(For USM Staff/Student Only)

EngLib USM > Ω School of Electrical & Electronic Engineering >

Real-time detection system for elephant and dangerous wildlife intrusion along the forest-outskirt villages

Real-time detection system for elephant and dangerous wildlife intrusion along the forest-outskirt villages / Law Hoi Chian
Sejak zaman dahulu, untuk tujuan pembangunan, manusia terpaksa mengeksploitasi alam sekitar. Aktiviti- aktiviti manusia ini telah menyebabkan hidupan-hidupan liar yang kehilangan habitatterpaksa mencari habitat yang lebih selamat dan penuh dengan sumber-sumber seperti makanan dan bekalan air, iaitu habitat manusia yang tinggal berhampiran dengan kawasan hutan. Gajah, beruang madu, harimau dan babi hutan merupakan hidupan yang paling kerap menceroboh ke dalam habitat manusia. Oleh itu, projek ini bertujuan untuk merangka satu sistem pintar yang mampu mengesan pencerobohan haiwan-haiwan yang berbahaya ke habitat manusia sebelum mereka merosakkan harta benda manusia seperti kawasan perladangan dan kawasan perumahan. Haiwan-haiwan sasaran projek ini ialah gajah, beruang madu, harimau dan babi hutan. Sistem ini dirangka dengan dua kaedah, iaitu Beg Perkataan Visi (BoW) dan Rangkaian Neural Buatan (ANN). Algoritma mengekstrak ciri-ciri imej yang digunakan ialah Ciri-ciri Teguh yang Dipercepatkan (SURF) dan ciri-ciri berdasarkan warna. Manakala algoritma pembelajaran mesin yang digunakan untuk membina sistem pintar ini adalah Mesin sokongan vector (SVM) dan Rangkaian Neural Suapan Rambatan Balik ke Hadapan. Beberapa prosedur yang berbeza telah dijalankan untuk mengaji pengaruh parameter-parameter sistem pintar tersebut terhadap prestasi sistem tersebut untuk mengesan haiwan-haiwan sasaran. Untuk sistem Beg Perkataan Visi (BoW), parameter yang diuji adalah jenis set imej untuk melatih sistem pintar tersebut, teknik-teknik pra-proses imej latih dan uji, Nisbah latih dengan uji untuk set gambar latihan dan nilai ambang untuk sistem. Bagi sistem Rangkaian Neural Buatan (ANN), parameter yang diuji adalah jenis set imej untuk melatih sistem pintar tersebut, teknik-teknik pra-proses imej latih dan uji, bilangan neuron dalam rangkaian neural, nilai ambang dan bilangan kali sistem pintar dilatih semula. Selepas dibina dengan parameter yang terunggul, prestasi kedua-dua sistem pintar tersebut telah dinilai. Sistem BoW telah dipilih untuk membina sistem pintar tersebut, dengan ketepatan 69.29%, sensitiviti 76.32% dan kekhususan 61.02%, berbanding dengan sistem ANN yang bernilai 55.93%, 66.89% dan 41.38%. _______________________________________________________________________________________________________ Over the years, human have to exploit the nature for the sake of globalization growth. These activitiesare forcing the wildlife to move their habitat to a safer and more resourceful place; i.e., the habitat of human beings. Dangerous animals may intrude into the habitat of human beings for food or shelter. Elephant herds, tiger, sun bear and wild boar contributes the most for human-wildlife conflict in peninsular Malaysia. Hence, this project aims to build a real-time image detection of the intrusion of animalsmoving towards human population, especially at forest-outskirt areas. In this project, computer vision based detection is used, which an artificial intelligent system is developed to detect the intrusion of dangerous wildlife before they approach and destroy to the fences of the village. The target animals to be detected are elephant, sun bear, tiger and wild boar. The recognition system is modelled by two different methods; the Bag of Words (BoW) and Artificial Neural Network (ANN). Feature extraction algorithm to extract the features of the training images are Speeded-up Robust Features (SURF) and Color-based features. The learning algorithm to develop the recognition model is Support Vector Machines and feed forward back propagation of Artificial Neural Networks. Different testing procedures are done to test the performance of the recognition model and perform fine tuning of the training parameter of both the recognition models. The parameters to be tested to the BoW model develop the ideal recognition are: the type of training images, image preprocessing methods to the training and testing data set, train to validation ratio of the training image set and threshold value. For ANN recognition model, the parameters to be tested are type of training images, image preprocessing methods to the training and testing data set, number of neurons in the hidden layer, threshold value and number of retrain. The performance of two recognition models with ideal parameters are compared in terms of accuracy, sensitivity and specificity. The BoW model is selected to develop the recognition model, with accuracy of 69.29%, 76.32% overall sensitivity and 61.02% specificity, higher than the ANN model of 55.93% accuracy, 66.89% overall sensitivity and 41.38% specificity.
Contributor(s):
Law Hoi Chian - Author
Primary Item Type:
Final Year Project
Identifiers:
Accession Number : 875007208
Barcode : 00003107086
Language:
English
Subject Keywords:
globalization growth; wildlife; move their habitat
First presented to the public:
6/1/2017
Original Publication Date:
4/17/2018
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Electrical & Electronic Engineering
Citation:
Extents:
Number of Pages - 99
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2018-04-17 12:52:56.061
Date Last Updated
2019-01-07 11:24:32.9118
Submitter:
Mohd Jasnizam Mohd Salleh

All Versions

Thumbnail Name Version Created Date
Real-time detection system for elephant and dangerous wildlife intrusion along the forest-outskirt villages1 2018-04-17 12:52:56.061