random from each new training set that is created to help the tree grow. The tree
is constructed using the remaining samples in the bag. The model performance
can be evaluated using the OOB samples (Nguyen et al., 2015). Random forest is
very flexible, has very high accuracy and also works better than a single decision
tree. It does not suffer from the over fitting problem (Breiman, 2001).
Support Vector Machine (SVM) idea was developed by Cortes and Vapnik in
1995, which is a supervised learning method usually utilized in remote sensing
applications. The main aim of SVM is to find the best hyperplane that divides
the training data into several groups (Mountrakis et al., 2011).
Originally, Support Vector Machine (SVM) was to identify a linear class boun-
dary. In order to overcome this restriction, the feature space was projected to a
higher dimension on the grounds that a linear boundary might be present in a
higher dimensional feature space. This projection to a higher dimensionality is
known as the kernel trick. Kernel increases the number of dimensions in non-
separable issues to make them separable. As a result of this, SVM becomes more
powerful, adaptable and precise (Maxwell et al., 2018).
Artificial Neural Network works like the human brain and the building blocks
are neurons. Each neuron has synaptic weights, which are specific coefficients
that link it to other neurons. During training, information is sent to these join-
ing points (Mijwil, 2018). Artificial Neural Network can learn complex configu-
rations, taking into consideration any nonlinear complex relationship between
the independent and the dependent variables (Jamali, 2021).
The High Forest Zone (HFZ) of Ghana, which contains the cocoa landscape,
comprises 8.2 million hectares amounting to 34% of the country’s total land
area, with vegetation varying from wet evergreen to dry semi-deciduous (Fore-
stry Commission, 2016; Indufor, 2015). Ghana’s HFZ is made up of a complex
web of forest, cocoa farms, croplands and human settlements (National REDD+
Secretariat, Forestry Commission, 2017). Implementing forest monitoring sys-
tems at the landscape level forms part of the HFZ’s climate-smart and sustaina-
ble landscape activities. As prescribed by the Intergovernmental Panel on Cli-
mate Change, wall-to-wall mapping is essential for the proper execution of these
forest monitoring systems (Mitchell et al., 2017).
Land cover maps that are precise and current are essential for environmental

DOI: 10.4236/gep.2022.1011018 268 Journal of Geoscience and Environment Protection

 E. Donkor et al.

research, climate change monitoring and natural resource management (Pelleti-

er et al., 2019). The cocoa landscape of Ghana is complex and diverse in nature,
composing of mixture of closed forest, open forest settlements, croplands and
cocoa farms making the mapping of the landscape very difficult. Some studies
have been carried out in mapping the cocoa landscape of Ghana using one clas-
sification algorithm for each study (Benefoh et al., 2018; Ashiagbor et al., 2020).
However, other classification algorithms for mapping the cocoa landscape of
Ghana have not been fully explored. Hence there is the need to explore the per-
formances of other image classifiers to know which algorithm is best suited for
the classification of the cocoa landscape of Ghana as well as other countries with
similar cocoa landscapes like Ghana. Benefoh et al. (2018) used Landsat 8 optical
dataset applying image-fusion on vegetation indices (VIs) and digital elevation
model (DEM) using maximum likelihood algorithm to detect and distinguish
cocoa plantation from forest and other land use classes in the Krokosua Hills
Forest Reserve catchment of Ghana. Also, in the Juaboso-Bia cocoa landscape of
Ghana, Ashiagbor et al. (2020) used Sentinel-1 and Sentinel-2 satellite images to
map mono cocoa, cocoa agroforestry, forest lands and other land use classes us-
ing random forest classifier.
The aim of this research is to assess and compare the performances of SVM,
ANN, RF and ML classifiers to know which classifier is best suited for mapping
the cocoa landscape of Ghana using the Juaboso and Bia West districts of Ghana
as the study area.

2. Materials and Methods

2.1. Study Area
The study was carried out in Juaboso and Bia West districts in the Western
North region of Ghana. The Western North region is the leading cocoa produc-
ing region in Ghana. Juaboso and Bia West districts are among the highest cocoa
producing districts in the Western North region of Ghana. The study area is si-
tuated between latitude 6˚13'N to 6˚50'N and longitude 2˚40'W to 3˚16'W, cov-
ering an area of 2571.26 square kilometres or 257,126 hectares (Figure 1).
With a mean annual temperature between 25.5˚C and 26.5˚C, the area has a
tropical climate marked by warm temperatures. The annual rainfall levels are
between the ranges of 1250 - 2000 mm with June and October as its peak months
(Ghana Statistical Service, 2014). The rainy and dry seasons are experienced
within the study area; the wet season is from April to October, while the dry
season lasts from November to March. Numerous food and commercial crops,
particularly cocoa, are favoured by the comparatively long rainy season (Ghana
Statistical Service, 2014). The elevation ranges between 137 - 594 m above sea
level with Krokosua Hills in North West - South West part with the rest of the
area on relatively lower elevation. The soils are primarily Oxysols and Ochrosols
with Birimian and Hornblende as the parent rocks (Ghana Statistical Service,
2014).

DOI: 10.4236/gep.2022.1011018 269 Journal of Geoscience and Environment Protection

E. Donkor et al.

Figure 1. Map of the study area showing the cocoa mosaic landscape of Juaboso and Bia
West districts of Ghana. (Image source: Landsat and Copernicus from goggle earth en-
gine).

The study area falls within Moist Evergreen (ME), Moist Semideciduous
North West (MSNW) and Moist Semideciduous South East (MSSE) subtypes
ecological zones (Hall & Swaine, 1981). There are two forest reserves and one
game reserve (protected area) in the study area. The forest reserves are Krokosua
Hills and a portion of Bia Tributaries North and Bia National Park (protected
area) all are under the administration of the Ghana Forestry Commission (Fore-
stry Commission, 2016). The rest of the area is covered by farmlands mostly co-
coa and communities in relatively low lying areas (Ghana Statistical Service,
2014).

DOI: 10.4236/gep.2022.1011018 270 Journal of Geoscience and Environment Protection

 E. Donkor et al.

2.2. Data Acquisition

Sentinel-2 images for 2020 with cloud cover of less than 10% were downloaded
from the Copernicus Open Access Hub (https://scihub.copernicus.eu/). Senti-
nel-2 images are delivered in tiles, with each tile bearing a distinct name and the
study area falls within tiles S2A-MSIL1C-20200105T103421-T30NVM, S2A-
MSIL1C-20200204T103221-T30NWN and S2A-MSIL1C-20201210T103429-
T30NVN. These images were downloaded as zip files and extracted into their
various bands with each image having 13 bands.
Field data for the classification of the image were collected using representa-
tive sampling method, where points were collected based on the dominance of
the land use/land cover types. At each location, the coordinates were picked us-
ing handheld GPS, the land use description at that point and the adjoining land
use were recorded on the field sheet.
Data were collected from the following land use/land cover classes; closed
forest, open forest, cocoa, settlement/bare surface and other vegetation (Table
1). In all, one thousand, two hundred and forty-six (1246) points were collected.
These were made up of; 133 for closed forest, 152 in open forest, 728 for cocoa,
121 in settlement/bare surface and 112 for other vegetation. These sample points
were divided at random into 869 which form 70% for classification and 377
which constitute 30% of the total sample points for validation.

2.3. Sentinel-2 Image Pre-Processing

The sentinel-2 images that were downloaded were in Level-1C processing format
had only undergone geometric and radiometric corrections but were not at-
mospherically corrected. The images were converted into Level-2A product and
atmospheric correction done using Sen2C or processor (Drusch et al., 2012;
Müller-Wilm, 2018). Ten (10) bands; 2, 3, 4, 5, 6, 7, 8, 8A, 11 and 12 were

Table 1. Description of LULC types for the image classification.

LULC Type Definition

Woody vegetation with a minimum mapping unit of 1 hectare

and a canopy cover greater than 60% at a height of 5 m.
Closed forest
Closed forest is mostly found in gazetted forest reserves and
national Park with a small portion in off reserve sacred groves.

Open forest areas are those that have a canopy cover within
Open forest 15% to 60%. The low canopy cover may be due to excessive timber
logging, mining and other environmental factor like bush fires.

Farmlands cultivated with cocoa. It includes cocoa farms without

Cocoa
trees (mono cocoa) and those with trees (cocoa agroforestry).

Include annual food crop farm, fallowland and other tree crops
Other Vegetation
like oil palm, citrus and rubber.

Settlement/Bare These include areas with no vegetation such as human settlements,

surface barren lands and mined-out areas.

DOI: 10.4236/gep.2022.1011018 271 Journal of Geoscience and Environment Protection

E. Donkor et al.

stacked for each tile to produce composite image. The stacked images were mo-
saicked to form one composite image. The study area shapefile was used to sub-
set the area of interest from the composite image and haze correction applied.

2.4. Supervised Image Classification and Accuracy Assessment

Supervised image classification was carried out using sentinel-2 imagery to de-
termine the LULC types of the research area. The classification was performed
using maximum likelihood, random forest, support vector machine and artificial
neural network algorithms in R software. Stacked sentinel-2 image, training data
and validation data in polygon shapefiles storing the identity for each land cover
type were imported into R. Caret, Rstoolbox, rgdal and raster packages were
imported into R for the classification. Sentinel-2 image was imported as raster
brick using the brick function in raster package.
Maximum Likelihood (ML) classification was done using “rasclass” package
in R to train and fit ML model. Random Forest (RF) classification was executed
by the “randomForest” package in R. Support Vector Machine (SVM) classifica-
tion was done using “e1071” package and artificial neural network classification
using “neuralnet” package in R.
Using the training data in combination with a classifier, the pixel values in the
training area for every band in sentinel-2 were extracted and stored in a data
frame with its corresponding LULC class ID (Table 2) to train and fit the model.
The classification was carried out separately for each classifier and saved the
output classified image file. The output classified images were filtered to remove
the speckles from the classified images to enhance their appearance. The final
maps were prepared and area for the landuse classes was calculated.
Accuracy assessment was performed for each classifier by using the classified
image and 377 validation points in R to generate the confusion matrix, the over-
all accuracy and the kappa. User’s Accuracy (UA) and Producer’s Accuracy (PA)
were calculated from the confusion matrix.

3. Results
3.1. Image Classification
Supervised classification was used to categorize the research area into five (5)

Table 2. Class ID and LULC name.

Class ID LULC Name

1 Closed forest

2 Open forest

3 Cocoa

4 Other Vegetation

5 Settlement/Bare surface

DOI: 10.4236/gep.2022.1011018 272 Journal of Geoscience and Environment Protection

 E. Donkor et al.

LULC classes; closed forest, open forest, cocoa, other vegetation and settle-
ment/bare surface using the four classifiers with combined maps shown in Fig-
ure 2.
Maximum likelihood classifier map shows five classes; closed forest, open for-
est, cocoa, other vegetation and settlement/bare surface. Other vegetation class
has a small class area, hence appearing patchy on the map. Random forest, sup-
port vector machine and artificial neural network classifiers maps display all the
five LULC classes well; this implies these classifiers separated all the classes well
under this study.
A summary of the LULC classes areas for the four classifiers is presented in
(Table 3) with its bar chart (Figure 3).

3.2. Accuracy Assessment

The accuracy assessment based on the classified images in R generated the con-
fusion matrix, overall accuracy and kappa.
The confusion matrix and the accuracy report for the four classifiers are pre-
sented in Table 4.

Figure 2. LULC maps of the study area using (a) ML; (b) RF; (c) SVM; (d) ANN.

DOI: 10.4236/gep.2022.1011018 273 Journal of Geoscience and Environment Protection

E. Donkor et al.

Table 3. LULC areas for each classifier.

CLASSIFIER ML RF SVM ANN

LULC Class Area (ha) Area (%) Area (ha) Area (%) Area (ha) Area (%) Area (ha) Area (%)

Closed Forest 57057.40 22.19 47314.91 18.40 48913.36 19.02 48010.46 18.67

Open Forest 43908.38 17.08 47958.84 18.65 43637.70 16.97 44070.35 17.14

Cocoa 143767.99 55.91 127943.8 49.76 135329.8 52.63 145006.76 56.40

Other Vegetation 7829.25 3.04 28645.41 11.14 24808.61 9.65 15127.17 5.88

Settlement/Bare Surface 4562.98 1.78 5262.99 2.05 4436.46 1.73 4911.26 1.91

TOTAL 257126 100 257126 100 257126 100 257126 100

Table 4. Confusion matrix and accuracy report for the four classifiers.

Closed Open Other Settlement/Bare PA UA

LULC Cocoa Total
Forest Forest Vegetation Surface (%) (%)

Maximum Likelihood Classifier

Closed Forest 36 20 6 1 0 63 81.82 57.14

Open Forest 8 30 14 1 0 53 60 56.6

Cocoa 0 0 184 16 9 209 89.32 88.04

Other Vegetation 0 0 2 19 4 25 51.35 76

Settlement/Bare Surface 0 0 0 0 27 27 67.5 100

Total 44 50 206 37 40 377

Overall Accuracy 78.51%

Kappa Statistics 0.6668

Random Forest Classifier

Closed Forest 34 10 0 1 0 45 77.27 75.56

Open Forest 10 38 0 0 0 48 70 79.17

Cocoa 0 2 188 9 5 204 91.26 92.16

Other Vegetation 0 0 17 27 5 49 72.97 55.1

Settlement/Bare Surface 0 0 1 0 30 31 75 96.77

Total 44 50 206 37 40 377

Overall Accuracy 84.08%

Kappa Statistics 0.7559

Support Vector Machine Classifier

Closed Forest 34 9 0 1 0 44 77.27 77.27

Open Forest 10 39 0 0 0 49 78 79.59

Cocoa 0 1 195 8 6 210 94.66 92.86

Other Vegetation 0 1 11 28 4 44 75.68 63.64

DOI: 10.4236/gep.2022.1011018 274 Journal of Geoscience and Environment Protection

 E. Donkor et al.

Continued

Settlement/Bare Surface 0 0 0 0 30 30 75 100

Total 44 50 206 37 40 377

Overall Accuracy 86.47%

Kappa Statistics 0.7902

Artificial Neural Network

Closed Forest 38 10 0 0 0 48 86.36 79.17

Open Forest 6 38 3 0 0 47 76.00 80.85

Cocoa 0 2 191 15 1 209 92.72 91.39

Other Vegetation 0 0 8 22 7 37 59.46 59.46

Settlement/Bare Surface 0 0 4 0 32 36 80.00 88.89

Total 44 50 206 37 40 377

Overall Accuracy 85.15%

Kappa Statistics 0.7700

Figure 3. LULC classes areas per classifier.

The highest overall accuracy is 86.47% for the SVM classifier, followed by
ANN (85.15%), RF (84.08%) and the least is ML (78.51%). In addition, the kappa
statistics of 0.7902 is the highest for SVM, next is ANN (0.7700), followed by RF
(0.7559) with ML having the least (0.6668).
The Kappa is the agreement between the model prediction and observed (Delgado
& Tibau, 2019). It provides a more accurate indicator of the overall performance
of the classifier. This is due to the possibility of a simple accuracy can be skewed
if the class distribution is also skewed. Van Ness et al. (2008) considered kappa
more than 0.75 as excellent and between 0.4 to 0.75 to be fair to good, hence
SVM, ANN and RF are excellent classifiers while ML is a good classifier per this
study.

DOI: 10.4236/gep.2022.1011018 275 Journal of Geoscience and Environment Protection

E. Donkor et al.

Overall accuracy is somewhat inadequate for summarizing the accuracy of

LULC classification. This is because the overall accuracy generated from the ac-
curacy assessment is an average value that does not tell whether the error was
evenly distributed across the LULC classes. Consequently, two other measures
are often used, which are the producer’s accuracy and the user’s accuracy. Pro-
ducer’s accuracy indicates for a given class the proportion of the reference data
that are classified correctly. User’s accuracy calculates for a given class how
many pixels are actually what the classification claims they are (Rwanga & Ndam-
buki, 2017). From (Table 4) the overall accuracy for ML classifier is 78.51% with
closed forest having 81.82% and 57.14% as PA and UA respectively. This means
that 81.82% of the closed forest area has been identified correctly with 57.14%
identified as truly closed forest from classification perspective. Using the ML
classifier the highly reliable LULC class associated with this classification is the
cocoa class from the producer’s accuracy and user’s accuracy viewpoint with PA
(89.32%) and UA (88.04%). Similarly using RF, SVM and ANN classifiers, the
classification was able to map the cocoa class very well based on the producer’s
accuracy and user’s accuracy with the highest in SVM (94.66% and 92.86%) for
PA and UA respectively.

4. Discussion
The major LULC in the study area is cocoa as obtain from all four (4) classifiers,
with the highest area obtained for ANN (56.40%) and the least in RF (49.76%).
After cocoa, closed forest is the next in terms of area coverage with the highest
area occurring in ML (22.19%) and the least in RF (18.40%). Open forest follows
with RF (18.65%) as the highest and SVM (16.97%) as the least in terms of area
coverage in this LULC class. Other vegetation follows with RF (11.14%) as the
highest with the least in ML (3.04%) as an area in this class. The smallest area is
the settlement/bare surface LULC class with RF having the highest area (2.05%)
and the least in SVM (1.73%).
Support vector machine produced the highest overall accuracy and kappa of
86.47%, 0.7902 respectively, followed by ANN (85.15%, 0.7700), RF (84.08%,
0.7559) and the least is ML (78.51%, 0.6668). Support vector machine has the
ability to handle minimal training data sets and usually produce higher classifi-
cation accuracy (Bouaziz et al., 2017). Khatami et al. (2016) revealed that SVM
was the best among numerous classifiers, including random forest, neural net-
work and decision trees.
Random forest and artificial neural network also performed very well and
their performances are closed to that of SVM. Each decision tree is constructed
with random forest using a subset of the features. This is favourable because
each decision tree may make a precise classification determination that is based
only on useful features and the decision trees perform voting to come out with
the final classification (Tian et al., 2016). Artificial Neural Network performs
supervised classification using small data and the ability to integrate multiple

DOI: 10.4236/gep.2022.1011018 276 Journal of Geoscience and Environment Protection

 E. Donkor et al.

types of data in the study, because there are no assumptions about the data used
(Mas & Flores, 2008).
Maximum likelihood classifier performance is fairly good, as it was able to se-
gregate closed forest, open forest, cocoa and settlement/bare surface LULC classes
with some misclassification in the other vegetation class. The inability of ML to
classify the other vegetation very well may be as a result of the mixed and com-
plex environment of the landscape. Parametric classifiers such as ML is not best
suited for complex systems (Mishra, 2018).
The classification results in a research earlier conducted by Benefoh et al.
(2018) in the Krokosua Hills forest reserve catchment of Ghana using maximum
likelihood method gave an overall accuracy of 82.6% and a kappa of 0.73. Also
Ashiagbor et al. (2020) classification results in the Juaboso-Bia cocoa landscape
of Ghana using sentinel-2 bands and its Vegetation Indices (VIs) with random
forest classifier produced overall accuracy of 79.02% and kappa of 0.748.
One drawback observed in this study is imbalanced classification, thus, where
the training dataset is biased or skewed towards a class or classes. Imbalanced
classifications present a challenge for predicting models because the majority of
machine learning algorithms for classification were built on the premise that
there should be an equal number of samples in each class. As a result, models
perform poorly when making predictions, especially for the class with small train-
ing samples (Browniee, 2019).
A total of 869 training samples were used for the classification, with closed
forest class constituting 10.70%, open forest 12.20%, cocoa 58.45%, settlement
9.67% and other vegetation 8.98%. From Figure 2, other vegetation class was not
well represented and this is due to the small number of training samples used for
the classification and cocoa class was visibly represented because more samples
of cocoa was used.

5. Conclusion and Recommendation

This research has demonstrated the comparative ability of SVM, ANN, RF and
ML classifiers to map the cocoa landscape of the Juaboso and Bia West districts
of Ghana. Based on overall classification accuracy, kappa statistics, producer’s
accuracy and user’s accuracy, SVM is the best among the four classifiers for map-
ping LULC categories of the study area. Hence SVM classifier map could be used
as the final LULC map for the study area. Also classification accuracy of the
SVM, ANN and RF methods were close to each other and higher than ML,
which indicates that non-parametric algorithms like machine learning tech-
niques can deliver more precise results than parametric algorithm like maximum
likelihood classifier. The reason is that, the method of the mapping function is
not assumed by non-parametric classifiers.
The final LULC map produced in this research provides useful information on
the area used for cocoa farming in the study area with 53% of the total land mass
under cocoa cultivation. The LULC map can be used to monitor cocoa driven

DOI: 10.4236/gep.2022.1011018 277 Journal of Geoscience and Environment Protection

E. Donkor et al.

deforestation especially in the gazetted forest and game reserves. This map will
also be very useful in the national forest monitoring framework for the REDD +
cocoa landscape project in Ghana.
It is recommended that, in future using machine learning algorithms to per-
form supervised image classification for complex ecosystems like the cocoa
landscape of Ghana, the training samples to be used should be almost the same
for each class in order minimize the problem of imbalanced classification.

Conflicts of Interest
The authors declare no conflict of interest with respect to the publication of this
paper.

Author Statement
Emmanuel Donkor (ED): Conceptualization, Investigation, Formal Analysis,
Writing-Original Draft, Writing-Review & Editing, Visualization.
Edward Matthew Osei Jnr (EMO): Writing-Original Draft, Writing-Review &
Editing.
Stephen Adu-Bredu (SAB): Formal Analysis, Writing-Original Draft, Writ-
ing-Review & Editing.
Samuel A, Andam-Akorful (SAAA): Formal Analysis, Writing-Original Draft,
Writing-Review & Editing, Supervision.
Efiba Vidda Senkyire Kwarteng (EVSK): Formal Analysis, Writing-Original
Draft, Writing-Review & Editing.
Lily Lisa Yevugah (LLY): Formal Analysis, Writing-Original Draft, Writ-
ing-Review & Editing.

References
Araki, S., Shima, M., & Yamamoto, K. (2018). Spatiotemporal Land Use Random Forest
Model for Estimating Metropolitan NO2 Exposure in Japan. Science of the Total Envi-
ronment, 634, 1269-1277. https://doi.org/10.1016/j.scitotenv.2018.03.324
Ashiagbor, G., Forkuo, E. K., Asante, W. A., Acheampong, E., Quaye-Ballard, J. A, Boa-
mah, P., Yakubu, M., & Foli, E. (2020). Pixel-Based and Object-Oriented Approaches
in Segregating Cocoa from Forest in the Juabeso-Bia Landscape of Ghana. Remote
Sensing Applications: Society and Environment, 19, Article ID: 100349.
https://doi.org/10.1016/j.rsase.2020.100349
Benefoh, D. T., Villamor, G. B., Van Noordwijk, M., Borgemeister, C., Asante, W. A., & Asu-
bonteng, K. O. (2018). Assessing Land-Use Typologies and Change Intensities in a Structu-
rally Complex Ghanaian Cocoa Landscape. Applied Geography, 99, 109-119.
https://doi.org/10.1016/j.apgeog.2018.07.027
Betts, M. G., Wolf, C., Ripple, W. J., Phalan, B., Millers, K. A., Adam, D., Butchart, S. H.
M., & Levi, T. (2017). Global Forest Loss Disproportionately Erodes Biodiversity in In-
tact Landscapes. Nature, 547, 441-444. https://doi.org/10.1038/nature23285
Bolstad, P. V., & Lillesand, T. M. (1991). Rapid Maximum Likelihood Classification.
Photogrammetric Engineering & Remote Sensing, 57, 67-74.
Bouaziz, M., Eisold, S., & Guermai, E. (2017). Semiautomatic Approach for Land Cover

DOI: 10.4236/gep.2022.1011018 278 Journal of Geoscience and Environment Protection

 E. Donkor et al.

Classification: A Remote Sensing Study for Arid Climate in Southeastern Tunisia. Eu-
ro-Mediterranean Journal for Environmental Integration, 2, Article No. 24.
https://doi.org/10.1007/s41207-017-0036-7
Breiman, L. (2001). Random Forests. Machine Learning, 45, 5-32.
https://doi.org/10.1023/A:1010933404324
Browniee, J. (2019). Python Machine Learning.
https://machinelearningmastery.com/hyperparameters-for-classification-machine-lear
ning-algorithms
Caetano, M. (2009). Image Classification. ESA Advanced Training Course on Land Re-
mote Sensing.
Cortes, C., & Vapnik, V. (1995). Support-Vector Networks. Machine Learning, 20, 273-297.
https://doi.org/10.1007/BF00994018
Delgado, R, & Tibau, X.-A. (2019). Why Cohen’s Kappa Should Be Avoided as Perfor-
mance Measure in Classification. PLOS ONE, 14, e0222916.
https://doi.org/10.1371/journal.pone.0222916
Di Franco, G., & Santurro, M. (2021). Machine Learning, Artificial Neural Networks and
Social Research. Quality & Quantity, 55, 1007-1025.
https://doi.org/10.1007/s11135-020-01037-y
Drusch, M., Del Bello, U., Carlier, S., Colin, O., Fernandez, V., Gascon, F., Hoersch, B. et
al. (2012). Sentinel-2: ESA’s Optical High-Resolution Mission for GMES Operational
Services. Remote Sensing of Environment, 120, 25-36.
https://doi.org/10.1016/j.rse.2011.11.026
Foody, M. G., & Mathur, A. (2004). Toward Intelligent Training of Supervised Image Clas-
sifications: Directing Training Data Acquisition for SVM Classification. Remote Sens-
ing of Environment, 93, 107-117. https://doi.org/10.1016/j.rse.2004.06.017
Forestry Commission (2016). Ghana REDD+ Strategy.
https://www.forestcarbonpartnership.org/sites/fcp/files/2015/April/Ghana%20National
%20REDD%2B%20Strategy%20Final.pdf
Ghana Statistical Service (2014). District Analytical Report: Juaboso and Bia West Dis-
tricts.
https://www2.statsghana.gov.gh/docfiles/2010_District_Report/Western/Juaboso.pdf
https://www2.statsghana.gov.gh/docfiles/2010_District_Report/Western/Bia%20West.p
df
Goodin, D. G., Anibas, K. L., & Bezymennyi, M. (2015). Mapping Land Cover and Land
Use from Object-Based Classification: An Example from a Complex Agricultural Land-
scape. International Journal of Remote Sensing, 36, 4702-4723.
https://doi.org/10.1080/01431161.2015.1088674
Hall, J. B., & Swaine, M. D. (1981). Distribution and Ecology of Vascular Plants in a
Tropical Rain Forest: Forest Vegetation in Ghana. In Geobotany (Vol. 1). Springer Na-
ture. https://doi.org/10.1007/978-94-009-8650-3
Hopkins, P. F., Maclean, A. L., & Lillesand, T. M. (1988). Assessment of Thematic Map-
per Imagery for Forestry Applications under Lakes States Conditions. Photogrammetry
Engineering & Remote Sensing, 54, 61-70.
Indufor, O. (2015). Development of Reference Emissions Levels and Measurement, Report-
ing and Verification System in Ghana.
https://www.forestcarbonpartnership.org/system/files/documents/Ghana%20MRV%20
Final%20Report%20%28ID%2067024%29.pdf
Jamali, A. (2021). Improving Land Use Land Cover Mapping of a Neural Network with
Three Optimizers of Multi-Verse Optimizer, Genetic Algorithm, and Derivative-Free
Function. The Egyptian Journal of Remote Sensing and Space Science, 24, 373-390.

DOI: 10.4236/gep.2022.1011018 279 Journal of Geoscience and Environment Protection

E. Donkor et al.

https://doi.org/10.1016/j.ejrs.2020.07.001
Khatami, R., Mountrakis, G., & Stehan, V. S. (2016). A Meta-Analysis of Remote Sensing
Research on Supervised Pixel-Based Land-Cover Image Classification Processes: Gen-
eral Guidelines for Practitioners and Future. Remote Sensing of Environment, 177,
89-100. https://doi.org/10.1016/j.rse.2016.02.028
Kulkarni, A. D. (2016). Random Forest Algorithm for Land Cover Classification. Interna-
tional Journal on Recent and Innovation Trends in Computing and Communication, 4,
2321-8169.
Lillesand, T. M., & Kiefer, R. W. (1987). Remote Sensing and Image Interpretation (2nd
ed.). John Wiley & Sons.
Mansaray, L. R., Wang, F., Huang, J., Yang, L., & Kanu, A. S. (2020). Accuracies of Sup-
port Vector Machine and Random Forest in Rice Mapping with Sentinel-1A, Landsat-8
and Sentinel-2A Datasets. Geocarto International, 35, 1088-1108.
https://doi.org/10.1080/10106049.2019.1568586
Mas, J. F., & Flores, J. J. (2008). The Application of Artificial Neural Networks to the
Analysis of Remotely Sensed Data. International Journal of Remote Sensing, 29, 617-663.
https://doi.org/10.1080/01431160701352154
Maxwell, A. E., Warner, T. A., Fang, F., Maxwell, A. E., & Warner, T. A. (2018). Imple-
mentation of Machine-Learning Classification in Remote Sensing: An Applied Review.
International Journal of Remote Sensing, 39, 2784-2817.
https://doi.org/10.1080/01431161.2018.1433343
Meltzer, R. (2021). What Is Random Forest?
https://careerfoundry.com/en/blog/data-analytics/what-is-random-forest
Mijwil, M. M. (2018). Artificial Neural Networks Advantages and Disadvantages. Univer-
sity of Baghdad.
Mishra, S. (2018). Parametric and Non-Parametric Methods.
https://www.linkedin.com/pulse/parametric-non-parametric-methods-satya-mishra
Mitchell, A. L., Rosenqvist, A., & Mora, B. (2017). Current Remote Sensing Approaches
to Monitoring Forest Degradation in Support of Countries Measurement, Reporting
and Verification (MRV) Systems for REDD+. Carbon Balance and Management, 12,
Article No. 9. https://doi.org/10.1186/s13021-017-0078-9
Mountrakis, G., Im, J., & Ogole, C. (2011). Support Vector Machines in Remote Sensing:
A Review. ISPRS Journal of Photogrammetry and Remote Sensing, 66, 247-259.
https://doi.org/10.1016/j.isprsjprs.2010.11.001
Müller-Wilm, U. (2018). S2 MPC: Sen2Cor Configuration and User Manual.
https://step.esa.int/thirdparties/sen2cor/2.5.5/docs/S2-PDGS-MPC-L2A-SUM-V2.5.5_
V2.pdf
National REDD+ Secretariat, & Forestry Commission (2017). Ghana’s National Forest
Reference Level.
https://redd.unfccc.int/files/ghana__modified_frl_november_10_2017_clean.pdf
Nguyen, T.-T., Huang, J. Z., & Nguyen, T. T. (2015). Unbiased Feature Selection in
Learning Random Forests for High-Dimensional Data. Research and Development of
Advanced Computing Technologies, 2015, Article ID: 471371.
https://doi.org/10.1155/2015/471371
Nitze, I., Barrett, B., & Cawkwell, F. (2015). Temporal Optimisation of Image Acquisition
for Land Cover Classification with Random Forest and MODIS Time-Series. Interna-
tional Journal of Applied Earth Observation and Geoinformation, 34, 136-146.
https://doi.org/10.1016/j.jag.2014.08.001

DOI: 10.4236/gep.2022.1011018 280 Journal of Geoscience and Environment Protection

 E. Donkor et al.

Pelletier, C., Webb, G. I., & Petitjean, F. (2019). Temporal Convolutional Neural Network
for the Classification of Satellite Image Time Series. Remote Sensing, 11, Article No.
523. https://doi.org/10.3390/rs11050523
Rodriguez-Galiano, V. F., Ghimire, B., Rogan, J., Chica-Olmo, M., & Rigol-Sanchez, J. P.
(2012). An Assessment of the Effectiveness of a Random Forest Classifier for Land-Cover
Classification. ISPRS Journal of Photogrammetry and Remote Sensing, 67, 93-104.
https://doi.org/10.1016/j.isprsjprs.2011.11.002
Rogan, J., Miller, J., Stow, D., Franklin, J., Levien, L., & Fischer, C. (2002). Land-Cover
Change Monitoring with Classification Trees Using Landsat TM and Ancillary Data.
Photogrammetric Engineering & Remote Sensing, 69, 793-804.
https://doi.org/10.14358/PERS.69.7.793
Rwanga, S. S., & Ndambuki, J. M. (2017). Accuracy Assessment of Land Use/Land Cover
Classification Using Remote Sensing and GIS. International Journal of Geosciences, 8,
611-622. https://doi.org/10.4236/ijg.2017.84033
Saeed, O., Hamid, E., & Ahmadi, F. F. (2015). Using Artificial Neural Network for Classi-
fication of High Resolution Remotely Sensed Images and Assessment of Its Perfor-
mance Compared with Statistical methods. American Journal of Engineering, Tech-
nology and Society, 2, 1-8.
Shao, Y., & Lunetta, R. S. (2012). Comparison of Support Vector Machine, Neural Net-
work, and CART Algorithms for the Land-Cover Classification Using Limited Training
Data Points. ISPRS Journal of Photogrammetry and Remote Sensing, 70, 78-87.
https://doi.org/10.1016/j.isprsjprs.2012.04.001
Tian, S., Zhang, X., Tian, J., & Sun, Q. (2016). Random Forest Classification of Wetland
Landcovers from Multi-Sensor Data in the Arid Region of Xinjiang, China. Remote
Sensing, 8, Article No. 954. https://doi.org/10.3390/rs8110954
Valero Medina, J. A., & Alzate Atehortúa, B. E. (2019). Comparison of Maximum Like-
lihood, Support Vector Machines, and Random Forest Techniques in Satellite Images
Classification. Tecnura, 23, 13-26. https://doi.org/10.14483/22487638.14826
Van Ness, P. H., Towle, V. R., & Juthani-Mehta, M. (2008). Testing Measurement Relia-
bility in Older Populations: Methods for Informed Discrimination in Instrument Selec-
tion and Application. Journal of Aging and Health, 20, 183-197.
https://doi.org/10.1177/0898264307310448
Woznicki, S. A., Baynes, J., Panlasigui, S., Mehaffey, M., & Neale, A. (2019). Development
of a Spatially Complete Floodplain Map of the Conterminous United States Using
Random Forest. Science of the Total Environment, 647, 942-953.
https://doi.org/10.1016/j.scitotenv.2018.07.353
Yu, L., Liang, L., Wang, J., Zhao, Y., Cheng, Q., Hu, L. et al. (2014). Meta-Discoveries
from a Synthesis of Satellite-Based Land-Cover Mapping Research. International Jour-
nal of Remote Sensing, 35, 4573-4588. https://doi.org/10.1080/01431161.2014.930206
Zagajewski, B., Kluczek, M., Raczko, E., Njegovec, A., Dabija, A., & Kycko, M. (2021).
Comparison of Random Forest, Support Vector Machines, and Neural Networks for
Post-Disaster Forest Species Mapping of the Krkonoše/Karkonosze Transboundary Bi-
osphere Reserve. Remote Sensing, 13, Article No. 2581.
https://doi.org/10.3390/rs13132581
Zhang, S. L., & Chang, T. C. (2015). A Study of Image Classification of Remote Sensing
Based on Back-Propagation Neural Network with Extended Delta Bar Delta. Mathe-
matical Problems in Engineering, 2015, Article ID: 178598.
https://doi.org/10.1155/2015/178598

DOI: 10.4236/gep.2022.1011018 281 Journal of Geoscience and Environment Protection