Detection of rain in acoustic recordings of the environment using machine learning techniques

This thesis is concerned with the detection and prediction of rain in environmental recordings using different machine learning algorithms. The results obtained in this research will help ecologists to efficiently analyse environmental data and monitor biodiversity.

Noise Filtering of Process Execution Logs based on Outliers Detection

This paper presents a technique for the automated removal of noise from process execution logs. Noise is the result of data quality issues such as logging errors and manifests itself in the form of infrequent process behavior. The proposed technique generates an abstract representation of an event log as an automaton capturing the direct follows relations between event labels. This automaton is then pruned from arcs with low relative frequency and used to remove from the log those events not fitting the automaton, which are identified as outliers. The technique has been extensively evaluated on top of various auto- mated process discovery algorithms using both artificial logs with different levels of noise, as well as a variety of real-life logs. The results show that the technique significantly improves the quality of the discovered process model along fitness, appropriateness and simplicity, without negative effects on generalization. Further, the technique scales well to large and complex logs.

Safety envelope for security

We present an approach for detecting sensor spoofing attacks on a cyber-physical system. Our approach consists of two steps. In the first step, we construct a safety envelope of the system. Under nominal conditions (that is, when there are no attacks), the system always stays inside its safety envelope. In the second step, we build an attack detector: a monitor that executes synchronously with the system and raises an alarm whenever the system state falls outside the safety envelope. We synthesize safety envelopes using a modified machine learning procedure applied on data collected from the system when it is not under attack. We present experimental results that show effectiveness of our approach, and also validate the several novel features that we introduced in our learning procedure.

Is the sun too bright in Queensland? An approach to robust outdoor colour beacon detection

Using cameras onboard a robot for detecting a coloured stationary target outdoors is a difficult task. Apart from the complexity of separating the target from the background scenery over different ranges, there are also the inconsistencies with direct and reflected illumination from the sun,clouds, moving and stationary objects. They can vary both the illumination on the target and its colour as perceived by the camera. In this paper, we analyse the effect of environment conditions, range to target, camera settings and image processing on the reported colours of various targets. The analysis indicates the colour space and camera configuration that provide the most consistent colour values over varying environment conditions and ranges. This information is used to develop a detection system that provides range and bearing to detected targets. The system is evaluated over various lighting conditions from bright sunlight, shadows and overcast days and demonstrates robust performance. The accuracy of the system is compared against a laser beacon detector with preliminary results indicating it to be a valuable asset for long-range coloured target detection.

Obstacle detection for a mining vehicle using a 2D laser

This paper discusses a number of key issues for the development of robust obstacle detection systems for autonomous mining vehicles. Strategies for obstacle detection are described and an overview of the state-of-the-art in obstacle detection for outdoor autonomous vehicles using lasers is presented, with their applicability to the mining environment noted. The development of an obstacle detection system for a mining vehicle is then detailed. This system uses a 2D laser scanner as the prime sensor and combines dead-reckoning data with laser data to create local terrain maps. The slope of the terrain maps is then used to detect potential obstacles.

Issues in obstacle detection for autonomous mining and construction vehicles

This paper discusses a Dumber of key issues for the development of robust, obstacle detection systems for autonomous mining and construction vehicles. A taxonomy of obstacle detection systems is described; An overview of the state-of- the-art in obstacle detection for outdoor autonomous vehicles is presented with their applicability to the mining and construction environments noted. The issue of so-called fail-safe obstacle detection is then discussed. Finally, we describe the development of an obstacle detection system for a mining vehicle.

Fault detection of wind turbine drivetrain utilizing power-speed characteristics

Wind energy, being the fastest growing renewable energy source in the present world, requires a large number of wind turbines to transform wind energy into electricity. One factor driving the cost of this energy is the reliable operation of these turbines. Therefore, it is a growing requirement within the wind farm community, to monitor the operation of the wind turbines on a continuous basis so that a possible fault can be detected ahead of time. As the wind turbine operates in an environment of constantly changing wind speed, it is a challenging task to design a fault detection technique which can accommodate the stochastic operational behavior of the turbines. Addressing this issue, this paper proposes a novel fault detection criterion which is robust against operational uncertainty, as well as having the ability to quantify severity level specifically of the drivetrain abnormality within an operating wind turbine. A benchmark model of wind turbine has been utilized to simulate drivetrain fault condition and effectiveness of the proposed technique has been tested accordingly. From the simulation result it can be concluded that the proposed criterion exhibits consistent performance for drivetrain faults for varying wind speed and has linear relationship with the fault severity level.

From ImageNet to mining: Adapting visual object detection with minimal supervision

This paper presents visual detection and classification of light vehicles and personnel on a mine site.We capitalise on the rapid advances of ConvNet based object recognition but highlight that a naive black box approach results in a significant number of false positives. In particular, the lack of domain specific training data and the unique landscape in a mine site causes a high rate of errors. We exploit the abundance of background-only images to train a k-means classifier to complement the ConvNet. Furthermore, localisation of objects of interest and a reduction in computation is enabled through region proposals. Our system is tested on over 10km of real mine site data and we were able to detect both light vehicles and personnel. We show that the introduction of our background model can reduce the false positive rate by an order of magnitude.

Content-independent embedding scheme for multi-modal medical image watermarking

Background As the increasing adoption of information technology continues to offer better distant medical services, the distribution of, and remote access to digital medical images over public networks continues to grow significantly. Such use of medical images raises serious concerns for their continuous security protection, which digital watermarking has shown great potential to address. Methods We present a content-independent embedding scheme for medical image watermarking. We observe that the perceptual content of medical images varies widely with their modalities. Recent medical image watermarking schemes are image-content dependent and thus they may suffer from inconsistent embedding capacity and visual artefacts. To attain the image content-independent embedding property, we generalise RONI (region of non-interest, to the medical professionals) selection process and use it for embedding by utilising RONI’s least significant bit-planes. The proposed scheme thus avoids the need for RONI segmentation that incurs capacity and computational overheads. Results Our experimental results demonstrate that the proposed embedding scheme performs consistently over a dataset of 370 medical images including their 7 different modalities. Experimental results also verify how the state-of-the-art reversible schemes can have an inconsistent performance for different modalities of medical images. Our scheme has MSSIM (Mean Structural SIMilarity) larger than 0.999 with a deterministically adaptable embedding capacity. Conclusions Our proposed image-content independent embedding scheme is modality-wise consistent, and maintains a good image quality of RONI while keeping all other pixels in the image untouched. Thus, with an appropriate watermarking framework (i.e., with the considerations of watermark generation, embedding and detection functions), our proposed scheme can be viable for the multi-modality medical image applications and distant medical services such as teleradiology and eHealth.

Evaluation of object detection proposals under condition variations

Object detection is a fundamental task in many computer vision applications, therefore the importance of evaluating the quality of object detection is well acknowledged in this domain. This process gives insight into the capabilities of methods in handling environmental changes. In this paper, a new method for object detection is introduced that combines the Selective Search and EdgeBoxes. We tested these three methods under environmental variations. Our experiments demonstrate the outperformance of the combination method under illumination and view point variations.

Information Security and Privacy : Proceedings of the 9th Australasian Conference, ACISP 2004, Sydney, Australia, July 13-15, 2004.

Preface The 9th Australasian Conference on Information Security and Privacy (ACISP 2004) was held in Sydney, 13–15 July, 2004. The conference was sponsored by the Centre for Advanced Computing – Algorithms and Cryptography (ACAC), Information and Networked Security Systems Research (INSS), Macquarie University and the Australian Computer Society. The aims of the conference are to bring together researchers and practitioners working in areas of information security and privacy from universities, industry and government sectors. The conference program covered a range of aspects including cryptography, cryptanalysis, systems and network security. The program committee accepted 41 papers from 195 submissions. The reviewing process took six weeks and each paper was carefully evaluated by at least three members of the program committee. We appreciate the hard work of the members of the program committee and external referees who gave many hours of their valuable time. Of the accepted papers, there were nine from Korea, six from Australia, five each from Japan and the USA, three each from China and Singapore, two each from Canada and Switzerland, and one each from Belgium, France, Germany, Taiwan, The Netherlands and the UK. All the authors, whether or not their papers were accepted, made valued contributions to the conference. In addition to the contributed papers, Dr Arjen Lenstra gave an invited talk, entitled Likely and Unlikely Progress in Factoring. This year the program committee introduced the Best Student Paper Award. The winner of the prize for the Best Student Paper was Yan-Cheng Chang from Harvard University for his paper Single Database Private Information Retrieval with Logarithmic Communication. We would like to thank all the people involved in organizing this conference. In particular we would like to thank members of the organizing committee for their time and efforts, Andrina Brennan, Vijayakrishnan Pasupathinathan, Hartono Kurnio, Cecily Lenton, and members from ACAC and INSS.

The security and privacy of usage policies and provenance logs in an Information Accountability Framework

The potential benefits of shared eHealth records systems are promising for the future of improved healthcare. However, the uptake of such systems is hindered by concerns over the security and privacy of patient information. The use of Information Accountability and so called Accountable-eHealth (AeH) systems has been proposed to balance the privacy concerns of patients with the information needs of healthcare professionals. However, a number of challenges remain before AeH systems can become a reality. Among these is the need to protect the information stored in the usage policies and provenance logs used by AeH systems to define appropriate use of information and hold users accountable for their actions. In this paper, we discuss the privacy and security issues surrounding these accountability mechanisms, define valid access to the information they contain, discuss solutions to protect them, and verify and model an implementation of the access requirements as part of an Information Accountability Framework.

Complete-linkage clustering for voice activity detection in audio and visual speech

We propose a novel technique for conducting robust voice activity detection (VAD) in high-noise recordings. We use Gaussian mixture modeling (GMM) to train two generic models; speech and non-speech. We then score smaller segments of a given (unseen) recording against each of these GMMs to obtain two respective likelihood scores for each segment. These scores are used to compute a dissimilarity measure between pairs of segments and to carry out complete-linkage clustering of the segments into speech and non-speech clusters. We compare the accuracy of our method against state-of-the-art and standardised VAD techniques to demonstrate an absolute improvement of 15% in half-total error rate (HTER) over the best performing baseline system and across the QUT-NOISE-TIMIT database. We then apply our approach to the Audio-Visual Database of American English (AVDBAE) to demonstrate the performance of our algorithm in using visual, audio-visual or a proposed fusion of these features.

The QUT-NOISE-SRE protocol for the evaluation of noisy speaker recognition

The QUT-NOISE-SRE protocol is designed to mix the large QUT-NOISE database, consisting of over 10 hours of back- ground noise, collected across 10 unique locations covering 5 common noise scenarios, with commonly used speaker recognition datasets such as Switchboard, Mixer and the speaker recognition evaluation (SRE) datasets provided by NIST. By allowing common, clean, speech corpora to be mixed with a wide variety of noise conditions, environmental reverberant responses, and signal-to-noise ratios, this protocol provides a solid basis for the development, evaluation and benchmarking of robust speaker recognition algorithms, and is freely available to download alongside the QUT-NOISE database. In this work, we use the QUT-NOISE-SRE protocol to evaluate a state-of-the-art PLDA i-vector speaker recognition system, demonstrating the importance of designing voice-activity-detection front-ends specifically for speaker recognition, rather than aiming for perfect coherence with the true speech/non-speech boundaries.

Robotic detection and tracking of Crown-Of-Thorns starfish

This paper presents a novel vision-based underwater robotic system for the identification and control of Crown-Of-Thorns starfish (COTS) in coral reef environments. COTS have been identified as one of the most significant threats to Australia's Great Barrier Reef. These starfish literally eat coral, impacting large areas of reef and the marine ecosystem that depends on it. Evidence has suggested that land-based nutrient runoff has accelerated recent outbreaks of COTS requiring extensive use of divers to manually inject biological agents into the starfish in an attempt to control population numbers. Facilitating this control program using robotics is the goal of our research. In this paper we introduce a vision-based COTS detection and tracking system based on a Random Forest Classifier (RFC) trained on images from underwater footage. To track COTS with a moving camera, we embed the RFC in a particle filter detector and tracker where the predicted class probability of the RFC is used as an observation probability to weight the particles, and we use a sparse optical flow estimation for the prediction step of the filter. The system is experimentally evaluated in a realistic laboratory setup using a robotic arm that moves a camera at different speeds and heights over a range of real-size images of COTS in a reef environment.