INFS 5095 – Big Data Basics (SP2 2021)

Assignment 1 – Technology Review

[Fadel Thariq Gifari]

[10/04/2021]

Page 1 of 9
Contents
Introduction. 3

Overview of Agriculture. 3

Overview of Convolutional Neural Network. 4

Limitations and Issues to Using Convolutional Neural Network. 7

Conclusion. 7

References 8

Page 2 of 9
Introduction.
Technology has been prevalent in the current age of mankind, ranging from inner
implementation and outer implementation.Initially all data is acquired via its respective
technology.In big data there are a multitude of technologies that can be utilized to gather and
distribute the data related to the industry via techniques from each technology.In this review, the
main focus of implementation is of Artificial Intelligence, that are categorized as analytical big data
technology, that has an overview of the annual statistics revolving different aspects that are indulged
in the workings of Agriculture.Currently the production of newly developed AI is increasing in a
structured manner, ranging from Reactive Machin AI to Self aware AI.
For Reactive Machine AI primarily responds to inferred data from the researchers while also
conducting pre - defined tasks, which defines their one dimensional characteristics.Past research
data is utilized by Limited Memory AI that read certain critical parts which will further improve future
related processes For Example: the AI of a self driving car will investigate the past GPS data in order
to route the car correctly, while also avoiding obstacles along the way For psychology purposes, the
Theory of Mind AI is better suited due to its emotional analytical capabilities, that allows for
researchers to expand on its territory.

Overview of Agriculture.
Agriculture is an everlasting industry that is essential in the survival of humans in its
entirety,it also generates products that will decrease the amount of poverty that exists within this
world. According to Worldbank.org Statistics infer that “25% of GDP” is of agriculture in countries
that are in the “developing” phase.There are several factors that will result in the ramification of
Agriculture Produce, including the main factors which are : weather and climate change,viability of
the land,the level of synchronization the chosen crops has with the soil, to simulate all the data
acquired is proven to be difficult, but quite detrimental to the outcome for the Industry, due to the
instability of the factors listed beforehand.For example the weather change preceded by Climate
change, has the possibility to result in a flooding over the preceding soil for the crops, resulting in a
growth impairment for a substantial amount of the crops. In today's age, the most commonly
produced crops are whole wheat and coarse wheat products.

Page 3 of 9
 Figure 1.1
The diagram above infers that human activities involving the prominence of outside
intervention towards oxygen will result in the production of Reactive Oxygen Species, which
increases the oxidative stress for each analyzed crop.its possible to generalize That aspect of
Agriculture via Convolutional Neural Network.

Overview of Convolutional Neural Network.

The simplest of Convolutional Neural network revolves around using tensorflow in order to
detect and classify certain images, utilizing hidden layers between the input and output layer to
produce a filter.The layers itself includes aspects that uphold its structure.To create the hidden layer,
a kernel size of 1x1,2x2,3x3,4x4 for Small images and 5x5 for larger images has to be set in order to
properly set the detection parameters for the inputted picture for further accuracy in detecting the
anomalies within each picture,after setting the Kernel Size a padding is required to Uniformize the
pixel existing around and within the picture.The last one is Stride that mainly decreases the amount
of pixel implemented, while also reducing the amount of time require to finish the training
data,finally a Pooling Layer to lower the size of the image in order for the image to be used in future
layers to reduce the discrepancy between the image and the layer.

Figure 1.2

Page 4 of 9
Using Convolutional Neural Network Methods in Agriculture.
For Agriculture, using the Convolutional Neural Network for its economic and practical
analysis is highly intensified in the industry. On the practical side, Convolutional Neural Network is
able to detect the anomalies within a specific crop field and segment all the desired results, in order
to classify the percentage of each factor revolving the practice of agriculture, specifically to identify
the components that increase or decrease the overall growth of a crop field.For example in a
research conducted by Cambridge University
A set of Aerial Photos were taken of an abandoned Plantation Field, using a training data
ratio of 85:15 , it was possible to remotely label and differentiate SugarCane from Soil on the
plantation field, while the miscellaneous items were labelled as others.

Figure 1.3
From the result, it can be deduced that the matrix displays labels in an orderly manner,Data as such
can be applied in researching an adaptable and feasible land, suitable for massively produced
crops(Kamilaris,2018).

There is a segmentation between pictures to test the overall accuracy and loss rate of the
CNN model, it is highly admissible to create a program that has all of the attributes for a CNN and a
sample dataset unrelated to the topic to avoid any baseline errors, in addition, multiple parameters
that were defined including sigmoid and feedforward functions that increases the overall efficiency
of the model before training it iteratively, by defining each parameter individually, it will allow the
program to predict the overall accuracy of the model when presented with 3 classes containing 100
images each.While also using a form of kernel that allows each image to be separated individually.
With the goal of preventing errors during classification. In Essence,represents the CNN model’s
epoch compilations.From these results it can be summarized that the cnn model is able to procure
the proper results from this sample datasets to be used for further research, and proven to be viable

Page 5 of 9
for image classification due to its low epoch loss and high accuracy while also maintaining its original
model without any drawbacks. The reason this model was used is due to the similarity of factors to
be detected within that of weather forecast images, including the air currents, heat precipitation,
and rain precipitation that plays a role in the overall growth of the crops. According to a research
done by Dong- a University that focused on the contents within the crops, mainly Tomatoes, to
determine nutrient deficiency within the crops.

Figure 2.1
The data above indicates the deficiency between the three micronutrients are prevalent in
the tomatoes,by adding multiple Convolutional layers , in supervised learning the data has been
labeled for each form of tomato, to indicate its deficiency.According to the researchers, in order to
set the proper boundaries for the Image Segmentation to successfully Operate,first it is required to
“tune the Inception‐ResNet v2 model based on the CNN architecture” (Tran,2019,p.6),followed by
the implementation of normalization in order to display the vanishing gradient prominent within the
specified images.The objects that are displayed within the image are the Tomato fruit and Leaves.

Figure 2.2
For the experimental results, the images showed that the nitrogen deficiency is detected
only in the tomato leaves, while calcium and other nutrients are found on the Tomato Fruit itself with
the accuracy resulting in 87.27 % , with each of the values producing its own individual maximum
value of 0.989,0.999, and 1.0 (Tran,2019,). The results of the supervised learning proved to be
substantially accurate, all the data acquired from the Tomato Research is viably utilized by farmers, to
determine what type of tomato seed is preferable compared to the other existing ones,from this it
will be possible to set the tomato seeds so that it will result in a hybrid produce, that will improve
and avoid any deficiency evolving within the tomatoes.

Page 6 of 9
Limitations and Issues to Using Convolutional Neural Network.

In using Convolutional Neural Network for Agriculture, there are multiple factors that hinders
the overall result of every case of this method. One being the error rate for each image
classification,which can result in the misclassification of certain objects within the image.That alone
will increase the possibility of a fallacy within the acquired data that the respective stakeholders will
take into account.
Another restriction that exist when using this method is a multitude of training data is
required in order for the model to conclude a result,this particular aspect of Convolutional Neural
Network will increase the time required to create a solution regarding a Agriculture Related Problem
For Example: Dynamical Structure of the land, The Elements that Construct a specific soil, Crop
acclimation to its environment,etc.
All of the hindrances mentioned are prevalent in creating and validating a set Dataset that
are involved in deciding the viability of a land for Crop fields and for certain cattles,therefore it is
important to note that using Convolutional Neural Network will impact the Agricultural Processes
both Positive and Negatively

Conclusion.
It can be concluded that the baseline for Convolutional methods stem from the overall
productivity of a crop field or other Agricultural Related Businesses.In essence this method has the
ability to segment the crucial aspects of Agriculture in Image Classification in order to gather the
necessary data that are vital for the advancement of the Agricultural Industry, that increases the
percentage of profits obtained by each stakeholder involved in operating the Industry.

Page 7 of 9
 References

Bank, W., 2021. Overview. [online] World Image Patches on Small Datasets. Applied
Bank. Available at: Sciences, 9(21), p.4500.
<https://www.worldbank.org/en/topic/agricu
lture/overview>[Accessed 18 April 2021]. Sharma, M. et al., 2017. Effects of Reactive
Oxygen Species on Crop Productivity.
Chui, M. et al., 2019. Notes from the AI Reactive Oxygen Species in Plants,
frontier: Applications and value of deep pp.117–136.
learning. McKinsey & Company. Available
at: Tran, T.-T. et al., 2019. A Comparative Study
<https://www.mckinsey.com/featured-insig of Deep CNN in Forecasting and
hts/artificial-intelligence/notes-from-the-ai-f Classifying the Macronutrient Deficiencies
rontier-applications-and-value-of-deep-lear on Development of Tomato Plant. Applied
ning> [Accessed April 18, 2021]. Sciences, 9(8), p.1-19.

Jan, B. et al., 2017. Deep learning in big

data Analytics: A comparative study.
Computers & Electrical Engineering.
Available at:
<https://www.sciencedirect.com/science/art
icle/pii/S0045790617315835> [Accessed
April 18, 2021].

Kamilaris, A. & Prenafeta-Boldú, F.X., 2018.

A review of the use of convolutional neural
networks in agriculture. The Journal of
Agricultural Science, 156(3), pp.312–322.

Kamilaris, A., Kartakoullis, A. and

Prenafeta-Boldú, F., 2017. A review on the
practice of big data analysis in agriculture.
[online]
Www-sciencedirect-com.access.library.uni
sa.edu.au. Available at:
<https://www-sciencedirect-com.access.lib
rary.unisa.edu.au/science/article/pii/S01681
69917301230?via%3Dihub>[Accessed 18
April 2021].

Phung & Rhee, 2019. A High-Accuracy

Model Average Ensemble of Convolutional
Neural Networks for Classification of Cloud

Page 8 of 9
Page 9 of 9