Object categorization via sparse representation of local features

Sparse representation has been introduced to address many recognition problems in computer vision. In this paper, we propose a new framework for object categorization based on sparse representation of local features. Unlike most of previous sparse coding based methods in object classification that only use sparse coding to extract high-level features, the proposed method incorporates sparse representation and classification into a unified framework. Therefore, it does not need a further classifier. Experimental results show that the proposed method achieved better or comparable accuracy than the well known bag-of-features representation with various classifiers.

Human identification from ECG signals via sparse representation of local segments

This work proposes a novel framework to extract compact and discriminative features from Electrocardiogram (ECG) signals for human identification based on sparse representation of local segments. Specifically, local segments extracted from an ECG signal are projected to a small number of basic elements in a dictionary, which is learned from training data. A final representation is extracted by performing a max pooling procedure over all the sparse coefficient vectors in the ECG signal. Unlike most of existing methods for human identification from ECG signals which require segmentation of individual heartbeats or extraction of fiducial points, the proposed method does not need to segment individual heartbeats or detect any fiducial points. The method achieves an 99.48% accuracy on a 100 subjects dataset constructed from a publicly available database, which demonstrates that both local and global structural information are well captured to characterize the ECG signals.

Automatic attribution of ancient Roman imperial coins

Classification of coins is an important but laborious aspect of numismatics - the field that studies coins and currency. It is particularly challenging in the case of ancient coins. Due to the way they were manufactured, as well as wear from use and exposure to chemicals in the soil, the same ancient coin type can exhibit great variability in appearance. We demonstrate that geometry-free models of appearance do not perform better than chance on this task and that only a small improvement is gained by previously proposed models of combined appearance and geometry. Thus, our first major contribution is a new type of feature which is efficient in terms of computational time and storage requirements, and which effectively captures geometric configurations between descriptors corresponding to local features. Our second contribution is a description of a fully automatic system based on the proposed features, which robustly localizes, segments out and classifies coins from cluttered images. We also describe a large database of ancient coins that we collected and which will be made publicly available. Finally, we report the results of empirical comparison of different coin matching techniques. The features proposed in this paper are found to greatly outperform existing methods.

Walkability of local communities: using geographic information systems to objectively assess relevant environmental attributes.

Geographic Information Systems (GIS) can be used to objectively measure features of the built environment that may influence adults’ physical activity, which is an important determinant of chronic disease. We describe how a previously developed index of walkability was operationalised in an Australian context, using available spatial data. The index was used to generate a stratified sampling frame for the selection of households from 32 communities for the PLACE (Physical Activity in Localities and Community Environments) study. GIS data have the potential to be used to construct measures of environmental attributes and to develop indices of walkability for cities, regions or local communities.

Objectively assessing "Walkability" of local communities: using GIS to identify the relevant environmental attributes.

Geographic Information Systems (GIS) may be used to measure objectively, those features of the built environment that may influence walking. Public health research on environmental determinants of physical activity in adults shows that different factors can influence walking for recreation, compared to walking for transport. Most studies have used perceived (self-report) rather than objective measures of potentially relevant environmental attributes. We describe how a previously-developed index of ‘walkability’ was operationalized in an Australian context, using available spatial data.               Attributes believed to be of relevance to walking for transport, that are measurable using GIS, are: Dwelling density (higher-density neighborhoods support greater retail and service variety, resulting in shorter, walkable distances between facilities; driving and parking are more difficult and time consuming). Connectivity (higher intersection densities provide people with a greater variety of potential routes, easier access to major roads where public transport is available and shorter times to get to destinations). Land use mix (the more varied the land use mix and built form, then the more conducive it is to walk to various destinations). Net retail area (there are more options for destinations where goods and services may be purchased and more local employment opportunities that can be reached by walking). The associations of these attributes with walking behaviors can be  examined separately, or in combination. Such GIS data are very helpful in fundamental studies of the environmental determinants of behavior, and also in applied policy research for cities, regions or local communities, to
address public health and environmental issues.

Measuring the walkability of local communities using geographic information systems data

Researchers are working to identify and promote environment and policy initiatives to encourage more active and healthy communities. Measuring environmental attributes through objective means can verify which physical environment factors are most important. We describe how Geographic Information Systems (GIS) may be used to measure objectively, the features of the built environment that may influence walking. We show how four key attributes currently believed to be of most relevance to walking for transport may be used to create a ‘walkability’ index. These are dwelling density (higher-density neighbourhoods support greater retail and service variety, resulting in shorter, walkable distances between facilities; driving and parking are more difficult); street connectivity (higher intersection density provides people with a greater choice of potential routes, easier access to major roads where public transport is available and shorter times to get to destinations); land use mix (the more varied the land use mix and built form, the more conducive it is to walk to various destinations); and net retail area (people who live near multiple and diverse retail opportunities are able to make more frequent and shorter shopping trips by walking and can walk to more local employment opportunities). The potential relationships between each of the objective environmental-attribute measures and walking behaviours is discussed, together with suggestions as to how such measures might be used to guide community infrastructure planning. GIS mapping can assist decision makers in where to focus transportation investments and where to guide future growth. Readily accessible GIS data can be used to guide and support urban planning and infrastructure investment decisions in both the private and public sectors, to increase walking in communities.

Extracting fingerprint features using textures

Personal identification of individuals is becoming increasingly adopted in society today. Due to the large number of electronic systems that require human identification, faster and more secure identification systems are pursued. Biometrics is based upon the physical characteristics of individuals; of these the fingerprint is the most common as used within law enforcement. Fingerprint-based systems have been introduced into the society but have not been well received due to relatively high rejection rates and false acceptance rates. This limited acceptance of fingerprint identification systems requires new techniques to be investigated to improve this identification method and the acceptance of the technology within society. Electronic fingerprint identification provides a method of identifying an individual within seconds quickly and easily. The fingerprint must be captured instantly to allow the system to identify the individual without any technical user interaction to simplify system operation. The performance of the entire system relies heavily on the quality of the original fingerprint image that is captured digitally. A single fingerprint scan for verification makes it easier for users accessing the system as it replaces the need to remember passwords or authorisation codes. The identification system comprises of several components to perform this function, which includes a fingerprint sensor, processor, feature extraction and verification algorithms. A compact texture feature extraction method will be implemented within an embedded microprocessor-based system for security, performance and cost effective production over currently available commercial fingerprint identification systems. To perform these functions various software packages are available for developing programs for windows-based operating systems but must not constrain to a graphical user interface alone. MATLAB was the software package chosen for this thesis due to its strong mathematical library, data analysis and image analysis libraries and capability. MATLAB enables the complete fingerprint identification system to be developed and implemented within a PC environment and also to be exported at a later date directly to an embedded processing environment. The nucleus of the fingerprint identification system is the feature extraction approach presented in this thesis that uses global texture information unlike traditional local information in minutiae-based identification methods. Commercial solid-state sensors such as the type selected for use in this thesis have a limited contact area with the fingertip and therefore only sample a limited portion of the fingerprint. This limits the number of minutiae that can be extracted from the fingerprint and as such limits the number of common singular points between two impressions of the same fingerprint. The application of texture feature extraction will be tested using variety of fingerprint images to determine the most appropriate format for use within the embedded system. This thesis has focused on designing a fingerprint-based identification system that is highly expandable using the MATLAB environment. The main components that are defined within this thesis are the hardware design, image capture, image processing and feature extraction methods. Selection of the final system components for this electronic fingerprint identification system was determined by using specific criteria to yield the highest performance from an embedded processing environment. These platforms are very cost effective and will allow fingerprint-based identification technology to be implemented in more commercial products that can benefit from the security and simplicity of a fingerprint identification system.

Are safety-related features of the road environment associated with smaller declines in physical activity among youth?

This study examined how objective measures of the local road environment related to safety were associated with change in physical activity (including active transport) among youth. Few longitudinal studies have examined the impact of the road environment on physical activity among children/adolescents in their neighborhoods. Participants were children aged 8–9 years (n=170) and adolescents aged 13–15 years (n=276) in 2004. Data were collected in 2004 and 2006 during followup of participants recruited initially in 2001 from 19 primary schools in Melbourne, Australia. Walking/cycling to local destinations was parent-reported for children and self-reported by adolescents. Moderate-to-vigorous physical activity (MVPA) during nonschool hours was recorded using accelerometers. Road environment features in each participant’s neighborhood (area within 800 m radius of their home) were measured objectively using a Geographical Information System. Linear regression analyses examined associations between road features and changes in active transport (AT) and MVPA over 2 years. Children’s AT increased but MVPA levels decreased in both age groups; on average, younger girls recorded the greatest declines. The number of traffic/pedestrian lights was associated with ΔAT among younger girls (B=0.45, p=0.004). The total length of walking tracks (in meters) was associated with ΔAT among younger girls (B=0.0016, p=0.015) and adolescent girls (B=0.0016, p=0.002). For adolescent boys, intersection density was associated with ΔAT (B=0.03, p=0.030). Slow points were associated with ΔMVPA among younger boys before school (B=1.55, p=0.021), while speed humps were associated with ΔMVPA among adolescent boys after school (B=0.23, p=0.015). There were many associations for adolescent girls: for example, the total length of local roads (B=0.49, p=0.005), intersection density (B=0.05, p=0.036), and number of speed humps (B=0.33, p=0.020) were associated with ΔMVPA during nonschool hours. Safety-related aspects of the built environment are conducive to physical activity among youth and may help stem age-related declines in physical activity. Passive road safety interventions may promote AT and physical activity among less active girls, in particular.

Geographical Information Systems (GIS) and their role in sustainable planning : a case study from a Local Government Area (LGA) in Australia

A case study is used to demonstrate the application of Geographical Information Systems (GIS) to inform sustainable development. The suitability of the landscape to support tourism accommodation in a Local Government Area (LGA) is modelled by integrating existing datasets, including conservation areas, residential zones, major roads and known locations of tourism operators into a logistic regression framework. By using a data-driven approach an indication of the relative importance of each explanatory variable can be accounted for, therefore informing planners of the importance of different assets. In a region where tourism is reliant upon natural features, this use of information systems in conjunction with quantitative statistical modelling can value-add to existing datasets. The provision of this kind of knowledge is important as it would otherwise not factor into the decision-making process had the datasets been considered independently of each other – a concept that applies to both the public and private sectors.

Local energy and pre-envelope

We observe that the local energy is the pre-envelope for analytic function. The maxima and phase of this function can be used to compute and classify visual features such as motion and stereo disparity, texture, etc. We examine the construction of new filters for computing Local Energy, and compare these filters with the Gabor filters and the three-point-filter of Venkatesh and Owens.

On the classification of image features

While the primary purpose of edge detection schemes is to be able to produce an edge map of a given image, the ability to distinguish between different feature types is also of importance. In this paper we examine feature classification based on local energy detection and show that local energy measures are intrinsically capable of making this classification because of the use of odd and even filters. The advantage of feature classification is that it allows for the elimination of certain feature types from the edge map, thus simplifying the task of object recognition.

Local energy, the pre-envelope, and filter resolution

We examine the construction of new filters for computing local energy, and compare these filters with the Gabor filters and the three-point-filter of Venkatesh [l]. Further, we demonstrate that the effect of convolution with complex Gabor filters is to band-pass (with some differentiating effect) and compute the local energy of the result. The magnitude of the resulting local energy is then used to detect features [2], [3] (step features, texture etc.), and the phase is used to classify the detected features [l], [4] or provide disparity information for stereo [5] and motion work [6], [7]. Each of these types of information can be obtained at multiple resolutions, enabling the use of course to fine strategies for computing disparity, and allowing the discrimination of image textures on the basis of which parts of the Fourier domain they dominate [8], [9].

Multi-scale local pattern co-occurrence matrix for textural image classification

Textural image classification technologies have been extensively explored and widely applied in many areas. It is advantageous to combine both the occurrence and spatial distribution of local patterns to describe a texture. However, most existing state-of-the-art approaches for textural image classification only employ the occurrence histogram of local patterns to describe textures, without considering their co-occurrence information. And they are usually very time-consuming because of the vector quantization involved. Moreover, those feature extraction paradigms are implemented at a single scale. In this paper we propose a novel multi-scale local pattern co-occurrence matrix (MS_LPCM) descriptor to characterize textural images through four major steps. Firstly, Gaussian filtering pyramid preprocessing is employed to obtain multi-scale images; secondly, a local binary pattern (LBP) operator is applied on each textural image to create a LBP image; thirdly, the gray-level co-occurrence matrix (GLCM) is utilized to extract local pattern co-occurrence matrix (LPCM) from LBP images as the features; finally, all LPCM features from the same textural image at different scales are concatenated as the final feature vectors for classification. The experimental results on three benchmark databases in this study have shown a higher classification accuracy and lower computing cost as compared with other state-of-the-art algorithms.

Quantifying heteroskedasticity using slope of local variances index

In econometrics, heteroskedasticity refers to the case when the variances of the error terms of the data in hand are not equal. Heteroskedastic time series are challenging to different forecasting models. However, all available solutions adopt the strategy of accommodating heteroskedasticity in the time series and consider it as a type of noise. Some statistical tests were developed over the past three decades to determine whether a time series features heteroskedastic behaviour. This paper presents a novel strategy to handle this problem by deriving a quantifying measure for heteroskedasticity. The proposed measure relies on the definition of heteroskedasticity as a time-variant variance in the time series. In this work, heteroskedasticity is measured by calculating local variances using linear filters, estimating variance trends, calculating changes in variance slopes, and finally obtaining the average slope angle. The results confirm that the proposed index complies with the widely popular heteroskedasticity tests.