Open Access. Powered by Scholars. Published by Universities.®
Physical Sciences and Mathematics Commons™
Open Access. Powered by Scholars. Published by Universities.®
- File Type
Articles 31 - 60 of 76
Full-Text Articles in Physical Sciences and Mathematics
Human Detection Using Local Shape And Non-Redundant Binary Patterns, Duc Thanh Nguyen, Wanqing Li, Philip Ogunbona
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: …
Texture Analysis Using Gabor Wavelets, Golshah Naghdy, Jianli Wang, Philip Ogunbona
Texture Analysis Using Gabor Wavelets, Golshah Naghdy, Jianli Wang, Philip Ogunbona
Professor Philip Ogunbona
Receptive field profiles of simple cells in the visual cortex have been shown to resemble even- symmetric or odd-symmetric Gabor filters. Computational models employed in the analysis of textures have been motivated by two-dimensional Gabor functions arranged in a multi-channel architecture. More recently wavelets have emerged as a powerful tool for non-stationary signal analysis capable of encoding scale-space information efficiently. A multi-resolution implementation in the form of a dyadic decomposition of the signal of interest has been popularized by many researchers. In this paper, Gabor wavelet configured in a 'rosette' fashion is used as a multi-channel filter-bank feature extractor for …
Image Content Annotation Based On Visual Features, Lei Ye, Philip Ogunbona, J. Wang
Image Content Annotation Based On Visual Features, Lei Ye, Philip Ogunbona, J. Wang
Professor Philip Ogunbona
Automatic image content annotation techniques attempt to explore structural visual features of images that describe image content and associate them with image semantics. In this paper, two types of concept spaces, atomic concept and collective concept spaces, are defined and the annotation problems in those spaces are formulated as feature classification and Bayesian inference, respectively. A scheme of image content annotation in this framework is presented and evaluated as an application of photo categorization using MPEG-7 VCE2 dataset and its ground truth. The experimental results show a promising performance.
Edge Image Description Using Fractal Interpolation, P Motallebi, P O. Ogunbona
Edge Image Description Using Fractal Interpolation, P Motallebi, P O. Ogunbona
Professor Philip Ogunbona
Edge images derived from compressed image databases are described using fractal techniques. The proposed method is able to give affine transformation-invariant description suitable for use in a query-by-example database application. Comparison among the proposed method, polynomial interpolation and spline interpolation is given. It is concluded that fractal interpolation can give a compact description of image contours and is able to cope with random perturbation of the coordinates of the contour points by as much as 25 percent.
Real-Time Facial Feature Point Extraction, Ce Zhan, Wanqing Li, Philip Ogunbona, Farzad Safaei
Real-Time Facial Feature Point Extraction, Ce Zhan, Wanqing Li, Philip Ogunbona, Farzad Safaei
Professor Philip Ogunbona
Localization of facial feature points is an important step for many subsequent facial image analysis tasks. In this paper, we proposed a new coarse-to-fine method for extracting 20 facial feature points from image sequences. In particular, the Viola-Jones face detection method is extended to detect small-scale facial components with wide shape variations, and linear Kalman filters are used to smoothly track the feature points by handling detection errors and head rotations. The proposed method achieved higher than 90% detection rate when tested on the BioID face database and the FG-NET facial expression database. Moreover, our method shows robust performance against …
Detecting Humans Under Occlusion Using Variational Mean Field Method, Duc Thanh Nguyen, Philip Ogunbona, Wanqing Li
Detecting Humans Under Occlusion Using Variational Mean Field Method, Duc Thanh Nguyen, Philip Ogunbona, Wanqing Li
Professor Philip Ogunbona
This paper proposes a human detection method using variational mean field approximation for occlusion reasoning. In the method, parts of human objects are detected individually using template matching. Initial detection hypotheses with spatial layout information are represented in a graphical model and refined through a Bayesian estimation. In this paper, mean field method is employed for such an estimation. The proposed method was evaluated on the popular CAVIAR-INRIA dataset. Experimental results show that the proposed algorithm is able to detect humans in severe occlusion within reasonable processing time.
Coding Gain And Spatial Localisation Properties Of Discrete Wavelet Transform Filters For Image Coding, J Andrew, P Ogunbona, F Paoloni
Coding Gain And Spatial Localisation Properties Of Discrete Wavelet Transform Filters For Image Coding, J Andrew, P Ogunbona, F Paoloni
Professor Philip Ogunbona
The authors consider coding gain and spatial localisation properties of DWT filters for still image compression. Using a JPEG type quantisation and encoding method several images are compressed using a DWT implemented using various two-band subband filter sets. It is concluded that a relatively high coding gain (relative to a highly correlated source) is necessary, but not sufficient, for good image coding performance. Further, it is observed that low spatial width filters are desirable, particularly in regard to reduced ringing distortion. In terms of the tradeoff between coding gain and spatial localisation, and in terms of actual coding performance, it …
A Novel Template Matching Method For Human Detection, Duc Thanh Nguyen, Wanqing Li, Philip Ogunbona
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.
A Real-Time Facial Expression Recognition System For Online Games, Ce Zhan, Wanqing Li, Philip Ogunbona, Farzad Safaei
A Real-Time Facial Expression Recognition System For Online Games, Ce Zhan, Wanqing Li, Philip Ogunbona, Farzad Safaei
Professor Philip Ogunbona
Multiplayer online games (MOGs) have become increasingly popular because of the opportunity they provide for collaboration, communication, and interaction. However, compared with ordinary human communication, MOG still has several limitations, especially in communication using facial expressions. Although detailed facial animation has already been achieved in a number of MOGs, players have to use text commands to control the expressions of avatars. In this paper, we propose an automatic expression recognition system that can be integrated into an MOG to control the facial expressions of avatars. To meet the specific requirements of such a system, a number of algorithms are studied, …
On The Step Response Of The Dct, Jim Andrew, Philip Ogunbona
On The Step Response Of The Dct, Jim Andrew, Philip Ogunbona
Professor Philip Ogunbona
We show that the discrete cosine transform (DCT) is the best orthogonal transform, in terms of energy packing efficiency, for coding input steps of uniformly distributed random phase. Over sufficiently small block sizes, edges in an image can be modeled as such step inputs. This characteristic of the DCT, coupled with its high energy packing efficiency for highly correlated data, helps explain the impressive performance of the DCT for image compression.
Method Of Color Interpolation In A Single Sensor Color Camera Using Green Channel Separation, Chaminda Weerasinghe, Igor Kharitonenko, Philip Ogunbona
Method Of Color Interpolation In A Single Sensor Color Camera Using Green Channel Separation, Chaminda Weerasinghe, Igor Kharitonenko, Philip Ogunbona
Professor Philip Ogunbona
This paper presents a color interpolation algorithm for a single sensor color camera. The proposed algorithm is especially designed to solve the problem of pixel crosstalk among the pixels of different color channels. Interchannel cross-talk gives rise to blocking effects on the interpolated green plane, and also spreading of false colors into detailed structures. The proposed algorithm separates the green channel into two planes, one highly correlated with the red channel and the other with the blue channel. These separate planes are used for red and blue channel interpolation. Experiments conducted on McBeth color chart and natural images have shown …
A Part-Based Template Matching Method For Multi-View Human Detection, Duc Thanh Nguyen, Wanqing Li, Philip Ogunbona
A Part-Based Template Matching Method For Multi-View Human Detection, Duc Thanh Nguyen, Wanqing Li, Philip Ogunbona
Professor Philip Ogunbona
This paper proposes a part-based template matching method for multi-view human detection. The proposed method includes two stages: matching and verification. In particular, the best individual matching parts given a detection window are determined using an improved template matching algorithm. The hypothesis of the matched parts forming a human is then verified by employing a Bayesian-based model. The verification is not only based on the matching costs of individual parts but also how well the combining the matched parts satisfying the configuration constraints of the human body. Experimental results have shown that the proposed method is robust for detecting humans …
A Pixel-Based Robust Image Watermarking System, Wenming Lu, Wanqing Li, Rei Safavi-Naini, Philip Ogunbona
A Pixel-Based Robust Image Watermarking System, Wenming Lu, Wanqing Li, Rei Safavi-Naini, Philip Ogunbona
Professor Philip Ogunbona
Robust image watermarking systems are required to be resistant to geometric attacks in addition to common image processing tasks, such as JPEG compression. However, robustness against geometric attacks, such as rotation, scaling and translation, still remains one of the most challenging research topics in image watermarking. We propose a new pixel-based watermarking system in which a binary logo is embedded, a bit per pixel, in the pixel domain of an image. The encoder of the proposed system is based on a sliding window embedding scheme that applies the local average quantization index modulation (QIM), to achieve geometric attack robustness. The …
Semi-Supervised Maximum A Posteriori Probability Segmentation Of Brain Tissues From Dual-Echo Magnetic Resonance Scans Using Incomplete Training Data, Wanqing Li, P Ogunbona, C Desilva, Y Attikiouzel
Semi-Supervised Maximum A Posteriori Probability Segmentation Of Brain Tissues From Dual-Echo Magnetic Resonance Scans Using Incomplete Training Data, Wanqing Li, P Ogunbona, C Desilva, Y Attikiouzel
Professor Philip Ogunbona
This study presents a stochastic framework in which incomplete training data are used to boost the accuracy of segmentation and to optimise segmentation when images under consideration are corrupted by inhomogeneities. The authors propose a semi-supervised maximum a posteriori probability (ssMAP) segmentation method that is able to utilise any amount of training data that are usually insufficient for supervised segmentation. The ssMAP unifies supervised and unsupervised segmentation and takes the two as its special cases. To deal with inhomogeneities, the authors propose to incorporate a bias field into the ssMAP and present an algorithm (referred to as ssMAPe) for simultaneous …
Visual Information Processing And Content Management: An Overview, Philip Ogunbona
Visual Information Processing And Content Management: An Overview, Philip Ogunbona
Professor Philip Ogunbona
Visual information processing and the management of visual content has become a significant part of contemporary economy. The visual information processing pipeline is divided into several modules including, (i) capture and enhancement, (ii) efficient representation for storage and transmission, (iii) processing for efficient and secure distribution, and, (iv) representation for efficient archiving and retrieval. Advances in semiconductor technology and optimum signal processing models and algorithms, provide tools to improve each module of the processing pipeline. Insight from other areas of study including psychology augments and informs the models being developed to understand and design efficient visual content management systems. The …
Similarity Measures For Compressed Image Databases, P Sangassapaviriya, Philip Ogunbona
Similarity Measures For Compressed Image Databases, P Sangassapaviriya, Philip Ogunbona
Professor Philip Ogunbona
For image database applications it is desirable that functions such as searching, browsing and partial recall be done without the need to totally decompress the image. This has the advantage of alleviating possible burden and degradation that the network may suffer. Edge images derived from wavelet-compressed images are considered as index that can be queried by example. Zernike moment invariants are used as descriptors for the index edge image and the query sketch image. The descriptions are compared for the purpose of database searching. The query images were allowed to undergo translation, rotation, scaling and some deformation. Simulation results gave …
Wavelet-Based Feature-Adaptive Adaptive Resonance Theory Neural Network For Texture Identification, Jiazhao Wang, Golshah Naghdy, Philip Ogunbona
Wavelet-Based Feature-Adaptive Adaptive Resonance Theory Neural Network For Texture Identification, Jiazhao Wang, Golshah Naghdy, Philip Ogunbona
Professor Philip Ogunbona
A new method of texture classification comprising two processing stages, namely a low-level evolutionary feature extraction based on Gabor wavelets and a high-level neural network based pattern recognition, is proposed. The design of these stages is motivated by the processes involved in the human visual system: low-level receptors responsible for early vision processing and the high-level cognition. Gabor wavelets are used as extractors of ‘‘lowlevel’’ features that feed the feature-adaptive adaptive resonance theory (ART) neural network acting as a high-level ‘‘cognitive system.’’ The novelty of the model developed in this paper lies in the use of a self-organizing input layer …
Content-Based Retrieval From Compressed-Image Databases, Philip Ogunbona, P Sangassapaviriya
Content-Based Retrieval From Compressed-Image Databases, Philip Ogunbona, P Sangassapaviriya
Professor Philip Ogunbona
There is an enormous amount of multi-media data including images, video, speech, audio and text, distributed among the various computer nodes on the Internet. The extent to which a user wiU be able to derive useful information from these data depends largely on the ease with which required data can be retrieved from the databases. The share volume of the data also poses a storage constraint on the databases; hence these data will need to exist in the compressed form on the databases. In this paper we concentrate on image data and propose a new paradigm in which a compressed …
Index Compressed Image Adaptive Vector Quantisation, Jamshid Shanbehzadeh, Philip Ogunbona
Index Compressed Image Adaptive Vector Quantisation, Jamshid Shanbehzadeh, Philip Ogunbona
Professor Philip Ogunbona
This paper introduces an improved image adaptive vector quantisation technique - index compressed image adaptive vector quantisation (IC-IAVQ). Despite its advantage over the universal codebook VQ, basic image adaptive VQ (IAVQ) is still suboptimum; it neglects the correlation among block indices in the encoded image. The new technique, IC-IAVQ, overcomes this suboptimality through a pre-processing and lossless compression of block indices. Simulation results using several images show that IC-IAVQ outperforms IAVQ and entropy coded IAVQ, especially at low bit-rates by about 2dB on average.
New Feature-Based Image Adaptive Vector Quantisation Coder, Jamshid Shanbehzadeh, Philip O. Ogunbona
New Feature-Based Image Adaptive Vector Quantisation Coder, Jamshid Shanbehzadeh, Philip O. Ogunbona
Professor Philip Ogunbona
It is difficult to achieve a good low bit rate image compression performance with traditional block coding schemes such as transform coding and vector quantization, without regard for the human visual perception or signal dependency. These classical block coding schemes are based on minimizing the MSE at a certain rate. This procedure results in more bits being allocated to areas which may not be visually important and the resulting quantization noise manifests as a blocking artifact. Blocking artifacts are known to be psychologically more annoying than white noise when the human visual response is considered. While image adaptive vector quantization …
Securing Wavelet Compression With Random Permutations, Takeyuki Uehara, Reihaneh Safavi-Naini, Philip Ogunbona
Securing Wavelet Compression With Random Permutations, Takeyuki Uehara, Reihaneh Safavi-Naini, Philip Ogunbona
Professor Philip Ogunbona
Wavelet compression for digital images achieves very high compression with reasonably high image quality and so is widely used in various applications. Adding security to compression algorithms has been proposed in a number of compression systems with the aim reducing the overall cost of compression and encryption. In this paper we propose a combined compression and encryption system based on wavelet transform and examine its security. Our results show that with a relatively small added cost varying degrees of security can be obtained while maintaining the performance of the compression system.
Message From The Dean Of The Faculty, Philip Ogunbona
Message From The Dean Of The Faculty, Philip Ogunbona
Professor Philip Ogunbona
The Faculty of Informatics at the University of Wollongong is proud to be co-hosting the 2010 IEEE International Symposium on Technology and Society. It is very fitting that this symposium is being held at the University of Wollongong where a purpose-built faculty has been created to educate graduates who will become leaders and provide technological solutions to the complex problems in the information and communication industry and society at large.
A Fast Algorithm For Color Image Segmentation, L. Dong, P. Ogunbona, Wanqing Li, G. Yu, L. Fan, G. Zheng
A Fast Algorithm For Color Image Segmentation, L. Dong, P. Ogunbona, Wanqing Li, G. Yu, L. Fan, G. Zheng
Professor Philip Ogunbona
Based on K-means and a two-layer pyramid structure, a fast algorithm is proposed for color image segmentation. The algorithm employs two strategies. Firstly, a two-layer structure of a color image is established. Then, an improved K-means with integer based lookup table implementation is applied to each layer. The clustering result on the upper layer (lower resolution) is used to guide the clustering in the lower layer (higher resolution). Experiments have shown that the proposed algorithm is significantly faster than the original K-means algorithm while producing comparable segmentation results.
Compression Tolerant Dct Based Image Hash, C. Kailasanathan, R. Safavi-Naini, P. Ogunbona
Compression Tolerant Dct Based Image Hash, C. Kailasanathan, R. Safavi-Naini, P. Ogunbona
Professor Philip Ogunbona
With the advent of Internet image authentication has become a central part of research in security. Since JPEG has recommended discrete cosine transform as one of the steps in image compression systems, a hash function which utilizes discrete cosine decomposition is desirable. In this paper, we propose a discrete cosine based hash function which distinguishes acceptable level of compression from image processing modifications such as Median filtering, Gaussian noise addition, and FMLR attack. To increase manipulation detection, we optimize the number of AC coefficients needed in smoothing.
Human Motion Simulation And Action Corpus, Gang Zheng, Wanqing Li, Philip Ogunbona, Liju Dong, Igor Kharitonenko
Human Motion Simulation And Action Corpus, Gang Zheng, Wanqing Li, Philip Ogunbona, Liju Dong, Igor Kharitonenko
Professor Philip Ogunbona
Acquisition of large scale good quality training samples is becoming a major issue in machine learning based human motion analysis. This paper presents a method to simulate continuous gross human body motion with the intention to establish a human motion corpus for learning and recognition. The simulation is achieved by a temporal-spatialtemporal decomposition of human motion into actions, joint actions and actionlets based on the human kinematic model. The actionlet models the primitive moving phase of a joint and represents the muscle movement governed by kinesiological principles. Joint actions and body actions are constructed from actionlets through constrained concatenation and …
Smoke Detection In Videos Using Non-Redundant Local Binary Pattern-Based Features, Hongda Tian, Wanqing Li, Philip Ogunbona, Duc Thanh Nguyen, Ce Zhan
Smoke Detection In Videos Using Non-Redundant Local Binary Pattern-Based Features, Hongda Tian, Wanqing Li, Philip Ogunbona, Duc Thanh Nguyen, Ce Zhan
Professor Philip Ogunbona
This paper presents a novel and low complexity method for real-time video-based smoke detection. As a local texture operator, Non-Redundant Local Binary Pattern (NRLBP) is more discriminative and robust to illumination changes in comparison with original Local Binary Pattern (LBP), thus is employed to encode the appearance information of smoke. Non-Redundant Local Motion Binary Pattern (NRLMBP), which is computed on the difference image of consecutive frames, is introduced to capture the motion information of smoke. Experimental results show that NRLBP outperforms the original LBP in the smoke detection task. Furthermore, the combination of NRLBP and NRLMBP, which can be considered …
Private Fingerprint Matching, Siamak Shahandashti, Reihaneh Safavi-Naini, Philip Ogunbona
Private Fingerprint Matching, Siamak Shahandashti, Reihaneh Safavi-Naini, Philip Ogunbona
Professor Philip Ogunbona
We propose a fully private fingerprint matching protocol that compares two fingerprints based on the most widely-used minutia-based fingerprint matching algorithm. The protocol enables two parties, each holding a private fingerprint, to find out if their fingerprints belong to the same individual. Unlike previous works, we do not make any simplifying assumption on the matching algorithm or use generic multiparty computation protocols in our constructions. We employ a commonly-used algorithm that works by first comparing minutia pairs from the two fingerprints based on their types, locations, and orientations, and then checking if the number of matching minutia pairs is more …
Illumination Invariant Face Detection Using Classifier Fusion, Alister Cordiner, Philip Ogunbona, Wanqing Li
Illumination Invariant Face Detection Using Classifier Fusion, Alister Cordiner, Philip Ogunbona, Wanqing Li
Professor Philip Ogunbona
An approach to the problem of illumination variations in face detection that uses classifier fusion is presented. Multiple face detectors are seperately trained for different illumination environments and their results are combined using a combination rule. To define the illumination environments, the training samples are clustered based on their illumination using unsupervised training. Different methods of clustering the samples and combining the outputs of the classifiers are examined. Experiments with the AR face database show that the proposed method achieves higher accuracy than the traditional monolithic face detection method.
Optimal Image Watermark Decoding, Wenming Lu, Wanqing Li, Reihaneh Safavi-Naini, Philip Ogunbona
Optimal Image Watermark Decoding, Wenming Lu, Wanqing Li, Reihaneh Safavi-Naini, Philip Ogunbona
Professor Philip Ogunbona
Not much has been done in utilizing the available information at the decoder to optimize the decoding performance of watermarking systems. This paper focuses on analyzing different decoding methods, namely, Minimum Distance, Maximum Likelihood and Maximum a-posteriori decoding given varying information at the decoder in the blind detection context. Specifically, we propose to employ Markov random fields to model the prior information given the embedded message is a structured logo. The application of these decoding methods in Quantization Index Modulation systems shows that the decoding performance can be improved by Maximum Likelihood decoding that exploits the property of the attack …
Age Estimation Based On Extended Non-Negative Matrix Factorization, Ce Zhan, Wanqing Li, Philip Ogunbona
Age Estimation Based On Extended Non-Negative Matrix Factorization, Ce Zhan, Wanqing Li, Philip Ogunbona
Professor Philip Ogunbona
Previous studies suggested that local appearance-based methods are more efficient than geometric-based and holistic methods for age estimation. This is mainly due to the fact that age information are usually encoded by the local features such as wrinkles and skin texture on the forehead or at the eye corners. However, the variations of theses features caused by other factors such as identity, expression, pose and lighting may be larger than that caused by aging. Thus, one of the key challenges of age estimation lies in constructing a feature space that could successfully recovers age information while ignoring other sources of …