Physical Sciences and Mathematics Commons^™

A Car Detection System Based On Hierarchical Visual Features, Fok Hing Chi Tivive, Abdesselam Bouzerdoum

Dr Fok Hing Chi Tivive

In this paper, we address the problem of detecting and localizing cars in still images. The proposed car detection system is based on a hierarchical feature detector in which the processing units are shunting inhibitory neurons. To reduce the training time and complexity of the network, the shunting inhibitory neurons in the first layer are implemented as directional nonlinear filters, whereas the neurons in the second layer have trainable parameters. A multi-resolution processing scheme is implemented so as to detect cars of different sizes, and to reduce the number of false positives during the detection stage, an adaptive thresholding strategy …

Kernel Pca Of Hog Features For Posture Detection, Peng Cheng, Wanqing Li, Philip Ogunbona

Associate Professor Wanqing Li

Motivated by the non-linear manifold learning ability of the Kernel Principal Component Analysis (KPCA), we propose in this paper a method for detecting human postures from single images by employing KPCA to learn the manifold span of a set of HOG features that can effectively represent the postures. The main contribution of this paper is to apply the KPCA as a non-linear learning and open-set classification tool, which implicitly learns a smooth manifold from noisy data that scatter over the feature space. For a new instance of HOG feature, its distance to the manifold that is measured by its reconstruction …

Human Detection Based On Weighted Template Matching, Duc Thanh Nguyen, Philip Ogunbona, Wanqing Li

Associate Professor Wanqing Li

This paper proposes a new two-stage human detection method involving matching and verification. A Bayesian framework is developed to verify the matching score obtained from a weighted distance measure. Performance evaluation indicates that the proposed method is able to utilize the flexible matching scheme and produce superior true positive, true negative and low misclassification rates.

Human Detection Using Local Shape And Non-Redundant Binary Patterns, Duc Thanh Nguyen, Wanqing Li, Philip Ogunbona

Associate Professor Wanqing Li

Motivated by the advantages of using shape matching technique in detecting objects in various postures and viewpoints and the discriminative power of local patterns in object recognition, this paper proposes a human detection method combining both shape and appearance cues. In particular, local shapes of the body parts are detected using template matching. Based on body parts' shapes, local appearance features are extracted. We introduce a novel local binary pattern (LBP) descriptor, called Non-Redundant LBP (NRLBP), to encode local appearance of human. The proposed method was evaluated and compared with other state-of-the-art human detection methods on two commonly used datasets: …

Ontology-Based Knowledge Representation For A P2p Multi-Agent Distributed Intrusion Detection System, Dayong Ye, Quan Bai, Minjie Zhang

Associate Professor Minjie Zhang

Many research efforts on application of ontology in network security have been done in the past decade. However, they mostly stop at initial proposal or focus on framework design without detailed representation of intrusion or attack and relevant detection knowledge with ontology. In this paper, the design and implementation of ontology-based knowledge representation for a peer-to-peer multi-agent distributed intrusion detection system (ontology-based MADIDS) are introduced. An example which demonstrates the representation of an attack with ontology and the relevant detection process is also presented. In ontology-Based MADIDS, ontology technique enables peers in the system and agents in one peer to …

A Car Detection System Based On Hierarchical Visual Features, Fok Hing Chi Tivive, Abdesselam Bouzerdoum

Professor Salim Bouzerdoum

In this paper, we address the problem of detecting and localizing cars in still images. The proposed car detection system is based on a hierarchical feature detector in which the processing units are shunting inhibitory neurons. To reduce the training time and complexity of the network, the shunting inhibitory neurons in the first layer are implemented as directional nonlinear filters, whereas the neurons in the second layer have trainable parameters. A multi-resolution processing scheme is implemented so as to detect cars of different sizes, and to reduce the number of false positives during the detection stage, an adaptive thresholding strategy …

Markov Random Fields For Abnormal Behavior Detection On Highways, Son Lam Phung, Philippe L. Bouttefroy, Abdesselam Bouzerdoum, Azeddine Beghdadi

Professor Salim Bouzerdoum

This paper introduces a new paradigm for abnormal behavior detection relying on the integration of contextual information in Markov random fields. Contrary to traditional methods, the proposed technique models the local density of object feature vector, therefore leading to simple and elegant criterion for behavior classification. We develop a Gaussian Markov random field mixture catering for multi-modal density and integrating the neighborhood behavior into a local estimate. The convergence of the random field is ensured by online learning through a stochastic clustering algorithm. The system is tested on an extensive dataset (over 2800 vehicles) for behavior modeling. The experimental results …

Reduced Training Of Convolutional Neural Networks For Pedestrian Detection, Giang Hoang Nguyen, Son Lam Phung, Abdesselam Bouzerdoum

Professor Salim Bouzerdoum

Pedestrian detection is a vision task with many practical applications in video surveillance, road safety, autonomous driving and military. However, it is much more difficult compared to the detection of other visual objects, because of the tremendous variations in the inner region as well as the outer shape of the pedestrian pattern. In this paper, we propose a pedestrian detection approach that uses convolutional neural network (CNN) to differentiate pedestrian and non-pedestrian patterns. Among several advantages, the CNN integrates feature extraction and classification into one single, fully adaptive structure. It can extract two-dimensional features at increasing scales, and it is …

Direct Detection Of Additives And Degradation Products From Polymers By Liquid Extraction Surface Analysis Employing Chip-Based Nanospray Mass Spectrometry, Martin Paine, Phillip Barker, Shane A. Maclaughlin, Todd W. Mitchell, Stephen J. Blanksby

Stephen Blanksby

Rationale: Polymer-based surface coatings in outdoor applications experience accelerated degradation due to exposure to solar radiation, oxygen and atmospheric pollutants. These deleterious agents cause undesirable changes to the polymers aesthetic and mechanical properties reducing its lifetime. The use of antioxidants such as hindered amine light stabilisers (HALS) retard these degradative processes, however, mechanisms for HALS action and polymer degradation are poorly understood. Methods: Detection of the hindered amine light stabiliser (HALS) TINUVIN®123 (bis (1-octyloxy-2,2,6,6-tetramethyl-4-piperidyl) sebacate) and the polymer degradation products directly from a polyester-based coil coating was achieved by liquid extraction surface analysis (LESA) coupled to a triple quadrupole QTRAP® …

Kernel Pca Of Hog Features For Posture Detection, Peng Cheng, Wanqing Li, Philip Ogunbona

Professor Philip Ogunbona

Motivated by the non-linear manifold learning ability of the Kernel Principal Component Analysis (KPCA), we propose in this paper a method for detecting human postures from single images by employing KPCA to learn the manifold span of a set of HOG features that can effectively represent the postures. The main contribution of this paper is to apply the KPCA as a non-linear learning and open-set classification tool, which implicitly learns a smooth manifold from noisy data that scatter over the feature space. For a new instance of HOG feature, its distance to the manifold that is measured by its reconstruction …

Human Detection Based On Weighted Template Matching, Duc Thanh Nguyen, Philip Ogunbona, Wanqing Li

Professor Philip Ogunbona

This paper proposes a new two-stage human detection method involving matching and verification. A Bayesian framework is developed to verify the matching score obtained from a weighted distance measure. Performance evaluation indicates that the proposed method is able to utilize the flexible matching scheme and produce superior true positive, true negative and low misclassification rates.

Human Detection Using Local Shape And Non-Redundant Binary Patterns, Duc Thanh Nguyen, Wanqing Li, Philip Ogunbona

Professor Philip Ogunbona

Motivated by the advantages of using shape matching technique in detecting objects in various postures and viewpoints and the discriminative power of local patterns in object recognition, this paper proposes a human detection method combining both shape and appearance cues. In particular, local shapes of the body parts are detected using template matching. Based on body parts' shapes, local appearance features are extracted. We introduce a novel local binary pattern (LBP) descriptor, called Non-Redundant LBP (NRLBP), to encode local appearance of human. The proposed method was evaluated and compared with other state-of-the-art human detection methods on two commonly used datasets: …

A Novel Template Matching Method For Human Detection, Duc Thanh Nguyen, Wanqing Li, Philip Ogunbona

Professor Philip Ogunbona

This paper proposes a novel weighted template matching method. It employs a generalized distance transform (GDT) and an orientation map (OM). The GDT allows us to weight the distance transform more on the strong edge points and the OM provides supplementary local orientation information for matching. Based on the matching method, a two-stage human detection method consisting of template matching and Bayesian verification is developed. Experimental results have shown that the proposed method can effectively reduce the false positive and false negative detection rates and perform superiorly in comparison to the conventional Chamfer matching method.

Object Detection Using Non-Redundant Local Binary Patterns, Duc Thanh Nguyen, Zhimin Zong, Philip Ogunbona, Wanqing Li

Professor Philip Ogunbona

Local Binary Pattern (LBP) as a descriptor, has been successfully used in various object recognition tasks because of its discriminative property and computational simplicity. In this paper a variant of the LBP referred to as Non-Redundant Local Binary Pattern (NRLBP) is introduced and its application for object detection is demonstrated. Compared with the original LBP descriptor, the NRLBP has advantage of providing a more compact description of object’s appearance. Furthermore, the NRLBP is more discriminative since it reflects the relative contrast between the background and foreground. The proposed descriptor is employed to encode human’s appearance in a human detection task. …

Preliminary Investigations Of Pigment Responses To Phylloxera Infestation, A L. Blanchfield, K S. Powell, Sharon A. Robinson

Sharon Robinson

Early detection of grape phylloxera (Daktulosphaira vitifoliae) infestation is vital for the implementation of post-outbreak quarantine in Australia. Remote sensing systems exploit changes in leaf pigment content associated with plant stress and offer a real possibility of a phylloxera-specific detection system. Pre-visual, symptomatic changes in the pigment content of phylloxera-infested grapevine leaves were investigated using high performance liquid chromatography (HPLC) as a potential aid to improve current phylloxera detection methods. A glasshouse trial was established to characterize the response of two grapevine varieties, Vitis vinifera ‘Cabernet Sauvignon’ and ‘Shiraz’, to phylloxera infestation, in a controlled environment. Field trials were conducted …

Predicting Disease Outbreaks Using A Support Vector Machine Model, Nicolae Dragu

Senior Theses and Projects

The purpose of this research is to create an efficient way of detecting disease outbreaks from news articles using Support Vector Machines (SVM). An SVM is a supervised machine learning method used for classification and regression problems. The role of the SVM in this project is to “learn” to distinguish between news articles that may indicate a disease outbreak and those that do not.

A series of health-related articles from the World Health Organization is parsed using a Java program in order to create vectors for the SVM. Each such article thus results in a vector. A basic negation detection …

Reaction Of The C2h Radical With 1-Butyne (C4h6): Low Temperature Kinetics And Isomer-Specific Product Detection, Satchin Soorkia, Adam J. Trevitt, Talitha M. Selby, David L. Osborn, Craig A. Taatjes, Kevin R. Wilson, Stephen R. Leone

Adam Trevitt

No abstract provided.

Reactions Of The Cn Radical With Benzene And Toluene: Product Detection And Low-Temperature Kinetics, Adam J. Trevitt, Fabien Goulay, Craig A. Taatjes, David L. Osborn, Stephen R. Leone

Adam Trevitt

Low-temperature rate coefficients are measured for the CN + benzene and CN + toluene reactions using the pulsed Laval nozzle expansion technique coupled with laser-induced fluorescence detection. The CN + benzene reaction rate coefficient at 105, 165, and 295 K is found to be relatively constant over this temperature range, (3.9−4.9) × 10−10 cm3 molecule−1 s−1. These rapid kinetics, along with the observed negligible temperature dependence, are consistent with a barrierless reaction entrance channel and reaction efficiencies approaching unity. The CN + toluene reaction is measured to have a rate coefficient of 1.3 × 10−10 cm3 molecule−1 s−1 at 105 …

A Novel Route To Copper(Ii) Detection Using 'Click' Chemistry-Induced Aggregation Of Gold Nanoparticles, Carol Hua, William H. Zhang, Swahnnya De Almeida, Simone Ciampi, Danmar Gloria, Guozhen Liu, Jason Brian Harper, J Justin Gooding

Australian Institute for Innovative Materials - Papers

A simple colorimetric method for the detection of copper ions in water is described. This method is based on the 'click' copper(i)-catalyzed azide-alkyne cycloaddition reaction and its use in promoting the aggregation of azide-tagged gold nanoparticles by a dialkyne cross-linker is described. Nanoparticle cross-linking, evidenced as a colour change, is used for the detection of copper ions. The lowest detected concentration by the naked eye was 1.8 μM, with the response linear with log(concentration) between 1.8-200 μM. The selectivity relative to other potentially interfering ions was evaluated.

Direct Detection Of Additives And Degradation Products From Polymers By Liquid Extraction Surface Analysis Employing Chip-Based Nanospray Mass Spectrometry, Martin Paine, Phillip Barker, Shane A. Maclaughlin, Todd W. Mitchell, Stephen J. Blanksby

Faculty of Science - Papers (Archive)

Rationale: Polymer-based surface coatings in outdoor applications experience accelerated degradation due to exposure to solar radiation, oxygen and atmospheric pollutants. These deleterious agents cause undesirable changes to the polymers aesthetic and mechanical properties reducing its lifetime. The use of antioxidants such as hindered amine light stabilisers (HALS) retard these degradative processes, however, mechanisms for HALS action and polymer degradation are poorly understood. Methods: Detection of the hindered amine light stabiliser (HALS) TINUVIN®123 (bis (1-octyloxy-2,2,6,6-tetramethyl-4-piperidyl) sebacate) and the polymer degradation products directly from a polyester-based coil coating was achieved by liquid extraction surface analysis (LESA) coupled to a triple quadrupole QTRAP® …

Detecting, Tracking, And Recognizing Activities In Aerial Video, Vladimir Reilly

Electronic Theses and Dissertations

In this dissertation, we address the problem of detecting humans and vehicles, tracking them in crowded scenes, and finally determining their activities in aerial video. Even though this is a well explored problem in the field of computer vision, many challenges still remain when one is presented with realistic data. These challenges include large camera motion, strong scene parallax, fast object motion, large object density, strong shadows, and insufficiently large action datasets. Therefore, we propose a number of novel methods based on exploiting scene constraints from the imagery itself to aid in the detection and tracking of objects. We show, …