design and implementation of a 2.4 ghz embedded rfid...

DESIGN AND IMPLEMENTATION OF A 2.4 GHZ EMBEDDED RFID

SYSTEM FOR FRUIT MATURITY EVALUATION FOR AGRICULTURAL

INDUSTRIES UTILIZING WIRELESS MESH SENSOR NETWORK

AZLINA BINTI OTHMAN

A project report submitted in partial fulfillment of the requirements for the award of

Master of Electrical Engineering (Telecommunication)

Faculty of Electric and Electronic Engineering

Universiti Tun Hussein Onn Malaysia

Jan 2018

iv

Special dedication to my loving father Othman Bin Yusuf , my mother Siti

Sareah Binti Japri, my loving husband Abd Jalil Bin Elias and my beloved

daughters son Qaisara. Qayyim and Qasrina, my kind hearted supervisor Dr

Farhana Binti Ahmad Poad, and my dearest friends.

v

ACKNOWLEDGEMENT

My sincerest thanks go to my supervisor, Dr. Farhana Binti Ahmad Poad for

her dedication to her students and patience in assisting me with this thesis. I

appreciate her valuable advice and efforts offered during the course of my studies. I

would also like to thank to Pm. Madya Dr Afandi Bin Ahmad lecturer subjects for

Research Methodology for their equally valuable support generously given during

do my thesis.

Special thanks go to my loving husband Abd Jalil bin Elias and my dearest

son Qaisara, Qayyim and Qasrina for giving me of moral support. I appreciate their

encouragement and sympathetic help which made my life easier and more pleasant

during graduate studies. Lastly, I would like to thank my parents for their enormous

encouragement and assistance, for without them, this work would not have been

possible. Last but not least, my special thanks to those who have directly or

indirectly contributed to the success of my project.

vi

ABSTRACT

Papaya is one of the major fruits exported as well as favorite fruit among all

ages. However, it is difficult for fruit sellers or consumers to identify the best quality of

papaya and to detect the amount of mature papaya in a short period of time. Therefore, a

simple system is proposed to determine the fruit maturity by utilizing active Radio

Frequency Identification (RFID) and Wireless Sensor Network (WSN). The main

objective of this project is to determine the fruit maturity. An algorithm based on

MATLAB has been developed using image processing toolbox to identify the color of

the fruit. The color is analyzed by setting the average color, which must be over 400

pixels. Based on the results, the papaya can be categorized as mature when the color

value exceeds 400 pixels and immature when the color value is less than 400 pixels. The

accuracy of the proposed system is about 80% of which 8 of 10 papaya are able to give

good prediction. To determine the amount of mature fruit within a short period of time,

RFID system has been applied by using Xbee Pro S2b as a medium of transmission.

Several tests have been conducted to evaluate performance of the proposed system

which are the data transmission in the indoor location and outdoor location. In terms of

time and distance, it is proven that when the distance exceeds 100m, the transmission is

weak or no longer available. It can be concluded that, by combining image processing

method with RFID technology, the maturity of fruits can be easily obtain at anywhere

with reliable results.

vii

ABSTRAK

Betik merupakan salah satu buah utama negara yang dieksport ke luar negara dan

juga buah kegemaran di kalangan semua peringkat umur. Walaubagaimanapun adalah

sukar untuk penjual buah-buahan atau pengguna untuk mengenalpasti kualiti betik yang

terbaik serta dapat mengesan jumlah buah betik yang masak dalam tempoh masa yang

singkat. Oleh itu satu kaedah yang mudah akan dihasilkan dalam projek ini bagi

menentukan kemasakan buah dengan menggunakan RFID dan WSN. Persamaan telah

dibangunkan dalan perisian MATLAB dengan menggunakan konsep pemprosesan imej.

Pemprosesan imej yang digunakan adalah dengan mengenalpasti warna buah tersebut.

Warna tersebut dianalisis dengan menetapkan purata warna iaitu mestilah melebih 400

piksel. Berdasarkan kepada keputusan, betik tersebut dapat dikategorikan masak apabila

nilai warna melebihi 400 piksel dan tidak masak apabila nilai warna kurang daripada

400 piksel. Dengan menggunakan sistem ini, ketepatan sistem ini adalah kira-kira 80%

dimana 8 daripada 10 betik tersebut berjaya menghasilkan keputusan yang betul. Untuk

menentukan jumlah buah matang dalam masa yang singkat, sistem RFID telah

digunakan dengan menggunakan Xbee Pro S2b sebagai medium penghantaran. Terdapat

dua jenis pengujian yang boleh dilakukan iaitu pertama penghantaran data pada

persekitaran dalaman dan persekitaran luaran. Manakala dari segi aspek masa dan jarak

penghantaran, adalah terbukti mengikut spesikasi yang ditetapkan iaitu apabila jarak

melebihi 100m penghantaran signal yang lemah dan tiada penghantaran signal dapat

dilakukan. Dapat disimpulkan bahawa dengan menggabungkan kaedah pemprosesan

imej dan teknologi RFID, kemasakan buah-buahan dapat diperolehi dengan mudah di

mana saja dengan keputusan yang tepat.

viii

TABLE OF CONTENTS

CHAPTER ITEM

TITLE

PAGE

i

DECLARATION ii

iii

DEDICATION

ACKNOWLEDGEMENTS

iii

v

ABSTRACT vi

CONTENT viii

LIST OF TABLE

xi

LIST OF FIGURES xii

LIST OF ABBREVIATIONS xv

I INTRODUCTION 1

1.1 Introduction 1

1.2 Problem Statement 2

1.3 Objectives 3

1.4 Scopes Project 3

1.5 Thesis Outline 4

II LITERATURE REVIEW

2.1 Introduction 5

2.2 Overview of Fruit Maturity 5

2.2.1 Non- Colour Based Grading 6

2.2.1.1 Fruit Taste 6

2.2.1.2 Fruit Smell 7

ix

2.2.1.3 Fruit Shape and Size 8

2.2.1.4 Fruit Weight 9

2.2.2 Colour Based Grading 9

2.2.2.1 RGB Colour Space 10

2.2.2.2 HSI Colour Space 10

2.2.2.3 L*a*b* Colour Space 11

2.2.2.4 YCbCr Colour Space 12

2.3 Overview of RFID 12

2.4 Part of RFID 13

2.4.1 RFID Antenna 14

2.4.2 RFID Tag 14

2.4.2.1 Passive RFID Tag 15

2.4.2.2 Partial Passive RFID Tag 16

2.4.2.3 Active RFID Tag 17

2.4.3 Frequency Range of RFID 18

2.5 Overview of Wireless Sensor Network 18

2.6 Overview of Wireless Mesh Sensor Network 20

2.7 Overview of Zig Bee 20

2.8 Summary 23

III PROJECT METHODOLOGY 26

3.1 Project Workflow 26

3.2 Overall Project Methodology 27

3.3 Software Implementation 28

3.4.1 Input Image 29

3.4.2 Pre-processing 29

3.4.3 Feature Extraction 29

3.4.4 Classifier 30

3.4 MATLAB Software 31

3.5 X-CTU Software 32

3.6 Hardware Implementation 33

x

3.6.1 Webcam 34

3.6.2 Arduino Uno 35

3.6.3 Xbee Pro S2b 36

3.6.4 Xbee Shield 37

3.6.5 LCD Display 36

3.7 RFID Receiver 39

3.8 RFID Transmitter 40

3.9 Summary 40

IV RESULT AND DISCUSSIONS 42

4.1 Introduction 42

4.2 Experimentation Analysis and Result 42

4.2.1 Software MATLAB Validation 44

4.2.2 Software X-CTU Validation 51

4.2.2.1 Test 1- Indoor for Radio Range Test and

Throughput

52

4.2.2.2 Test 2- Outdoor for Radio Range Test

and Throughput

54

V CONCLUSION AND RECOMENDATION 58

5.1 Introduction 58

5.2 Recommendations 59

REFERENCES 60

APPENDICES

Appendix A – Source Code Arduino

– Source Code Matlab

Appendix B – Datasheet Arduino Uno

– Datasheet SkXbee

xi

LIST OF TABLES

NO

TITLE

PAGE

2.1 Frequency range in RFID with each applications. 18

2.2 Different between RFID and WSN 19

2.3 Comparison of The ZigBee, Bluetooth, Wi-Fi and UWB

Protocols

22

2.4 Comparison of previous works for non-color based grading 24

2.5 Comparison of previous works for color based grading

25

3.1 Camera Settings and Parameters 34

3.2 Pin connection between Xbee, Arduino Uno and LCD Display

39

4.1 Results Obtained Original Image and Grey Scale Image 50

4.2 Results Obtained In Terms Precision and Accuracy 51

4.3 Indoor test for distance 1m to 30m 56

4.4 Outdoor test for distance 1m to 30m 56

xii

LIST OF FIGURES

NO

TITLE

PAGE

2.1 The experimental setup for testing mango fruit using NIR 7

2.2 The schematic diagram of the electronic nose to detect the mandarin 8

2.3 Computer system vision 12

2.4 Architecture between RFID tags and RFID readers 13

2.5 Example of RFID Antenna 14

2.6 Types of RFID tag 15

2.7 Passive type RFID tag system 16

2.8 Partial passive RFID tag system 16

2.9 Active RFID tag system 17

2.10 The Concept of Wireless Monitoring Wireless Sensor Network 19

2.11 ZigBee Topologies 21

3.1 The Process of Project Methodology 27

3.2 A generalized block diagram of maturity detection of papaya fruits 28

3.3 Original RGB image versus Grey-scale image 29

3.4 Proposed algorithm 30

3.5 Steps for Determination of Fruit Maturity 31

3.6 Xbee series 2 attached with xbee wireless module 32

3.7 X-CTU Configuration Interface 33

xiii

3.8 Input, Image Processing and Output 33

3.9 Webcam Havic 6I6 34

3.10 Arduino Uno 35

3.11 Xbee Pro S2b 36

3.12 Xbee shield 37

3.13 LCD Display

37

3.14 Schematic for RFID Receiver 39

3.15 Connection for RFID Transmitter 40

4.1 Example of Mature Papaya 43

4.2 Example of Immature Papaya

44

4.3 Experimentation for Papaya 1 45

4.4 Experimentation for Papaya 2

41



46



47



48



49

4.13 Indoor- Radio Range Test and Throughput for 1m 52

4.14 Indoor-Radio Range Test and Throughput for 10m

52

xiv

4.15 Indoor-Radio Range Test and Throughput for 20m 53

4.16 Indoor-Radio Range Test and Throughput for 30m

53

4.17 Outdoor- Radio Range Test and Throughput for 1m 54

4.18 Outdoor-Radio Range Test and Throughput for 10m

55

4.19 Outdoor-Radio Range Test and Throughput for 20m 55

4.20 Outdoor-Radio Range Test and Throughput for 30m

56

4.21 Graph indoor versus outdoor for RSSI 57

4.22 Graph indoor versus outdoor for RSSI

57

xv

LIST OF ABBREVIATIONS

ANN - Artificial Neural Networks

GUI - Graphical User Interface

FUT - Fruit Under Test

H2H - Human To Human

H2M - Human To Machine

HF - High Frequency

LCD - Liquid Crystal Display

LF - Low Frequency

M2M - Machine To Machine

MLR - Multi Linear Regression

NIR - Near Infrared Reflectance

PCA - Principal Component Analysis

PLS - Partial Least Squares

RFID - Radio Frequency Identification

WMSN - Wireless Mesh Sensor Network

WSN - Wireless Sensor Network

ZC - Zigbee Coordinator

ZED - ZigBee End Device

ZR - ZigBee Router

CHAPTER 1

INTRODUCTION

1.1 Introduction

Agriculture is one of the contributors of the country economic policy,

especially in the third world such as Malaysia, China and many parts of Asia. There

are various methods to improve productivity and fruit quality such as color based

grading and non-color based grading. Most farmers have less knowledge on how to

manage their farm and they are unaware on how to improve their productivity of

agricultural practices. Therefore, an RFID-WMSN based system is proposed to

predict the maturity of fruits. In this work, papaya is selected as Fruit Under Test

(FUT), since it is one of the Malaysia’s local fruit with high demand request from

Malaysian.

In agriculture industries, determinations of fruit maturity and quality are very

important. There are two popular methods for fruit maturity prediction, which are

non-destructive and noninvasive technique. These two methods allow the physical

properties of the fruits is measured according to its maturity and quality indicator [1].

The papaya is said to be matured, if the flesh is soft, sugar is increased, soluble and

total solids as well as increase in pigmentations. [1]. Most of researchers [1] used

non-destructive testing technique to test fruits since this method is save for papaya,

provide real-time measurement, low in costing, and less time consuming.

In this work, a real time remote monitoring and control system is designed

and developed to predict the maturity of the fruits, which able reduce the number of

2

labor in future of fruit industries. In addition, an algorithm is proposed to

automatically identify the defect in the fruits and maturity of papaya using image

processing technique. The development of RFID-WMSN based system for

agriculture industries will possibly increase the efficiency, productivity and

profitability, as well as decreasing unintended effects on crops and the environment

in agriculture production.

1.2 Problem Statement

In fruit industries, the maturity of papaya is manually predicted by human

based on color which is prone to error and inefficiency due to the human subjective

nature. The effectiveness of this manual classification highly depends on the

experience, expertise and high concentration of the operator. Which leads to the

increment the cost of labor during maturity classification of papaya. The demand for

high quality fruits by consumers has increased day by day. Fast, efficient and cost

effective quality control is needed to achieve the target of production. Therefore, a

system that able to predict the maturity of papaya is developed by introducing RFID-

WMSN technologies on a single platform. The prediction of fruit maturity is done by

MATLAB software based on consecutive algorithm.

1.3 Objectives

The objectives of this project is as follows:

i) To design a new architecture of embedded system prototype components that

comprise of the RFID technology and WMSN platform.

ii) To develop fruit maturity system which can analyze, classify and identify

fruits based on color.

iii) To analyze and characterize the performance of the complete proposed

solution package of embedded system.

3

1.4 Scopes Project

There are 3 main part of the scopes.

i. Identify fruit maturity.

MATLAB Software was used to identify the fruit maturity based on

image processing concept. That means can differentiate by its color.

ii. Embedded system prototype components that comprise of the RFID

technology.

Used active RFID and arduino UNO as controller. RFID be able to

transmit between 30 to 60 meters indoor without router and more than

200 meters outdoor as a standalone device. The data, information and

number of products produced can be of real time controlled and

monitored by user.

iii. Utilizing with wireless mesh network.

Field monitoring uses 2.4 GHz operating frequency nodes for the

purpose of study. Used Xbee Pro S2b as a main component for data

transmission.

1.5 Thesis Outline

This thesis consists of five chapters which are Chapter 1: Introduction,

Chapter 2: Literature review, Chapter 3: Methodology, Chapter 4: Result and

Analysis and Chapter 5: Conclusion

Chapter 1 explains on the background of the project including introduction,

problem statement, objectives and scopes of the project.

Chapter 2 discusses on the method used to detect the maturity of fruit carried

out by previous researchers. The method categorized into two, which are color based

method and non–color based. Color based method can be categorized by 4 methods

which are RGB Color Space, HSI Color Space, L*a*b* Color Space and YCbCr

Color Space while non-color based method are categorized by 4 methods, which are

taste, smell, size and weight. In addition to fruit maturity prediction, this chapter also

4

highlighted on the existing technology used for data transmission as well as the

network topology available.

Chapter 3 discusses on the methodology of the project suggested in order to

complete the development. It can be divided into two parts namely hardware and

software. The hardware part focuses on the system development including the RFID

tag and reader portion. The software part refers to the development of algorithm that

use image processing method using MATLAB 2009B version, while the X-CTU

program is used to configure the Radio Frequency (RF) Module.

Chapter 4 focuses on the result and analysis of the fruit maturity system based

on RFID-WMSN platform. From here, it can be seen the performance of this project.

The performance of the develop system and algorithm are tested based on color.

Algorithm 1 is developed for pre-processing of the database and algorithm-2 is

proposed for identification of defects.

Finally, in Chapter 5 the conclusions and suggestions are made for future

work.

5

CHAPTER 2

LITERATURE REVIEW

2.1 Introduction

In this chapter, the important aspects of RFID are described and relevant

background research is later explained to understand the current state of the art and a

review of the previous works undertaken in designing RFID system for fruit maturity

prediction is performed to suit the requirements before a complete system can be

developed. Finally, this research work is put in context with the previous and current

research activities.

2.2 Overview of Fruit Maturity

Nowadays, the technology used for fruit maturity prediction is growing in

line with the current agricultural economic. Several research works have been done

to check maturity of fruits using electronic devices [2]. However, there are several

gaps regarding to the past methods on determination of the fruit maturity. On top of

this, this study is to explore the most feasible methodology in determining the

maturity of papaya. Papaya is one of the popular fruit in Malaysia. It also has been

marked as ‘Malaysia’s Best Fruit’ that approved by Ministry Of Agriculture [3].

Instead of papaya, other fruits such as apple, oranges, peaches, apricot also have been

tested for their maturity but different methods [4]. The methods can be divided into

6

two categories, which are first non-color based grading and color based grading. In

this project, maturity color based grading is the main inspection criteria for papaya.

The maturity of papaya changes from green (immature) to yellow (mature).

2.2.1 Non-color Based Grading

Preliminary, fruits in the market are inspected based on its maturity (non-

color based grading). Consequently, several works have been identified on the fruit

grading based on its taste, smell, weight, shape and size.

2.2.1.1 Fruit Taste

Fundamentally, the maturity of fruits can be determined through the taste of

sour (immature) to sweet (mature) [4]. Therefore, by using Near Infrared Reflectance

(NIR) method, the sucrose level content in various grades of mangoes maturity can

be determined. [4]. In the previous research performed by A. Afhzan et al. [4], the

acquired analysis indicates that there is a correlation between the absorption of

infrared and the percentage of the sucrose level for each watermelon. Multi-linear

regression (MLR), principal component analysis (PCA) and partial least squares

(PLS) regression with respect to the reflectance and its derivative, the logarithms of

the reflectance reciprocal and its second derivative are the vital components to

develop NIR models [4].

On the other side, bonding of organic molecules will change their vibration

energy from the peak of NIR frequency and exhibits an absorption process through

the spectrum [5]. Other than papaya, the maturity and firmness for another fruits such

as watermelon, peaches, and apricot are determined through this method [6]. Figure

2.1 shows the previous experiment for testing of mango fruit using NIR [7].

7

Figure 2.1 : The experimental setup for testing mango fruit using NIR [7]

In determining the sugar content of melons, sugar absorption band method

via NIR was used to avoid bias caused by the color information of melons proposed

by Tsuta [8]. From this related work, each of the second-derivative absorbencies at

874 nm and 902 nm had a high correlation with the sugar content of melons. Xiaobo

[9] used NIR to measure sugar content inside ‘Fuji’ apple. Using Multiple Linear

Regression (MLR), and it is observed that there is an existing correlation between

content of sugar and various NIR wavelengths in this particular work. Also, sugar

content of a fruit can be determined as well as it sweetness.

2.2.1.2 Fruit Smell

Durians have strong aroma which indicates that the fruit is mature. Several

works have been performed to investigate the aroma of the fruits and its maturity, for

example, electronic noise proposed by Brezmes [10]. In this research, an artificial

olfactory system is proposed in which is used as a non-destructive instrument to

measure the fruit maturity, as it was developed to follow the mammalian olfactory

system within an instrument. Moroever, the system is designed to obtain

measurements, identifications and categorizations of aroma mixtures. Nowadays,

electronic nose technology is improved to explore the fruit ripening stage.

8

A. H. Gómez et al. [11] has proposed an Electronic Nose to identify the

maturity of tomatoes, as the Electronic Nose device uses ten different metal oxides

sensors (portable E-nose, PEN 2) with respect to the changes of volatile production

[11]. Consequently, it has been proven that the Electronic Nose PEN 2 is a successful

methodology in identifying the maturity of tomato other fruit such as blueberry,

apple, and mandarin [12]

The Electronic Nose is an efficient process and non-destructive method in

practically sensing the scent of fruits as well as expecting the uppermost crop period

of time. Moreover, there is an available commercial Electronic Noses that uses an

array of sensors combined with pattern recognition software [13]. Additionally, there

have been several hearsays on electronic sensing in environmental control, medical

diagnostics in the food industry, as a good correlation between the electronic nose

signals and firmness, starch index, alongside with acidity parameters were found in

this procedure and electronic noses portray the potential of becoming a reliable

device to determine the fruit maturity [13]. Figure 2.2 shows that the schematic

diagram of the electronic nose method potential checking mandarin maturity [14].

Figure 2.2 : The schematic diagram of the electronic nose to detect the mandarin [14]

2.2.1.3 Fruit Shape and Size

Fruit shapes and sizes are an important aspect to determine the fruit maturity.

Mostly fruits with bigger size are much more ripped or mature in context, compare

with the smaller ones. As the fruit grew more mature, its size and shape changes and

reacts more toward the environment in the sense of geotropism. Furthermore, Liming

and Yanchao [15] have proposed that the grading strawberry fruit is based upon its

shape, size and color. Therefore, through investigated designed guidelines,

9

strawberry contour are extracted from its shape and in eliminating the influence of

the strawberry size, normalizing the line sequences eliminates its length of the shape.

Also, size grading is implemented in strawberry and a similar process is desired,

which is to reach the maximum horizontal diameter. Besides, from the fruit size the

maturity of strawberry can be estimated by contour line sequence.

2.2.1.4 Fruit Weight

Additionally, in determining the fruit maturity by its weight, the fruit density

or area of cross section of the fruit must be provided. Hsieh and Lee [16] introduced

a homemade hyper-spectral imaging system to detect internal and external quality of

papaya. Herein, to predict weight, the area parameter can be a linear correlation with

a coefficient of 0.93, where the absolute calibration error is around 7% when area of

side view cross section of the fruit was used.

2.2.2 Colour Based Grading

On the other side, from the color of the fruit, it can be determined if whether

the particular fruit has matured or otherwise. This is because fruit color changes from

dark green to a pale orange like color during its maturity process and thus the fruit

maturity can be determined through its color observation. Besides, color space is an

important factor in determining those algorithms that has to suit in the system

specification. Subsequently, RGB, HSI, L*a*b* and YCbCr color spaces are the

most common color space used by several researchers.

10

2.2.2.1 RGB Colour Space

Many researchers due to the imaging or features represented in the RGB

color space broadly utilize RGB color space. The methodology works feasibly

through an extract from the number of pixels of an image. Additionally, Abdul

Rahman [17] used and applied RGB color space in order to monitor the ripeness of

watermelon as the color space is used in extracting features from the watermelon.

Also, fuzzy logic system is utilized to classified color features extracted from the

three RGB components. The purpose is to determine the ripeness level of the

watermelon. Herein, 95%-100% accuracy of a total of 90 samples, which contains 30

samples for each category, was achieved.

2.2.2.2 HSI Colour Space

Notwithstanding on the method employed, HSI color space was also

performed by Abdullah [18] to classify starfruit maturity through its surface skin

color into its maturity index based on single Hue. Therefore, in establishing the color

recognition techniques, multivariate discriminant analysis and multi-layer perceptron

neural network was used. This work is practicable to differentiate the maturity of the

star fruit into four groups of unripe, under ripe, ripe, and overripe. Additionally,

about 92% of the tested samples could be classified in this technique. Besides, Mohd

Mokji has performed previous related work as well on the starfruit color maturity

classification [19]. The investigation was developed based on the FAMA’s old

version of starfruit color maturity standard, whereby the classification if performed at

six different indices and the accuracy of this work has achieved about 93.3% for

starfruit color maturity.

11

2.2.2.3 L*a*b* Colour Space

Also, L*a*b* color space was performed by Kang [20] in 2008 based on the

color changes of a mango. The investigation measured on the color value of a* and

b* on a curved surface that delivers about 55% to 69% of the percentage within its

range measured. Additionally, hue angle and chrominance have an average error of 2

and 2.5 respectively by using the measurement results. Syahrir [21] also uses the

L*a*b* color space in investigating if whether a tomato has rotten. The image was

later improved through a filter and threshold process before converting the image to

L*a*b* color space format. About 90% of the tomatoes tested were not rotten

according to the test outcomes and thus the judgment of tomato maturity and the

estimation of tomato’s expiry date were précised in this work. Besides, nine simple

features of the appearance extracted from images of bananas were used for grouping

purposes. The nine simple features are L*, a*, b* values; brown area percentage,

number of brown spots per cm2, homogeneity, contrast, correlation, and entropy of

image texture.

L* is the luminance or lightness component that goes from 0 (black) to 100

(white), and parameters a* (green to red) and b* (blue to yellow) are the 2 chromatic

components, varying from – 120 to +120. The results obtained are precisely match

with human observation that makes the system important to be used.

Figure 2.3 shows the computer system vision [22] for online prediction to

predict the maturing phases of bananas. The computer vision system (CVS) consisted

of Lighting system and Digital camera and image acquisition. The 1st critical step in

displaying out banana images is by segmentation of the image from the background.

A threshold of 50 in the grayscale combined with an edge detection technique based

on Laplacian-of Gauss (LoG) operator is the pre-process to remove background. All

banana images in their changed maturing stages were processed and segmented using

the same method.

12

Figure 2.3 : Computer system vision

2.2.2.4 YCbCr Color Space

Shah Rizam [23] measured the ripeness and quality of watermelon based on

YCbCr color space in 2009. Cb and Cr were calculated and computed by using

Artificial Neural Networks. The system shows that the system can be used to classify

the watermelon because the accuracy 86.51% was achieved.

2.3 Overview of RFID

Moreover, RFID stands for Radio Frequency Identification and the term RFID

is used to describe various technologies that use radio waves that automatically

identify a person or an object. RFID technology is a common way of storing or

receiving data remotely using a device that encompasses RFID tags and RFID

readers. RFID tags are small objects such as stickers, key chains or chip tags with

different range ranging from 10 mm to 6 meters. RFID tags can be either active or

passive. Passive RFID tags do not have their own power supply, so the price is

cheaper than the active tags. By only armed with the induced electrical induction on

the antenna caused by an incoming radio frequency scan, it is sufficient to provide

sufficient power for the RFID tag to transmit the feedback. Active RFID tags have

their own power supply and have longer range. Memory is also larger so that it can

accommodate various kinds of information in it. RFID tags that are widely circulated

13

now are RFID tags that are passive RFID is one of an operative automatic

identification technology for different things. The ultimate function of RFID is the

capability to trace the position of the tagged things.RFID tags attempt to receive and

act repercussions against an object to be impressed with the help of an antenna

emitted by RFID tags and RFID readers. Figure 2.4 shows the architecture between

RFID tags and RFID readers.

Figure 2.4 : Architecture between RFID tags and RFID readers.

2.4 Part of RFID

RFID is a system that can store and control data using wireless technology and

it operates at a frequency of 50 khz up to 2.5 Ghz. RFID technology can be divided

into three main parts: antenna, RFID tag and RFID reader to enable an object to be

detected according to the desired operating distance of the user based on the selection

of the tag itself which consists of passive RFID tag and active RFID tag.

14

2.4.1 RFID Antenna

RFID antenna is a device that transmits radio signals to enable RFID tags and

RFID readers. The RFID antenna is able to read, write and control data acquisition

and communication systems between RFID tags and RFID readers. The RFID

antenna can be designed according to its use in RFID applications such as the use of

home theft security tools, the RFID tag system applied in highway toll payments

more. Electromagnetic field generated by the RFID antenna can detect the presence

of RFID tags at a certain distance according to the type of RFID used. Figure 2.5

shows the example of RFID Antenna.

Figure 2.5 : Example of RFID Antenna

2.4.2 RFID Tag

The RFID tag requires an operating source to perform operations to transmit

radio signals to RFID readers. RFID tags gain power from battery cells or from

electromagnetic waves transmitted by RFID readers that induce the electric current

inside the tag. The required energy on the RFID tag depends on several factors

including the operating distance between the RFID tag and the RFID reader, the

radio frequency used by an RFID tag. In general, the more complex the functions

used by an RFID tag, the more energy it takes to interact with the RFID reader.

Figure 2.6 shows types of RFID tags.

15

Figure 2.6 : Types of RFID tag

2.4.2.1 Passive RFID Tag

This passive type RFID tag does not use internal power supply. The electric

current used on the antenna is derived from radio frequency signals which provide

sufficient power for an integrated circuit in the RDID tag operating and transmitting

information. Most of the passive tag signals are due to a carrier wave backdrop from

an RFID reader. This situation indicates that this antenna is designed for both

functions receiving signal and also transmitting back wave. Reactions from passive

tags are not just for identification numbers, but RFID tag chips can also contain non-

volatile memory to store data. Passive tag has a practical reading distance of 10cm,

which is 4 inches up to several meters and it relies on the selection of radio

frequency and the type and size of the antenna. Figure 2.7 shows a passive type

RFID tag system.

16

Figure 2.7 : Passive type RFID tag system.

2.4.2.2 Partial Passive RFID Tag

Partial passive type RFID tags are similar to active RFID tags for example

they have their own power source but the batteries used are just to empower

microchips instead of sending signals. Radio frequency energy will be reverted to

RFID readers such as passive RFID tags. Figure 2.8 shows partially active RFID tag

systems.

Figure 2.8 : Partial passive RFID tag system.

17

2.4.2.3 Active RFID Tag

Unlike the passive type tag, the active RFID tag has its own power supply,

which uses the power supply to integrated circuits and sends signals to RFID readers.

The RFID tag is activated by a power source from an integrated circuit so that the

power to be used is sufficient and also produces higher power than the passive type

RFID tags to enable it to work more effectively in challenging Radio Frequency

environments such as ice, iron or at a fair distance far away. Mostly active type tags

have a practical distance of 100 meters and the battery life exceeds 10 years. Active

RFID tags are equipped with sensors such as temperature sensors, which are used to

indicate the temperature of a material that is easily damaged or durable. Active RFID

tags typically have a distance of about 500 m and have greater memory than passive

type RFID tags. This RFID tag is also capable of storing additional information

transmitted by an information conveyor. Figure 2.9 shows an active type RFID tag

system.

Figure 2.9 : Active RFID tag system.

18

2.4.3 Frequency Range of RFID

The main aspect required in RFID selection is the appropriate frequency to be

applied to the system. This is because the appropriate frequency should be chosen to

allow wireless communication between the RFID reader and RFID tags to be

interconnected. RFID system fracture selection will influence the communication

distance of a system, frequency interference with other radio frequency and speed of

data to communicate with each other in the system. Table 2.1 shows the frequency

range in RFID with each applications.

Table 2.1 : Frequency range in RFID with each applications.

Frequency Range Applications

Low-frequency

125 - 148 KHz

up to 3 inches Pet and ranch animal identification;

car keylocks; factory data collection

High-frequency

13.56 MHz

up to 3 feet library book identification;

clothing identification; smart cards

Ultra-high frequency

433 MHz

up to 300 feet Defense applications w/ active tags

Ultra-high frequency

865 - 928 MHz

up to 36 feet Supply chain tracking:

Boxes, pallets, containers, trailers

2.5 Overview of Wireless Sensor Network

Wireless network refers to the technology to communicate and access the

internet without cable connection between computers and other electronic devices.

Sensor Network has contributed to several applications, and awareness has expended

to implement the technology into the agriculture environment. WSN is one of the

most important technologies in the 21st century [24]. WSN is an assembly of a

number of low-power, low-cost, multipurpose sensor nodes communicating wireless

upon a short distance.

The difference between a WSN and a RFID system is that RFID devices have

no cooperative capabilities, while WSN allows different network topologies and

multihop communication [25]. WSN can cut down the effort and time needed for

monitoring environment. As a result, money, water and labour costs can be reduced.

The technology allows for remote measurements. There seems to be increased

19

development towards wireless outcomes in comparison to wired-based systems.

Figure 2.10 shows the concept of wireless monitoring that is to be applied in the

agriculture environment.

Figure 2.10 : The Concept of Wireless Monitoring Wireless Sensor Network

This systems provides a full network coverage in large facilities such as a big

industries, typically massive lengths of cabling that leads to remarkable return on

investment. WSN provides an intelligent platform to gather and collect data from the

sensor nodes that can detect and interact with the physical environment [26].

Using a wireless mesh as a backbone network simplifies installation and

provides an affordable medium [27] WSN can be used to identify maturity to admit

the irrigation system and identify where and when to irrigate. Table 2.2 show the

different between RFID and WSN.

Table 2.2: Different between RFID and WSN (Hai liu et al, 2008)

RFID technology WSN

Purpose Detect presence and location of

tagged objects

Sense interested parameters in

environment and attach objects

Component Tags, readers Sensor nodes, relay nodes, sinks

Protocols RFID standards ZigBee, Wi-Fi

Communication Single hop Multihop

Mobility Tag move with attached objects Sensor nodes are usually static

Programmability Usually closed systems Programmable

Price Reader-expensive, tag-cheap Sensor node-medium

20

2.6 Overview of Wireless Mesh Sensor Network

The ZigBee platform over IEEE 802.15.4 is a protocol standard for short range

wireless communication, while RFID technology uses RF technology for

communication which is typically used for object identification and tracking system.

In this proposed system a new design of RFID WMSN-based system adopted

through ZigBee platform which upscales the active RFID system is developed to

suite with the indoor application providing robust and efficient data delivery in

industrial manufacturing environment. WMSN developed for this project is based on

three main components function explained previously in chapter 2 which namely

ZigBee end device or tag (ZED), ZigBee reader or coordinator (ZC) and ZigBee

router (ZR). The function process of WMSN components comprise of a plurality of

RFID tags, a plurality of transceiver to form a self-healing communication network,

and a host computer deployed.

2.7 Overview of Zig Bee

The well-known four protocol standards for short range wireless

communications with low power consumption is ZigBee platform over IEEE

802.15.4, Wi-Fi over IEEE 802.11 while Bluetooth over IEEE 80215.1 and ultra-

wideband (UWB) over IEEE 802.15.3. Latest technology known as WiMAX is

suitable for long range wireless communication up to 50km at 75Mb/s rate per

channel. WiMAX is Wireless WAN over IEEE 801.16 standard. Since, WiMAX

technology is not really suitable for industrial automation indoor application, due to

WiMAX application are essentially WISP (wireless internet service provider) which

suitable for outdoor applications in a large network environment [28] ZigBee, which

was originated in 1998, is based on the IEEE 802.15.4 standard and pioneered by

ZigBee Alliance, which was formed by several companies interested in defining low

cost, low power, and wireless network standard [29]. ZigBee can support large

number of nodes providing a low cost global network. The IEEE defines only the

PHY and MAC layers in its standard, while ZigBee defines the network and

application layers, application profile and security mechanism. Due to this design,

the consumption of power is minimal and the battery life span is longer.

21

Zigbee supports three topologies, which are star, mesh and cluster-tree, as

shown in Figure 2.11.

Figure 2.11 : ZigBee Topologies

In stars topology, each end node is connected to the coordinator and the

Zigbee Coordinator (ZC) carries out communication. In mesh topology, each device

communicates with any other device within its radio range or through multi-hop. In

cluster tree topology, there is a single routing path between any devices [30]. In the

Zigbee application, it is mostly used for mesh topology. In spite of that, for the

proposed system monitoring mesh topology was chosen. The various sense data from

moisture sensors go to WMSN, that integrates with RFID tag and sprinkler will turn

into a node. On the farm, there are plenty of nodes and each node will communicate

through this ZigBee technology platform. Based on that, the reader will read the

sensor data and stores the data to the server, which is used by a farmer for

monitoring. In the proposed system, field monitoring uses 2.4 GHz operating

frequency nodes for the purpose of study. Table 2.3 summarizes the main

comparison among the four protocols.

22

Table 2.3: Comparison of The ZigBee, Bluetooth, Wi-Fi and UWB Protocols).

Standard ZigBee Bluetooth Wi-Fi UWB

IEEE standard 802.15.4 802.15.1 802.11a/b/g 802.15.3a

Frequency band 868/915 MHz,

2.4GHz

2.4Ghz 2.4GHz, 5GHz 3.1-10.6 GHz

Max signal rate 250Kb/s 1Mb/s 54Mb/s 110Mb/s

Nominal range 10m -100m 10m 100m 10m

Nominal Tx Power (-25) – 0dBm 0 – 10dBm 15-20dBm -41.3

dBm/MHz

Number of RF

channels

1/10; 16 79 14(2.4GHz) 1-15

Channel bandwidth 0.3/0.6MHz,

2Mhz

1MHz 22MHz 500MHz-

7.5GHz

Modulation type BPSK(+ASK),O-

QP SK

GFSK BPSK,QPSK,COFDM,

CCK,M-QM

BPSK,QPSK

Spreading DSSS FHSS DSSS,CCK,OFDM DS-

UWB,MB-

OFDM

Coexistence

mechanism

Dynamic freq.

selection

Adaptive

freq. hoping

Dynamic freq.

selection, Tx power

control

Adaptive freq.

hoping

Basic cell Star Piconet BSS Piconet

Extension of the basic

cell

Cluster tree,

mesh

Scatternet ESS Peer-to-peer

Max number of cell

nodes

>6500 8 2007 8

Data protection 16bit CRC 16bit CRC 32bit CRC 32bit CRC

With reference to Table 2.3 the comparison is based on an IEEE standard.

Even ZigBee and Bluetooth give a lower data rate intended for WPAN

communication. Furthermore ZigBee enhanced mesh network up to 100 meters and

beyond, depends to the network deployment and applications. In this project, it

allows the coverage area to be extended when new routers are added. Wi-Fi is

oriented to WLAN about 100 m. UWB and Wi-Fi provide a higher data rate. The

nominal transmission power is 0 dBm for both Bluetooth and ZigBee while 20 dBm

for Wi-Fi. Obviously, ZigBee permits the maximum number of cell nodes to be

higher, up to >6500 nodes. For ZigBee and Bluetooth, Baker [31] studied their

strengths and weakness for industrial applications, and claimed that ZigBee over

802.15.4 protocol can meet a wider variety of real industrial needs than Bluetooth

due to its long term battery operation, greater useful range, flexibility in a number of

23

dimensions and reliability of the mesh networking architecture (Toscano and Bello,

2008).

2.8 Summary

This chapter explains all the previous work related to fruit maturity and RFID.

The different researchers performed their work and came out with their own systems.

There are several approached that have been carried out to detect the fruit maturity.

Non-color grading and color grading were discussed which are related to the fruit

maturity classification. For non-color-based grading can be detected through taste,

smell, shape and size and also weight. For this method, the researchers involved were

A.Afhzan, Tsuta, Xiabao, Brezmes, Limimg and Yanchas and Hsein & Lee. Each

approach have same purpose only different methods. For color-based grading there

are 4 types of approach which is RGB, HSL, L* a* b and YCbCr. Another that this

chapter presented on the active RFID tag and WMSN that can implemented in this

project. Zigbee is chosen for this proposed system due to the characteristics. The

Zigbee based on IEEE 802.15.14 standard that works at 2.4Ghz to increase the

system capabilities in terms of extending the range and providing wireless mesh

sensor network. The proposed method will be able to reduce the labor cost, hardware

cost and provide less circuit complexity.

Table 2.4 : Comparison of previous works for non-color based grading

Method

Researches Equipment Types of Fruits Descriptions

A. Afhzan Near Infrared Reflectance

(NIR)

Mangoes Correlation between absorption of

infrared and percentage of sucrose

level

Taste Tsuta Near Infrared Reflectance

(NIR)

Watermelon Sugar content of melons and sugar

absorption

Xiaboo Near Infrared Reflectance

(NIR)

Fuji Apple Using Multiple Linear Regression.

Correlation between content of

sugar and various wavelength NIR

Smell Brezmes Electronic Nose Orange Used non-destructive instrument

A.h.Gomez.et Electronic Nose Tomato Used 10 different metal oxide

sensors

Shape and Size Limimg & Yanchao Electronic Nose Strawberry Eliminate by normalizing the line

sequence

Weight Hseih & Lee Hyper Spectral Imaging Papaya Estimated fruit density or area of

cross section

60

REFERENCES

[1] A. Rocha, D. C. Hauagge, J. Wainer, and S. Goldenstein, “Automatic fruit

and vegetable classification from images,” Comput. Electron. Agric., vol. 70,

no. 1, pp. 96–104, 2010.

[2] A. M. Aibinu, M. J. E. Salami, A. A. Shafie, N. Hazali, and N. Termidzi,

“Automatic Fruits Identification System Using Hybrid Technique,” in

Proceedings of 6th IEEE International Symposium on Electronic Design, Test

and Application, 2011, pp. 217–221.

[3] “MBG Yes! Fruits Survey Findings 2013,” 2013. [Online]. Available:

http://mbg.com.my/MBG/news-a-updates/1322-mbg-yes-fruits-

surveyfindings-2013.html.

[4] A. Afhzan, A. Rahim, N. E. Abdullah, S. Lailatul, M. Hassan, and I.

Shairah, “A Numerical Analysis of Correlation Between Sucrose Level

Measurement and N ear-Infrared ( NIR ) for Various Grades of Watermelon,”

pp. 180–185, 2013.

[5] G. Carlomagno, L. Capozzo, G. Attolico, and a. Distante, “Non-destructive

grading of peaches by near-infrared spectrometry,” Infrared Phys. Technol.,

vol. 46, no. 1–2, pp. 23–29, Dec. 2004.

[6] S. Bureau, D. Ruiz, M. Reich, B. Gouble, D. Bertrand, J.-M. Audergon,

and C. M. G. C. Renard, “Rapid and non-destructive analysis of apricot

fruit quality using FT-near-infrared spectroscopy,” Food Chem., vol. 113,

no. 4, pp. 1323–1328, Apr. 2009.

[7] Z. Schmilovitch, A. Mizrach, A. Hoffman, H. Egozi, and Y. Fuchs,

“Determination of mango physiological indices by near-infrared

spectrometry,” Postharvest Biol. Technol., vol. 19, no. 3, pp. 245–252, Jul.

2000.

http://mbg.com.my/MBG/news-a-updates/1322-mbg-yes-fruits-surveyfindings-2013.html

http://mbg.com.my/MBG/news-a-updates/1322-mbg-yes-fruits-surveyfindings-2013.html

61

[8] Mizuki Tsuta, Junichi Sugiyamaand Yasuyuki Sagara (2002). Near-Infrared

Imaging Spectroscopy Based on Sugar Absorption Band for Melons. J. Agric.

Food Chem. 50 (1):48–52 Networks (SoftCOM), International Conference.

IEEE. Split.11-13 September 2012. 1-5.

[9] Zou Xiaobo and Zhao Jiewen (2005). Apple quality assessment by fusion

three sensors. IEEE Jurnal Sensors.Oct. 30 - Nov. 3. Irvine, CA: IEEE. 389-

392.

[10] Jesús Brezmes, Ma. Luisa López Fructuoso, Eduard Llobet, Xavier Vilanova,

Inmaculada Recasens, Jorge Orts, Guillermo Saiz and Xavier Correig (2005).

Evaluation of an Electronic Nose to Assess Fruit Ripeness. IEEE Jurnal

Sensors. 5(1):97-108.

[11] A. H. Gómez, G. Hu, J. Wang, and A. G. Pereira, “Evaluation of tomato

maturity by electronic nose,” Comput. Electron. Agric., vol. 54, no. 1, pp.

44–52, Oct. 2006.

[13] C. Di Natale, A. Macagnano, E. Martinelli, R. Paolesse, E. Proietti, and A.

D’Amico, “The evaluation of quality of post-harvest oranges and apples by

means of an electronic nose,” Sensors Actuators B Chem., vol. 78, no. 1–3,

pp. 26–31, Aug. 2001.

[14] A. H. Gómez, J. Wang, G. Hu, and A. G. Pereira, “Electronic nose

technique potential monitoring mandarin maturity,” Sensors Actuators B

Chem., vol. 113, no. 1, pp. 347–353, Jan. 2006.

[15] Xu Liming and Zhao Yanchao (2010). Automated strawberry grading system

based on image processing. Computers and Electronics in Agriculture.72:32–

39.

[16] C. L. Hsieh and M. H. Lee (2009). Applied Hyperspectral Image on

Detecting the Maturity, Weight, and Sugar Content of Tai-Farm No.2 Papaya.

American Society of Agricultural and Biological Engineers. June 21 - June

24, 2009. Reno, Nevada: Curran Associates, Inc. 392-394.

[17] Farah Yasmin Abdul Rahman, Shah Rizam Mohd Shah Baki, Ahmad Ihsan

Mohd Yassin, Nooritawati Md. Tahir and Wan Illia Wan Ishak (2009).

Monitoring of Watermelon Ripeness Based on Fuzzy Logic. World Congress

on Computer Science and Information Engineering. 20-22 October, 2009.San

Francisco, USA: Springer. 67-70

62

[18] M.Z. Abdullah, J. Mohamad-Saleh, A.S. Fathinul-Syahir and B.M.N. Mohd-

Azemi (2006). Discrimination and classification of fresh-cut starfruits

(Averrhoa carambola L.) using automated machine vision system. Journal of

Food Engineering. 76:506–523. 15

[19] Musa Bin Mohd Mokji (2009). Features Construction for Starfruit Quality

Inspection. Ph.D. Thesis. Universiti Teknologi Malaysia.

[20] S.P. Kang, A.R. East and F.J. Trujillo (2008). Colour vision system

evaluation of bicolour fruit: A case study with ‘B74’ mango. Postharvest

Biology and Technology. 49:77–85.

[21] W. Md. Syahrir, A. Suryanti and C. Connsynn (2009). Colour Grading in

Tomato Maturity Estimator using Image Processing Technique. 2nd IEEE

International Conference on Computer Science and Information Technology

(ICCSIT). August 08-August 11. Beijing, China: IEEE. 276 – 280.

[22] P. Properties, “E : Food Engineering and Physical Properties Application of

Image Analysis,” vol. 69, no. 9, pp. 471–477, 2004.

[23] Shah Rizam M. S. B., Farah Yasmin A.R., Ahmad Ihsan M. Y. and Shazana

K (2009). Non-destructive Watermelon Ripeness Determination Using Image

Processing and Artificial Neural Network (ANN). International Journal of

Computer Systems Science and Engineering. 4(6):391-395.

[24] Zukifli, C. Z. and Suparmin, S. 2015. Wireless Communication Network

Tools Kits With Visualization Techniques: Design Consideration For

Proposed Architecture. International Journal of Multimedia and Ubiquitous

Engineering. 10(7): 375–386.

[25] Raed, M. T. A. 2012. Design And Implementation Of MultiHop Active RFID

System With Embedded Wireless Sensor Network USM, Penang, Malaysia.

[26] Abdullah, S., Ismail, W. and Zulkifli, C. Z. 2013. Integrating ZigBee-based

Mesh Network for Production Management Automation with Embedded

Passive and Active RFID. Proceedings of 2013 IEEE RFID Technical &

Applications Conference, Johor Bharu, 4-5 September 2013, Malaysia.

[27] Zheng, J., and Jamalipour, A. 2009. Wireless Sensor Network: A Networking

Perspective. John Willey & Sons..

[28] Abdullah, S., Ismail, W., Halim, Z. A. and Zulkifli, C. Z. 2013. Integrating

Zigbee-Based Mesh Network With Embedded Passive And Active RFID For

Production Management Automation. In RFID-Technologies and

63

Applications (RFIDTA), 2013 IEEE International Conference. IEEE. Johor

Bharu, Malaysia. 4-5 September 2013. 1–6. 2013. M2M Communication In

Virtual Sensor Network For Shaal. Jurnal Teknologi. 65(1): 99–105.

[29] Zigbee, A. 2015. The ZigBee Alliance. 2015. [Online]. Available:

http://www.zigbee.org/. [Accessed: 23-Aug- 2015].

[30] Abdulla, R. and Ismail, W. 2011. Extend active RFID with A ZigBee

Network. Microwaves & RF. 50(2): 79.

[31] Baker, N. “ZigBee and Bluetooth: Strengths and weaknesses forindustrial

applications,” IEE Computing & Control Engineering , vol. 16,no. 2, pp 20-

25, April/May 2005

design and implementation of a 2.4 ghz embedded rfid...

Documents