TraSe algorithm : automatic evolutionary morphing of electronic dance music

The TraSe (Transform-Select) algorithm has been developed to investigate the morphing of electronic music through automatically applying a series of deterministic compositional transformations to the source, guided towards a target by similarity metrics. This is in contrast to other morphing techniques such as interpolation or parameters or probabilistic variation. TraSe allows control over stylistic elements of the music through user-defined weighting of numerous compositional transformations. The formal evaluation of TraSe was mostly qualitative and occurred through nine participants completing an online questionnaire. The music generated by TraSe was generally felt to be less coherent than a human composed benchmark but in some cases judged as more creative.

Semantic and context-based retrieval of digital cultural objects [Published in Chinese]

Cultural objects are increasingly generated and stored in digital form, yet effective methods for their indexing and retrieval still remain an important area of research. The main problem arises from the disconnection between the content-based indexing approach used by computer scientists and the description-based approach used by information scientists. There is also a lack of representational schemes that allow the alignment of the semantics and context with keywords and low-level features that can be automatically extracted from the content of these cultural objects. This paper presents an integrated approach to address these problems, taking advantage of both computer science and information science approaches. We firstly discuss the requirements from a number of perspectives: users, content providers, content managers and technical systems. We then present an overview of our system architecture and describe various techniques which underlie the major components of the system. These include: automatic object category detection; user-driven tagging; metadata transform and augmentation, and an expression language for digital cultural objects. In addition, we discuss our experience on testing and evaluating some existing collections, analyse the difficulties encountered and propose ways to address these problems.

Enhancing digital road map with lane details extracted from large-scale stereo aerial imagery using object-oriented image analysis

Precise, up-to-date and increasingly detailed road maps are crucial for various advanced road applications, such as lane-level vehicle navigation, and advanced driver assistant systems. With the very high resolution (VHR) imagery from digital airborne sources, it will greatly facilitate the data acquisition, data collection and updates if the road details can be automatically extracted from the aerial images. In this paper, we proposed an effective approach to detect road lane information from aerial images with employment of the object-oriented image analysis method. Our proposed algorithm starts with constructing the DSM and true orthophotos from the stereo images. The road lane details are detected using an object-oriented rule based image classification approach. Due to the affection of other objects with similar spectral and geometrical attributes, the extracted road lanes are filtered with the road surface obtained by a progressive two-class decision classifier. The generated road network is evaluated using the datasets provided by Queensland department of Main Roads. The evaluation shows completeness values that range between 76% and 98% and correctness values that range between 82% and 97%.

Analyzing web multimedia query reformulation behavior

Current multimedia Web search engines still use keywords as the primary means to search. Due to the richness in multimedia contents, general users constantly experience some difficulties in formulating textual queries that are representative enough for their needs. As a result, query reformulation becomes part of an inevitable process in most multimedia searches. Previous Web query formulation studies did not investigate the modification sequences and thus can only report limited findings on the reformulation behavior. In this study, we propose an automatic approach to examine multimedia query reformulation using large-scale transaction logs. The key findings show that search term replacement is the most dominant type of modifications in visual searches but less important in audio searches. Image search users prefer the specified search strategy more than video and audio users. There is also a clear tendency to replace terms with synonyms or associated terms in visual queries. The analysis of the search strategies in different types of multimedia searching provides some insights into user’s searching behavior, which can contribute to the design of future query formulation assistance for keyword-based Web multimedia retrieval systems.

Depression, anxiety and body image after treatment for invasive stage one epithelial ovarian cancer

Background: Diagnosis of epithelial ovarian cancer (EOC) in young women has major implications including those to their reproductive potential. We evaluated depression, anxiety and body image in patients with stage I EOC treated with fertility sparing surgery (FSS) or radical surgery (RS). We also investigated fertility outcomes after FSS.----- Methods: A retrospective study was undertaken in which 62 patients completed questionnaires related to anxiety, depression, body image and fertility outcomes. Additional information on adjuvant therapy after FSS and RS and demographic details were abstracted from medical records. Both bi and multivariate regression models were used to assess the relationship between demographic, clinical and pathological results and scores for anxiety, depression and body image.----- Results: Thirty-nine patients underwent RS and the rest, FSS. The percentage of patients reporting elevated anxiety and depression (subscores ≥ 11) were 27 % and 5% respectively. The median (inter quartile range) score for body image scale (BIS) was 6 (3-15). None of the demographic or clinical factors examined showed significant association with anxiety and BIS with the exception of ‘time since diagnosis’. For depression, post-menopausal status was the only independent predictor. Among those 23 patients treated by FSS, 14 patients tried to conceive (7 successful), resulting in 7 live births, one termination of pregnancy and one miscarriage.----- Conclusion: This study shows that psychological issues are common in women treated for stage I EOC. Reproduction after FSS is feasible and lead to the birth of healthy babies in about half of patients who wished to have another child. Further prospective studies with standardised instruments are required.

Design and evaluation of an image analysis platform for low-power, low-bandwidth camera networks

We describe the design and evaluation of a platform for networks of cameras in low-bandwidth, low-power sensor networks. In our work to date we have investigated two different DSP hardware/software platforms for undertaking the tasks of compression and object detection and tracking. We compare the relative merits of each of the hardware and software platforms in terms of both performance and energy consumption. Finally we discuss what we believe are the ongoing research questions for image processing in WSNs.

Improved GrabCut segmentation via GMM optimisation

Semi-automatic segmentation of still images has vast and varied practical applications. Recently, an approach "GrabCut" has managed to successfully build upon earlier approaches based on colour and gradient information in order to address the problem of efficient extraction of a foreground object in a complex environment. In this paper, we extend the GrabCut algorithm further by applying an unsupervised algorithm for modelling the Gaussian Mixtures that are used to define the foreground and background in the segmentation algorithm. We show examples where the optimisation of the GrabCut framework leads to further improvements in performance.

Automatic solder joint defect classification using the Log-Gabor filter

This paper proposes the validity of a Gabor filter bank for feature extraction of solder joint images on Printed Circuit Boards (PCBs). A distance measure based on the Mahalanobis Cosine metric is also presented for classification of five different types of solder joints. From the experimental results, this methodology achieved high accuracy and a well generalised performance. This can be an effective method to reduce cost and improve quality in the production of PCBs in the manufacturing industry.

Road network extraction with new vectorization and pruning from high-resolution RS images

With the increasing resolution of remote sensing images, road network can be displayed as continuous and homogeneity regions with a certain width rather than traditional thin lines. Therefore, road network extraction from large scale images refers to reliable road surface detection instead of road line extraction. In this paper, a novel automatic road network detection approach based on the combination of homogram segmentation and mathematical morphology is proposed, which includes three main steps: (i) the image is classified based on homogram segmentation to roughly identify the road network regions; (ii) the morphological opening and closing is employed to fill tiny holes and filter out small road branches; and (iii) the extracted road surface is further thinned by a thinning approach, pruned by a proposed method and finally simplified with Douglas-Peucker algorithm. Lastly, the results from some QuickBird images and aerial photos demonstrate the correctness and efficiency of the proposed process.

Image synthesis based on a model of human vision

Modern computer graphics systems are able to construct renderings of such high quality that viewers are deceived into regarding the images as coming from a photographic source. Large amounts of computing resources are expended in this rendering process, using complex mathematical models of lighting and shading. However, psychophysical experiments have revealed that viewers only regard certain informative regions within a presented image. Furthermore, it has been shown that these visually important regions contain low-level visual feature differences that attract the attention of the viewer. This thesis will present a new approach to image synthesis that exploits these experimental findings by modulating the spatial quality of image regions by their visual importance. Efficiency gains are therefore reaped, without sacrificing much of the perceived quality of the image. Two tasks must be undertaken to achieve this goal. Firstly, the design of an appropriate region-based model of visual importance, and secondly, the modification of progressive rendering techniques to effect an importance-based rendering approach. A rule-based fuzzy logic model is presented that computes, using spatial feature differences, the relative visual importance of regions in an image. This model improves upon previous work by incorporating threshold effects induced by global feature difference distributions and by using texture concentration measures. A modified approach to progressive ray-tracing is also presented. This new approach uses the visual importance model to guide the progressive refinement of an image. In addition, this concept of visual importance has been incorporated into supersampling, texture mapping and computer animation techniques. Experimental results are presented, illustrating the efficiency gains reaped from using this method of progressive rendering. This visual importance-based rendering approach is expected to have applications in the entertainment industry, where image fidelity may be sacrificed for efficiency purposes, as long as the overall visual impression of the scene is maintained. Different aspects of the approach should find many other applications in image compression, image retrieval, progressive data transmission and active robotic vision.

Improving automatic speaker verification using SVM techniques

Automatic recognition of people is an active field of research with important forensic and security applications. In these applications, it is not always possible for the subject to be in close proximity to the system. Voice represents a human behavioural trait which can be used to recognise people in such situations. Automatic Speaker Verification (ASV) is the process of verifying a persons identity through the analysis of their speech and enables recognition of a subject at a distance over a telephone channel { wired or wireless. A significant amount of research has focussed on the application of Gaussian mixture model (GMM) techniques to speaker verification systems providing state-of-the-art performance. GMM's are a type of generative classifier trained to model the probability distribution of the features used to represent a speaker. Recently introduced to the field of ASV research is the support vector machine (SVM). An SVM is a discriminative classifier requiring examples from both positive and negative classes to train a speaker model. The SVM is based on margin maximisation whereby a hyperplane attempts to separate classes in a high dimensional space. SVMs applied to the task of speaker verification have shown high potential, particularly when used to complement current GMM-based techniques in hybrid systems. This work aims to improve the performance of ASV systems using novel and innovative SVM-based techniques. Research was divided into three main themes: session variability compensation for SVMs; unsupervised model adaptation; and impostor dataset selection. The first theme investigated the differences between the GMM and SVM domains for the modelling of session variability | an aspect crucial for robust speaker verification. Techniques developed to improve the robustness of GMMbased classification were shown to bring about similar benefits to discriminative SVM classification through their integration in the hybrid GMM mean supervector SVM classifier. Further, the domains for the modelling of session variation were contrasted to find a number of common factors, however, the SVM-domain consistently provided marginally better session variation compensation. Minimal complementary information was found between the techniques due to the similarities in how they achieved their objectives. The second theme saw the proposal of a novel model for the purpose of session variation compensation in ASV systems. Continuous progressive model adaptation attempts to improve speaker models by retraining them after exploiting all encountered test utterances during normal use of the system. The introduction of the weight-based factor analysis model provided significant performance improvements of over 60% in an unsupervised scenario. SVM-based classification was then integrated into the progressive system providing further benefits in performance over the GMM counterpart. Analysis demonstrated that SVMs also hold several beneficial characteristics to the task of unsupervised model adaptation prompting further research in the area. In pursuing the final theme, an innovative background dataset selection technique was developed. This technique selects the most appropriate subset of examples from a large and diverse set of candidate impostor observations for use as the SVM background by exploiting the SVM training process. This selection was performed on a per-observation basis so as to overcome the shortcoming of the traditional heuristic-based approach to dataset selection. Results demonstrate the approach to provide performance improvements over both the use of the complete candidate dataset and the best heuristically-selected dataset whilst being only a fraction of the size. The refined dataset was also shown to generalise well to unseen corpora and be highly applicable to the selection of impostor cohorts required in alternate techniques for speaker verification.