CHAPTER ONE

INTRODUCTION

1. 1. Background of the Study

Agriculture has become than simply a means to feed the ever-growing populations of the world and citrus diseases are threatening the livelihood of this important source. Citrus diseases have cause major production and economic loss in agriculture and forestry in recent times. Therefore, early detection and identification of citrus diseases plays the utmost important role to take timely measures (Sankaran et al., 2010).

 The prevention and control of plant diseases have always been widely discussed because plants are exposed to outer environment and are highly prone to diseases. Normally, the accurate and rapid diagnosis of diseases plays an important role in controlling plant disease, since useful protection measures are often implemented after correct diagnosis (Guiling, 2018). Various efforts have been developed to prevent crop loss due to diseases. Historical approaches of widespread application of pesticides have in the past decade increasingly been supplemented by integrated pest management (IPM) approaches (Ehler, 2006).

There are several ways to detect plant diseases. Some diseases do not have any visible symptoms associated with them, and those which appear only do so when it is too late to act. In these cases, it is necessary to perform sophisticated analysis, usually by means of powerful microscopes. In some cases, the signs can only be detected in parts of the electromagnetic spectrum that are not visible to humans (Barbedo, 2013). Most diseases, however, generate some kind of manifestation in the visible spectrum. The diseases may exhibit symptoms on different parts of the plant, i.e. leaves, stem, fruits/seeds etc. Independent of the approach, identifying a disease correctly when it first appears is a crucial step for efficient disease management. Historically, disease identification has been supported by agricultural extension organizations or other institutions such as local plant clinics. In more recent times, such efforts have additionally, been supported by providing information for disease diagnosis online, leveraging the increasing internet penetration worldwide. Even more recently, tools based on mobile phones have proliferated, taking advantage of the historically unparalleled rapid uptake of mobile phone technology in all parts of the world. Smartphones in particular offer very novel approaches to help identify diseases because of their tremendous computing power, high-resolution displays and extensive built-in sets of accessories such as advanced High Definition (HD) cameras. It is widely estimated that there will be between 5 and 6 billion smart-phones on the globe by 2020. At the end of 2015, already 69% of the world’s population had access to mobile broad-band coverage and mobile broadband penetration reached 47% in 2015, a 12-fold increase since 2007 (Harvey et al., 2017).

1. 2. Statement of the Problem

Citrus is a major plant grown mainly in Nigeria and other tropical areas of the world due to its richness in vitamin C and other important nutrients. The production of the citrus fruit has been widely affected by various diseases which ultimately degraded the quality of citrus fruits and also causes financial loss to the growers. Hence, to enhance the productivity of the citrus fruit, there is need for a Convolutional Neural Networks based approach for citrus diseases detection using mobile devices to enable the early detection of citrus fruit diseases for prompt treatment, to foster better and qualitative citrus fruits yields.

1.3     Aim and Objectives of the Study  

The aim of this study is to develop a citrus disease detection system. The objectives of the study are as follows;

1. To Design a convolution neural network (CNN) model for citrus disease detection.
2. To Implement the system in (i) using java programming.
3. Deploy the system in (ii) on an android mobile device.
4. To perform functional test on the system deployed in (iii).

1.4     Significance of the Study

Crop diseases remain a major threat to food supply world-wide. This study demonstrates the technical feasibility of a deep learning approach to enable automatic disease detection through image recognition. This study can support an accurate detection of orange fruit with little computational effort. It will also minimize the time taken in detecting citrus plant diseases by Agricultural Extension Organizations (AEO), local farmers and other institutions such as local plant clinics. Hence, it will help farmers in deciding how to intervene and foster solution to stop or minimize the effects of disease on citrus plant. This system if implemented will enhance the productivity in citrus plant and equally boost the Agricultural Sector in Nigeria.

 1.5 Scope of the Study

This research work covers the development of a convolutional neural network based citrus disease detection software with special focus on citrus diseases in orange plants in Makurdi, Benue state.

1.6   Definition of Terms

AEO:             Agricultural Extension Organizations

Artificial Intelligence: It is an intelligence demonstrated by machines, in contrast to the natural intelligence displayed by humans and animals.

Artificial Neural Networks:    These are computing systems that are inspired by, but not necessarily identical to biological neural networks, that constitute animal brain. Such systems learn to perform task by considering examples, generally with being programmed with any task or specific rules.

Citrus Disease:     This refers to abnormal condition in citrus that impairs its physical functions, associated with specific symptoms and signs.

Citrus:    It is a tree or a genus that includes citron, lemon, lime, orange and grape fruit.

Convolutional Neural Network (CNN):   Is a class of Deep Neural Networks, most commonly applied to analyzing visual imagery.

Data Argumentation:   Is use in deep learning to regularize, normalize the information on data to be trained.

Data Notation:    This is the process of labeling data to make it usable for machine learning.

Dataset: A collection of related sets of information that is composed of separate elements but can be manipulated as a unit by a computer.

Deep Learning: It is a part of broader family of machine learning model base on artificial neural networks.

Disease Detection:    It is a process of determining which disease or condition explains a tree/persons symptoms and sign. It is most often referred to as diagnosis with the medical context.

Disease Diagnosis:   It is process of determining which disease or condition explains a tree/persons symptom and signs.

Image:     This refers to the optical counterpart of an object produced by an optical device (such as a lens or mirror) or an electronic device.

Machine Learning: It is a scientific study of algorithms and statistical model that computer system uses in other to perform a specific task effectively, without using an explicit instruction relying on patterns and inference instead. It is seen as subset of artificial intelligence.

Mobile App:  A mobile app or mobile application is a computer program or software application designed to run on a mobile device such as a phone-tablet or watch.

 Model: Is a description or analogy used to help visualize something that cannot be directly observed.

Multilayer Perceptron: This refers to fully connected networks, that is, each neuron in one layer is connected to all neurons in the next layer.

Pathogen:   It is a bacterium, virus, or other microorganism that can cause disease.

Perceptron: This refers to a linear classifier (binary) use in supervised learning. It helps to classify the given input data.

Pest:  Is a destructive insects or other animal that attacks crops, food, livestock etc.

Phytosanitary:   This refers to measures for the control of plant diseases especially in agricultural crops.

Pollinators: It is an animal that moves pollen from the male anther of flower to female stigma of a flower

Software:   It is a program and other operating information use by a computer system.

Tensor-Flow: is a free and open-source software library for dataflow and differentiable programming across a range of tasks. It is a symbolic math library, and is also used for machine learning applications such as neural networks.

Weka: also known as Waikato Environment for Knowledge Analysis is a collection of machine learning algorithms for data mining tasks. The algorithms can either be applied directly to a dataset or called from your own Java code. Weka contains tools for data pre-processing, classification, regression, clustering, association rules, and visualization.