Waleed Abdulla

University of Auckland · Electrical Computer and Software Engineering

Background: Hyperspectral imaging systems face numerous challenges in acquiring accurate spatial-spectral hypercubes due to sample surface heterogeneity, environmental instability, and instrumental noise. Preprocessing strategies such as outlier detection, calibration, smoothing, and normalization are typically employed to address these issues, sel...

This paper presents a study on improving the performance of the acoustic piezoelectric transducer system in air, as the low acoustic impedance of air leads to suboptimal system performance. Impedance matching techniques can enhance the acoustic power transfer (APT) system’s performance in air. This study integrates an impedance matching circuit int...

Human behavior recognition technology is widely adopted in intelligent surveillance, human-machine interaction, video retrieval, and ambient intelligence applications. To achieve efficient and accurate human behavior recognition, a unique approach based on the hierarchical patches descriptor (HPD) and approximate locality-constrained linear coding...

Optical coherence tomography (OCT) is a noninvasive imaging technique that provides high-resolution cross-sectional retina images, enabling ophthalmologists to gather crucial information for diagnosing various retinal diseases. Despite its benefits, manual analysis of OCT images is time-consuming and heavily dependent on the personal experience of...

The need for contactless vascular biometric systems has significantly increased. In recent years, deep learning has proven to be efficient for vein segmentation and matching. Palm and finger vein biometrics are well researched; however, research on wrist vein biometrics is limited. Wrist vein biometrics is promising due to it not having finger or p...

The deep convolutional neural network (DCNN) in computer vision has given promising results. It is widely applied in many areas, from medicine, agriculture, self-driving car, biometric system, and almost all computer vision-based applications. Filters or weights are the critical elements responsible for learning in DCNN. Backpropagation has been th...

This paper develops a new approach to fraud detection in honey. Specifically, we examine adulterating honey with sugar and use hyperspectral imaging and machine learning techniques to detect adulteration. The main contributions of this paper are introducing a new feature smoothing technique to conform to the classification model used to detect the...

Honey from different botanical and geographical origins differ significantly in their market value due to their quality, flavor or health benefits. However, the high value of honey and increasing demand have motivated fraudulent acts of honey. It is subjected to frequent adulteration by mislabeling, direct or indirect inclusion of cheaper sweetener...

Identifying honey botanical origins and analyzing honey products of the same floral origin from different honey product brands are crucial to protect consumers’ interest. Hyperspectral imaging is a promising approach to differentiate various honey products. In this study, the honey hyperspectral imaging dataset, which contains 56 New Zealand honey...

The use of deep learning methods to extract vascular biometric patterns from the palm surface has been of interest among researchers in recent years. In many biometric recognition tasks, there is a limit in the number of training samples. This is because of limited vein biometric databases being available for research. This restricts the applicatio...

This study carries out a comparative investigation on the classification performance of Support Vector Machine (SVM) and Convo-lutional Neural Networks (CNN) for palm vein recognition. A relatively small dataset was used to reduce the computational complexity associated with SVM. The investigation is carried out considering the HK PolyU Multispectr...

This paper develops an approach to Manuka honey quality classification, using feature reduction and support vector machines (SVMs). We have developed a database of hyperspectral images of honey, including different quality Manuka honey. This paper focusses on machine learning techniques, specifically SVMs, to classify the quality of Manuka honey. W...

The present study proposes a new approach to automated screening of Clinically Significant Macular Edema (CSME) and addresses two major challenges associated with such screenings, i.e., exudate segmentation and imbalanced datasets. The proposed approach replaces the conventional exudate segmentation based feature extraction by combining a pre-train...

Biometric‐enabled systems have become highly desirable to authenticate and identify people due to the proliferation of mobile technologies. Their application in general settings, such as automated border gates using ePassports or at entrances to buildings or rooms, has become popular as well. Regrettably, where current biometry scanning technologie...

Biometric recognition methods using human traits like fingerprint, face, voice, palm-print, and palm vein have developed significantly in recent years. Palm vein recognition has gained attention because of its unique characteristics and high recognition accuracy. Many palm vein recognition methods proposed recently suffer from the issue of having l...

The development of a robust honey classification model based on hyperspectral imaging requires finding significant wavelength bands describing honey botanical origins. Significant wavelength bands could be discovered by using feature selection methods that each method is commonly used as a standalone method. This paper proposes a strategy of combin...

The present study proposes a new approach to automated screening of Clinically Significant Macular Edema (CSME) and addresses two major challenges associated with such screenings, i.e., exudate segmentation and imbalanced datasets. The proposed approach replaces the conventional exudate segmentation based feature extraction by combining a pre-train...

This chapter shadows light over the fundamental imaging techniques used in the field of ophthalmology; namely fundus retinal imaging. Since it is a noninvasive way of imaging the retina, the method is popular among the ophthalmic community and the targeted population. Fundus retinal images reveal vital information about the health status of a perso...

The conventional adaptive noise control algorithms create a zone of quiet at the location of an error microphone. Several virtual active noise control (ANC) algorithms have been suggested in the literature to introduce a more flexible positioning of the zone of quiet. The main objective of the proposed ANC system in our research is to minimize the...

Active noise control (ANC) is an effective way to cancel the low-frequency noise. The conventional ANC system creates the 'zone of quiet' by minimizing the mean square error (MSE) at the location of an error microphone. However, in practical applications, sometimes it is not possible to achieve the noise attenuation at the desired location due to p...

Palm vein biometrics has received a lot of attention in recent years. This technology offers accuracy, robustness and is contactless, which makes it a promising option for clinical applications. It uses palm vascular patterns of individuals as identification metric to match the identity. As per observations, the vein structure beneath the palm surf...

Honey has been growing as a trade commodity that has a significant impact on the economy. Development of honey quality assessment methods is vital to protect consumers from fraudulence. Automatic and non-invasive methods are interesting concepts which can deal with problems of conventional chemical-based methods and may complement them. A combinati...

Diabetic retinopathy (DR) is one of the leading causes of avertible blindness worldwide. Early detection of the disease can help to save the vision of diabetic patients. Presence of exudates, hemorrhages, and microaneurysms indicate an unhealthy eye image. Deep learning models have triumphed in image recognition, object detection and biomedical sig...

Hyperspectral imaging as a fast and non-invasive method for honey analysis has great potential to overcome the drawbacks of the conventional chemically based assessment methods. This paper discusses segmentation and calibration techniques to acquire accurate and consistent spectra in the implementation of hyperspectral imaging. These essential tech...

Vessel segmentation from the fundus retinal images is highly significant in diagnosing many pathologies related to eye and other systemic diseases. Even though there are many methods in the literature focusing on this task, most of these methods are not focusing on the small peripheral vessels segmentation. In this paper, we propose a new approach...

Automatic retinal image analysis has remained an important topic of research in the last ten years. Various algorithms and methods have been developed for analysing retinal images. The majority of these methods use public retinal image databases for performance evaluation without first examining the retinal image quality. Therefore, the performance...

Feature extraction from retinal images is gaining popularity worldwide as many pathologies are proved having connections with these features. Automatic detection of these features makes it easier for the specialist ophthalmologists to analyse them without spending exhaustive time to segment them manually. The proposed method automatically detects t...

In the Computational Auditory Scene Analysis (CASA) method, the ideal ratio mask or alternatively the ideal binary mask is the key point to reconstruct the enhanced signal. The ratio mask in its Wiener filtering or its square root form is currently considered. However, this kind of ratio mask overlooked one important issue. It does not exploit the...

The implementation of the parametric acoustic array has invariably being based on Berktay's far-field solution, despite its limitations. In addition to the problems due to the approximations of the model, recent studies suggest that the frequency response of the transducers add substantial errors in the demodulation process, which are not considere...

In computational auditory scene analysis, the accurate estimation of binary mask or ratio mask plays a key role in noise masking. An inaccurate estimation often leads to some artifacts and temporal discontinuity in the synthesized speech. To overcome this problem, we propose a new ratio mask estimation method in terms of Wiener filtering in each Ga...

In conventional computational auditory scene analysis, the segment segregation and pitch estimation are necessary. However, it is hard to obtain an accurate pitch contour of clean speech for segregating the segments in low signal-to-noise ratio. This often leads to an inaccurate estimation of binary mask and produces artifacts and temporal disconti...

The rapid growth of the industry has a major effect on the environmental noise pollution, and it ranks second to the air pollution that adversely affects the human health. Passive noise control techniques are impractical and very expensive for low-frequency noises. To solve such acoustic problem active noise control has been studied since early 20t...

Diabetic retinopathy is a micro-vascular disease that affects the vision. People with diabetes are more likely to get affected by diabetic retinopathy. There are many screening techniques available to diagnose this pathology. Majority of this screening methods use the retinal image of the patient for the diagnosis. Recently many automatic retinal i...

Abstract: Segmentation of retinal vessels from retinal fundus images is the key step in the automatic retinal image analysis. In this paper we propose a new unsupervised automatic method to segment the retinal vessels from retinal fundus images. Contrast enhancement and illumination correction are carried out through a series of image processing st...

In this paper, we present a new approach for the progressive compression of three-dimensional (3D) mesh geometry using redundant frame dictionaries and sparse approximation techniques. We construct the proposed frames from redundant linear combinations of the eigenvectors of a combinatorial mesh Laplacian matrix. We achieve a sparse synthesis of th...

In this paper, we propose a novel noise masking method based on Computational Auditory Scene Analysis by using an adaptive factor. Although it has succeeded in the field of speech separation and speech enhancement to some extent, the usage of fixed thresholds used for segregation and labeling heavily affects the processing performance. Focusing on...

The explosive growth in fingerprint technologies within the past decade has seen the emergence of a dedicated field of research into securing fingerprint templates during storage in a database. While new fingerprint template protection techniques are often broadly classified as belonging to the well-known salting, noninvertible transforms, key bind...

As fingerprints continue toward ubiquity in human recognition applications, growing fingerprint databases will pose an increasingly greater risk of irreversible identity theft in the event of a database breach. Consequently, more focus is being placed on researching new and effective ways of securing fingerprint templates during database storage. R...

This paper recasts the problem of online secondary path modeling in the form of a statistical inverse problem. A statistical and, in particular, a Bayesian approach towards secondary path modeling is developed and the computational issues that emerge from this approach are discussed. All signals and parameters are modeled as random variables and th...

This paper investigates the propagation of primary and secondary waves produced by parametric loudspeakers of different sizes. The theory of the parametric acoustic array describes the nonlinear interaction of waves to be confined to the near-field, but the nonlinearities may remain over the far-field, producing different results. Four simulations...

The high demand for genuine honey leads to fraud practices in the market which have disadvantaged top graded genuine honey production. The conventional chemical analysis procedures are usually used to ensure the quality and authenticity of honey. Yet, some drawbacks, such as time-consuming, laborious, invasive and required complex sample preparatio...

A significant challenge in the development of automated fingerprint recognition algorithms is dealing with intra-class variance among multiple samples of the same fingerprint. A major contributor to this intra-class variance is the inconsistency with which a finger is presented to the fingerprint scanner across multiple authentication attempts. Thi...

In Chap. 4, the embedding and detection algorithms of the proposed audio watermarking scheme were analyzed theoretically. The aim of this chapter is to examine system performance in terms of imperceptibility, robustness, security, data payload, and computational complexity, as required in Sect. 1. 3. 1. .

In recent years, there has been considerable interest in the development of audio watermarking techniques. To clarify the essential principles underlying a diversity of sophisticated algorithms, this chapter gives an overview of basic methods for audio watermarking, such as least significant bit (LSB) modification, phase coding, spread spectrum wat...

Imperceptibility, robustness, and security are vital considerations in the design of any audio watermarking scheme for copyrights protection. In this chapter, a spread spectrum (SS)-based audio watermarking technique which involves the psychoacoustic model, multiple scrambling, adaptive synchronization, frequency alignment, and coded-image watermar...

Imperceptibility is a prerequisite to the use of the watermarked audio; hence, perceptual quality assessment is worthy of more attention. Objective quality measures have been widely used in speech quality evaluation. In this chapter, we introduce objective quality measures used for the first time in the perceptual quality evaluation of audio waterm...

Instead of using the entire minutiae template to generate a protected fingerprint template, recently a non-invertible cancellable fingerprint construct based on a 3-5 minutiae pattern was proposed as a safer alternative. This paper investigates the recognition accuracy attainable by this new fingerprint construct. It is found that using five sample...

A crucial property of an effective fingerprint template protection scheme is non-invertibility, which ensures that the original fingerprint template cannot be recovered from its secured counterpart. Since it is extremely difficult to design a function that achieves a high degree of non-invertibility, it is unsafe to use an entire fingerprint templa...

Psychoacoustics is the science of sound perception, i.e., investigating the statistical relationships between acoustic stimuli and hearing sensations [51]. This study aims to build up the psychoacoustic model, a kind of quantitative model, which could closely match the hearing mechanism. A good understanding of the sensory response of the human aud...

Available acoustical signal processing algorithms for active noise control (ANC) can only attenuate noise at a finite number of points. Accordingly, they are unable to create continuous quiet zones in 3-D space. This paper proposes a methodology for developing a family of acoustical signal processing algorithms that are able to control sound in thr...

Available adaptive active noise control (ANC) algorithms can only minimize the noise level at a point that an error microphone is placed. Consequently, a zone of quiet around this microphone is produced as a byproduct. However, they cannot technically control or, even, monitor the noise level within the zone of quiet unless they use several sensors...

Motivated by the inherent correlation between the speech features and their lexical words, we propose in this paper a new framework for learning the parameters of the corresponding acoustic and language models jointly. The proposed framework is based on discriminative training of the models' parameters using minimum classification error criterion....

This paper proposes a new, privacy-preserving fingerprint construct, which consists of a single pattern created from a small subset of minutiae from the corresponding minutiae template. The sparsity of the resulting feature vector ensures that it cannot be used to reconstruct the underlying fingerprint image, while simultaneously allowing for the c...

A significant challenge in the development of automated fingerprint recognition algorithms is dealing with missing minutiae. While it is generally assumed that some minutiae will always be missing between multiple samples of the same fingerprint, this assumption has never been empirically evaluated. An important factor influencing minutiae persiste...

This paper presents a novel method for creating a frame, to be used as an over complete dictionary for the progressive compression of 3D mesh geometry. The frame is computed from redundant linear combinations of the eigenvectors of a mesh Laplacian matrix, and atoms are selected by a Matching Pursuit algorithm. Experimental results show that a spar...

In this paper we study the robustness of a command decoding approach based on tiny decoding graphs for voice-based robotic interaction. This approach comprises the fusion of the grammar rules and the statistical n-gram language models to produce an elegant and quite efficient tiny decoding graph. The resulting tiny graph has several advantages such...

This paper proposes a discriminative learning algo-rithm for improving the accuracy of continuous speech recog-nition systems through optimizing the language model parame-ters on decoding graphs. The proposed algorithm employs soft margin estimation (SME) to build an objective function for maximizing the margin between the correct transcriptions an...

The development of automatic speech recognition (ASR) systems plays a crucial role in their performance as well as their integration with spoken dialogue systems for controlling service robots. However, to the best of our knowledge, there is no research in the literature addressing the development of ASR systems and their integration with service r...

In this paper, we propose a speech decoder, called SPeech lIght decoDER (SPIDER), for extracting the best decod-ing hypothesis from a search space constructed using weighted finite-state transducers. Despite existence of many speech de-coders, these decoders are quite complicated as they take into consideration many design goals, such as extraction...

In traditional models of speech recognition, acoustic and language models are treated in independence and usually estimated separately, which may yield a suboptimal recognition performance. In this paper, we propose a joint optimization framework for learning the parameters of acoustic and language models using minimum classification error criterio...

This paper develops a reliable methodology for active control of acoustic noise in free space. The core of this paper consists of a root locus analysis on the adaptation process performed in active noise control. Based on this analysis, a novel active noise control algorithm is derived. This algorithm is fully implemented in an experimental setup....

Adaptive active noise control (ANC) systems with single-channel structure create a silent point at the location of an error microphone. This silent point creates a zone of quiet surrounding itself, however, this zone has small dimensions. Moreover, the error microphone has to be located within the zone of quiet, limiting the effective space availab...

This study represents a stochastic model for the adaptation process performed on adaptive control systems by the filtered-x least-mean-square (FxLMS) algorithm. The main distinction of this model is that it is derived without using conventional simplifying assumptions regarding the physical plant to be controlled. This model is then used to derive...

This paper concerns adaptive active control of acoustic noise in free space. Traditional adaptive active noise control algorithms are efficient in acoustic ducts; however, they are very unstable and sensitive when being used in free space. An efficient adaptive active noise control algorithm for free space noise is derived based on a root locus ana...

This paper concerns about the theoretical stability of the adap­ tation process performed by the Filtered-x Least Mean Square (FxLMS) algorithm in active control of acoustic noise. A dynamic model for the Variance of Residual Acoustic Noise (VRAN) is developed and it is shown that the stability of this model is a sufficient condition for the stabil...

It is our believe that joint optimization of acoustic and language models meets the inherent correlation between them, and thus expected to achieve better recognition perfor-mance. This nice approach should be effective in achieving robust speech recognition where the testing conditions are different from those of training. The acoustic and languag...

The goal of automatic speech recognition (ASR) is to decode spoken utterances, by machine, to uncover their information content. Therefore, ASR is also called speech decoding. Over the last seven decades, Tremendous research efforts emerged competing together to achieve this goal in a real-time accurate way. One of the promising approaches to achie...

This research proposes a sub-graph extraction method to boost the discriminability of the generative acoustic models. The proposed method explicitly formalizes the notion that the words of reference decoding paths have not only one pronunciation but instead multiple pronunciations for the same word. The discriminative training framework is based on...

The document, which includes fourblank pages, was not presented at the conference and therefore was not made available for publication as part of the conference proceedings.

Recognising people from their gait is a challenging problem in biometric research. In this paper, we address the problem of gait identification based on a novel approach of sub-vector quantisation (SVQ) technique. A silhouette-based algorithm is utilised to capture the spatial-temporal information of the gait. A sequence of temporally ordered outer...

This paper studies theoretical analysis and practical implementation of adaptive active noise control (ANC) systems using Filtered-x Least Mean Square (FxLMS) algorithm. A comprehensive theoretical model for the adaptation process performed by the FxLMS algorithm is developed. Based on this model, closed-form math-ematical expressions for formulati...