Time filter

Source Type

Fu Z.,Monash University | Robles-Kelly A.,Canberra Research Laboratory | Robles-Kelly A.,Australian National University | Zhou J.,Canberra Research Laboratory | Zhou J.,Australian National University
IEEE Transactions on Pattern Analysis and Machine Intelligence | Year: 2011

Multiple instance learning (MIL) is a paradigm in supervised learning that deals with the classification of collections of instances called bags. Each bag contains a number of instances from which features are extracted. The complexity of MIL is largely dependent on the number of instances in the training data set. Since we are usually confronted with a large instance space even for moderately sized real-world data sets applications, it is important to design efficient instance selection techniques to speed up the training process without compromising the performance. In this paper, we address the issue of instance selection in MIL. We propose MILIS, a novel MIL algorithm based on adaptive instance selection. We do this in an alternating optimization framework by intertwining the steps of instance selection and classifier learning in an iterative manner which is guaranteed to converge. Initial instance selection is achieved by a simple yet effective kernel density estimator on the negative instances. Experimental results demonstrate the utility and efficiency of the proposed approach as compared to the state of the art. © 2006 IEEE.

Fu Z.,Monash University | Robles-Kelly A.,Canberra Research Laboratory
IEEE Transactions on Geoscience and Remote Sensing | Year: 2011

In this paper, we develop a novel approach to object-material identification in spectral imaging by combining the use of invariant spectral absorption features and statistical machine-learning techniques. Our method hinges on the relevance of spectral absorption features for material identification and casts the problem into a pattern-recognition setting by making use of an invariant representation of the most discriminant band segments in the spectra. Thus, here, we view the identification problem as a classification task, which is effected based upon those invariant absorption segments in the spectra which are most discriminative between the materials under study. To robustly recover those bands that are most relevant to the identification process, we make use of discriminant learning. To illustrate the utility of our method for purposes of material identification, we perform experiments on both terrestrial and remotely sensed hyperspectral imaging data and compare our results to those yielded by an alternative. © 2006 IEEE.

Kusakunniran W.,University of New South Wales | Kusakunniran W.,Neville Roach Laboratory | Wu Q.,University of Technology, Sydney | Zhang J.,University of New South Wales | And 3 more authors.
IEEE Transactions on Circuits and Systems for Video Technology | Year: 2012

It is well recognized that gait is an important biometric feature to identify a person at a distance, e.g., in video surveillance application. However, in reality, change of viewing angle causes significant challenge for gait recognition. A novel approach using regression-based view transformation model (VTM) is proposed to address this challenge. Gait features from across views can be normalized into a common view using learned VTM(s). In principle, a VTM is used to transform gait feature from one viewing angle (source) into another viewing angle (target). It consists of multiple regression processes to explore correlated walking motions, which are encoded in gait features, between source and target views. In the learning processes, sparse regression based on the elastic net is adopted as the regression function, which is free from the problem of overfitting and results in more stable regression models for VTM construction. Based on widely adopted gait database, experimental results show that the proposed method significantly improves upon existing VTM-based methods and outperforms most other baseline methods reported in the literature. Several practical scenarios of applying the proposed method for gait recognition under various views are also discussed in this paper. © 2012 IEEE.

Kusakunniran W.,University of New South Wales | Wu Q.,University of Technology, Sydney | Zhang J.,Neville Roach Laboratory | Zhang J.,University of Technology, Sydney | And 2 more authors.
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics | Year: 2012

Gait has been known as an effective biometric feature to identify a person at a distance. However, variation of walking speeds may lead to significant changes to human walking patterns. It causes many difficulties for gait recognition. A comprehensive analysis has been carried out in this paper to identify such effects. Based on the analysis, Procrustes shape analysis is adopted for gait signature description and relevant similarity measurement. To tackle the challenges raised by speed change, this paper proposes a higher order shape configuration for gait shape description, which deliberately conserves discriminative information in the gait signatures and is still able to tolerate the varying walking speed. Instead of simply measuring the similarity between two gaits by treating them as two unified objects, a differential composition model (DCM) is constructed. The DCM differentiates the different effects caused by walking speed changes on various human body parts. In the meantime, it also balances well the different discriminabilities of each body part on the overall gait similarity measurements. In this model, the Fisher discriminant ratio is adopted to calculate weights for each body part. Comprehensive experiments based on widely adopted gait databases demonstrate that our proposed method is efficient for cross-speed gait recognition and outperforms other state-of-the-art methods. © 1996-2012 IEEE.

Engerer N.A.,Australian National University | Engerer N.A.,Canberra Research Laboratory | Mills F.P.,Australian National University | Mills F.P.,Space Science Institute
Solar Energy | Year: 2014

The rapidly growing installed base of distributed solar photovoltaic (PV) systems is causing increased interest in forecasting their power output. A key step towards this is accurately estimating the output from a PV system based on the known output from a nearby PV system. However, each PV system is unique with its own hardware configuration, orientation, shading, etc. Thus, the process of using the power output from one system to estimate the power output of another nearby system is not necessarily straightforward. In order to address these challenges, a modified clear-sky index for photovoltaics is proposed. This index is the ratio of the instantaneous PV power output to the instantaneous theoretical clear-sky power output derived from a clear-sky radiation model and PV system simulation routine. This definition performs better than previous clear-sky indices when both PV systems' characteristics are known and the two PV systems have similar orientations. Through this index, the performance of a nearby PV system can be predicted quite accurately. This is demonstrated through the analysis of power output data from five residential PV systems in Canberra, Australia. © 2014 Elsevier Ltd.

Engerer N.A.,Australian National University | Engerer N.A.,Canberra Research Laboratory | Mills F.P.,Australian National University | Mills F.P.,Space Science Institute
Solar Energy | Year: 2015

There have been many validation studies of clear sky solar radiation models, however, to date, no such analysis has been completed for Australia. Clear sky models are essential for estimating the generation potential of various solar energy technologies, the basic calibration of radiation measuring equipment, quality control of solar radiation datasets, engineering design (e.g. heating and cooling of buildings) and in agricultural and biological sciences (e.g. forestry). All of these areas are of considerable interest to the Australian economy and will benefit from an assessment of clear sky radiation models. With the recent provision of one-minute interval radiation data by the Australian Bureau of Meteorology for 20 sites across Australia, such a study can now be undertaken at a level not previously possible. Using up to ten years of data from each of 14 of these sites, clear sky periods are extracted through an automated detection algorithm. With these clear sky periods identified, nine of the most prominent beam and global clear sky radiation models are assessed using the relative Mean Bias Error, relative Root Mean Square Error and Coefficient of Determination as metrics. Further testing assessed model performance as a function of solar zenith angle and apparent solar time. Results show that for global clear sky simulations, the Solis, Esra and REST2 approaches perform best, while the Iqbal, Esra and REST2 methods are the most proficient clear sky beam models. © 2015 Elsevier Ltd.

Wang H.,Australian National University | Wang H.,Canberra Research Laboratory | Zhou X.,University of New South Wales | Reed M.C.,Australian National University
IEEE Transactions on Wireless Communications | Year: 2013

This paper studies the information-theoretic secrecy performance in large-scale cellular networks based on a stochastic geometry framework. The locations of both base stations and mobile users are modeled as independent two-dimensional Poisson point processes. We consider two important features of cellular networks, namely, information exchange between base stations and cell association, to characterize their impact on the achievable secrecy rate of an arbitrary downlink transmission with a certain portion of the mobile users acting as potential eavesdroppers. In particular, tractable results are presented under diverse assumptions on the availability of eavesdroppers' location information at the serving base station, which captures the benefit from the exchange of the location information between base stations. © 2002-2012 IEEE.

Du L.,Macquarie University | Buntine W.,Canberra Research Laboratory | Jin H.,CSIRO
EMNLP-CoNLL 2012 - 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, Proceedings of the Conference | Year: 2012

Topic models are increasingly being used for text analysis tasks, often times replacing earlier semantic techniques such as latent semantic analysis. In this paper, we develop a novel adaptive topic model with the ability to adapt topics from both the previous segment and the parent document. For this proposed model, a Gibbs sampler is developed for doing posterior inference. Experimental results show that with topic adaptation, our model significantly improves over existing approaches in terms of perplexity, and is able to uncover clear sequential structure on, for example, Herman Melville's book "Moby Dick". © 2012 Association for Computational Linguistics.

Du L.,Macquarie University | Buntine W.,Canberra Research Laboratory | Johnson M.,Macquarie University
NAACL HLT 2013 - 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Main Conference | Year: 2013

We present a new hierarchical Bayesian model for unsupervised topic segmentation. This new model integrates a point-wise boundary sampling algorithm used in Bayesian segmentation into a structured topic model that can capture a simple hierarchical topic structure latent in documents. We develop an MCMC inference algorithm to split/merge segment(s). Experimental results show that our model outperforms previous unsupervised segmentation methods using only lexical information on Choi's datasets and two meeting transcripts and has performance comparable to those previous methods on two written datasets. © 2013 Association for Computational Linguistics.

Qian Y.,Zhejiang University | Zhou J.,Canberra Research Laboratory | Zhou J.,Australian National University | Ye M.,Zhejiang University | Wang Q.,Zhejiang University
International Geoscience and Remote Sensing Symposium (IGARSS) | Year: 2011

Sparse modeling is a powerful framework for data analysis and processing. It is especially useful for high-dimensional regression and classification problems in which a large number of feature variables exist but the amount of training samples is limited. In this paper, we address the problems of feature description, feature selection and classifier design for hyperspectral images using structured sparse models. A linear sparse logistic regression model is proposed to combine feature selection and pixel classification into a regularized optimization problem with the constraint of sparsity. To explore the structured features, three-dimensional discrete wavelet transform (3D-DWT) is employed, which processes the hyperspectral data cube as a whole tensor instead of adapting the data to a vector or matrix. This allows more effective capturing of the spatial and spectral structure. The structure of the 3D-DWT features is imposed on the sparse model by group LASSO which selects the features on the group level. The advantages of our method are validated on the real hyperspectral data. © 2011 IEEE.

Loading Canberra Research Laboratory collaborators
Loading Canberra Research Laboratory collaborators