UNIVERSITI PUTRA MALAYSIA

CHANGE DETECTION IN MANGROVE FOREST AREA USING LOCAL MUTUAL INFORMATION

MAHIRAH BINTI JAHARI

ITMA 2010 5

CHANGE DETECTION IN MANGROVE FOREST AREA USING LOCAL MUTUAL INFORMATION

MAHIRAH BINTI JAHARI

MASTER OF SCIENCE UNIVERSITI PUTRA MALAYSIA

2010

CHANGE DETECTION IN MAGROVE FOREST AREA USING LOCAL MUTUAL INFORMATION

By

MAHIRAH BINTI JAHARI

Thesis Submitted to the School of Graduate Studies, Universiti Putra Malaysia, in Fulfilment of the Requirements for the Degree of Master of Science

November 2010

Abstract of thesis presented to the Senate of Universiti Putra Malaysia in fulfilment of the requirement for the degree of Master of Science

CHANGE DETECTION IN MANGROVE FOREST AREA USING LOCAL

MUTUAL INFORMATION

By

MAHIRAH JAHARI

November 2010

Chairman: Siti Khairunniza bt Bejo, PhD Faculty: Institute of Advance Technology

This thesis unveils the potential and utilization of similarity measure for forest change

detection. A new simple similarity approach based on local mutual information is used

to detect any significant changes in the image of forest areas. Point similarity measure is

defined as a measure which is used to calculate the similarity of individual pixels. The

basic idea of the proposed method is that any change pixel will be maximally dissimilar,

i.e. the value of similarity of these pixels will be low. The method has been tested to

detect and identify changes caused by plant growth and plant loss in four sub-areas of

Matang Mangrove Forest, Perak. Image of SPOT 5 satellite taken from band 1, band 2,

band 3, and band 4 with the resolution of 10meter dated on 5 August 2005 and 13 June

2007 has been used to test the method. It is then compared with the results of Principal

Component Analysis 1 (PCA 1). The plant loss areas has been successfully identified as

any pixel with the value of local mutual information less than and equals to zero. The

method has been refined to accurately detect changes caused by the growth areas by

ii

thresholding the histogram of the average percentage of difference between joint

probability and marginal probability. Results from the experiment showed that a

threshold value of zero is the best threshold value to identify between changed and

unchanged areas in all cases of the images. In overall, band 3 gives the best results of

forest change detection compared to the other bands in all cases. Compared to the image

differencing and normalized differenced vegetation index (NDVI), the proposed method

not only can solve the problem on selecting the threshold value but also provides the

highest percentage of successful classification at the fourth, second and first study area

with the value of 95.07%, 89.47% and 87.66% respectively. From the results, it has

been concluded that local mutual information is not only can be effectively used for

change detection technique but also can be used to classify the plant growth and plant

loss areas.

iii

Abstrak tesis yang dikemukakan kepada Senat Universiti Putra Malaysia sebagai memenuhi keperluan untuk ijazah Master Sains

MENGESAN PERUBAHAN DI KAWASAN HUTAN PAYA BAKAU

MENGGUNAKAN PENENTUAN KESERUPAAN SETEMPAT

Oleh

MAHIRAH JAHARI

November 2010

Pengerusi: Siti Khairunniza bt Bejo, PhD Fakulti: Institut Teknologi Maju

Tesis ini menyingkap potensi penggunaan penentuan keserupaan setempat dalam

kaedah pengesanan perubahan bagi kawasan hutan. Satu pendekatan keserupaan yang

baharu dan mudah berdasarkan maklumat salingan tempatan digunakan bagi mengesan

apa sahaja perubahan penting dalam imej kawasan hutan. Titik keserupaan setempat

ditakrifkan sebagai satu langkah yang digunakan bagi mengira persamaan pada piksel

individu. Konsep asas bagi kaedah yang dicadangkan adalah piksel yang mempunyai

perubahan akan menghasilkan perbezaan yang maksimum, i.e. nilai persamaan piksel

ini akan menjadi rendah. Kaedah ini telah diuji bagi mengesan kawasan perubahan dan

mengenal pasti sama ada perubahan itu disebabkan oleh pertumbuhan pokok atau

kehilangan pokok dalam empat kawasan di Hutan Bakau Matang, Perak. Imej daripada

satelite SPOT 5 dengan jalur 1, jalur 2, jalur 3 dan jalur 4 yang mempuyai resolusi

10meter bertarikh 5 Ogos 2005 dan 13 Jun 2007 telah digunakan untuk menguji kaedah

ini. Ianya kemudian akan dibandingkan dengan keputusan daripada Analisis Komponen

iv

v

Utama 1 (PCA1). Kawasan yang mengalami kehilangan pokok telah dikenalpasti

dengan jayanya di mana nilai setempat maklumat salingan piksel di kawasan tersebut

adalah kurang dan sama dengan sifar. Kaedah ini diperhalusi lagi untuk mengesan

secara tepat kawasan-kawasan yang berubah akibat daripada pertumbuhan pokok

dengan menggunakan kaedah pengembangan titik ambang pada histogram peratusan

purata perbezaan antara kebarangkalian bercantum dan kebarangkalian sut. Keputusan

menunjukkan bahawa nilai ambang yang terbaik untuk menentukan kawasan yang

berubah dengan kawasan yang tidak berubah adalah nilai sifar. Secara keseluruhannya,

dalam penentuan perubahan di kawasan hutan, band 3 memberikan nilai yang terbaik

berbanding jalur-jalur lain dalam semua kes. Berbanding dengan kaedah pembezaan

imej dan kaedah perbezaan normalisasi indeks tumbuhan (NDVI), kaedah ini bukan

sahaja dapat menyelesaikan masalah pemilihan nilai ambang dalam kaedah penentuan

perubahan tetapi juga memberikan peratusan kejayaan yang tertinggi di kawasan

keempat, kawasan kedua dan kawasan pertama dengan nilai masing-masing adalah

95.07%, 89.47% dan 87.66%. Kesimpulan hasil daripada keputusan kajian

menunjukkan maklumat salingan tempatan bukan saja berkesan digunakan sebagai

teknik pengesanan perubahan tetapi juga boleh digunakan untuk mengenalpasti

kawasan-kawasan pertumbuhan pokok dan kehilangan pokok.

vi

ACKNOWLEDGEMENTS

I am thankful to Allah for making things possible, Alhamdulillah. I would like to

express my sincere gratitude and thanks to my supervisor Dr. Siti Khairunniza Bejo for

her untiring guidance, assistance, encouragement and supervision during the conduct of

the study. I would also like to express my deep appreciation and gratitude to Dr. Helmi

Zulhaidi b. Mohd Shafri and Assoc. Prof. Dr. Abdul Rashid Mohamed Shariff for their

co-supervision, guidance and suggestions.

I am thankful to Mr. Mohd Suhaimi b Mohd Noor (PPHD (Laut)) from Pejabat Hutan

Daerah Larut & Matang, Perak for guidance, assistance providing the data and field trip

to the study area. Special thanks to all members and technicians at Faculty of

Engineering, Universiti Putra Malaysia and Institute of Advance (ITMA) for allowing

and providing the facilities to be used to complete my research and graduate study. I

would also like to record my gratitude to my lecturers and friends who always give

motivation and support to accomplish my study.

Utmost appreciation and heartfelt gratitude goes to my beloved parents and all family

members for their patience, understanding, support and sacrifices throughout my study.

vii

I certify that an Examination Committee met on XXXX to conduct the final examination of Mahirah binti Jahari on her Master of Science thesis entitled “Change Detection In Forest Area using Local Similarity Measure” in accordance with Universiti Pertanian Malaysia (Higher Degree) Act 1980 and Universiti Pertanian Malaysia (Higher Degree) Regulations 1981. The committee recommends that the candidate be awarded the relevant degree. Members of the Examination Committee were as follows: Ishak Aris, Ph.D. Associate Professor Faculty of Engineering Universiti Putra Malaysia (Chairman) Abdul Rahman Ramli, PhD Associate Professor Faculty of Engineering Universiti Putra Malaysia (Internal Examiner) Syamsiah Mashohor, PhD Faculty of Engineering Universiti Putra Malaysia (Internal Examiner) Mohd Zubir Mat Jafri, PhD School of Physic Universiti Sains Malaysia (External Examiner) _______________________ BUJANG KIM HUAT, PhD Professor/Deputy Dean School of Graduate Studies Universiti Putra Malaysia Date:

Deleted: ¶

viii

This thesis was submitted to the Senate of Universiti Putra Malaysia and has been accepted as fulfilment of the requirement for the degree of Master of Science. The members of the Supervisory Committee were as follows: Siti Khairunniza Bejo, PhD Senior Lecturer Faculty of Engineering Universiti Putra Malaysia (Chairman) Abdul Rashid Mohamed Shariff, PhD Associate Professor Faculty of Engineering Universiti Putra Malaysia (Member) Helmi Zulhaidi Mohd Shafri, PhD Senior Lecturer Faculty of Engineering Universiti Putra Malaysia (Member) ________________________________ HASANAH MOHD GHAZALI, PhD

Professor and Dean School of Graduate Studies Universiti Putra Malaysia

Date: 10 March 2011

ix

DECLARATION

I declare that the thesis is my original work except for quotations and citations which have been duly acknowledged. I also declare that it has not been previously, and is not concurrently submitted for any degree at Universiti Putra Malaysia or other institutions.

_________________________

MAHIRAH BINTI JAHARI

Date: 3 November 2010

TABLE OF CONTENTS

Page

ABSTRACT ii ABSTRAK iv ACKNOWLEDGEMENTS vi APPROVAL vii DECLARATION ix LIST OF TABLES xii LIST OF FIGURES xiii LIST OF ABBREVIATIONS xiv CHAPTER 1 INTRODUCTION 1

1.1 Problem Statement and Motivation 3 1.2 Objectives 5 1.3 Scope and Limitation of the Study 5 1.4 Thesis Outline 6

2 LITERATURE REVIEW 7 2.1 Mangrove Forest Area 7 2.2 Remote Sensing 9 2.2.1 SPOT 5 Satellite 12

2.2.2 Application of Remote Sensing 13 2.3 Change Detection 15

2.4 Similarity Measure 26 2.5 Summary 29

3 METHODOLOGY 31 3.1 Research Design 31

3.2 Study Area 32 3.2.1 Area 1 35 3.2.2 Area 2 37 3.2.3 Area 3 38 3.2.4 Area 4 39 3.2.5 Reference Image 41 3.3 Preprocessing 42 3.3.1 Geometric Correction 43 3.3.2 Principal Component Analysis 45 3.4 Local Mutual Information 46 3.5 Comparison with the Other Change Detection Method 51

x

xi

3.5.1 Normalized Difference Vegetation Image (NDVI) 51 3.5.2 Image Differencing 52 3.6 Accuracy Assessment 52

4 RESULTS AND DISCUSSION 54

4.1 Background 54 4.2 Principal Component Analysis 54 4.3 Experimental Results 56 4.3.1 Threshold Value Selection 56 4.3.2 Band and PCA 1 Images Comparison 62 4.3.3 Band Images Analysis 66 4.4 Type of Changes: Changes Caused by the Plant Growth and Caused by the Plant Loss

69

4.5 Band Combination 84 4.6 Other Change Detection Method 89

4.6.1 Image Differencing 89 4.6.2 Normalized Difference Vegetation Index (NDVI) 97 4.7 Accuracy Assessment 102 5 CONCLUSION AND RECOMMENDATIONS 112

5.1 Conclusions 112 5.2 Recommendations for Further Studies 115

REFERENCES 117 APPENDICES 124 BIODATA OF STUDENT 158 LIST OF PUBLICATION 159