Paul Hill | University of Bristol (original) (raw)

Papers by Paul Hill

Multimedia Tools and Applications, 2016

In this paper, we present a method for autonomously detecting and extracting region(s)-of-interes... more In this paper, we present a method for autonomously detecting and extracting region(s)-of-interest (ROI) from surveillance videos using trajectorybased analysis. Our approach, localizes ROI in a stochastic manner using correlated probability density functions that model motion dynamics of multiple moving targets. The motion dynamics model is built by analyzing trajectories of multiple moving targets and associating importance to regions in the scene. The importance of each region is estimated as a function of the total time spent by multiple targets, their instantaneous velocity and direction of movement whilst passing through that region. We systematically validate our model and benchmark our technique against competing baselines through extensive experimentation using public datasets such as CAVIAR, ViSOR, and CUHK as well as a scenario-specific in-house surveillance dataset. Results obtained have demonstrated the superiority of the proposed technique against a few popular existing state-of-the-art techniques.

International Journal of Remote Sensing, 2016

This paper presents a supervised hierarchical remote sensing image segmentation technique using a... more This paper presents a supervised hierarchical remote sensing image segmentation technique using a committee of multi-scale convolutional neural networks. With existing techniques, segmentation is achieved through fine-tuning a set of predefined feature detectors. However, such a solution is not robust since the introduction of new sensors or applications would require novel features and techniques to be developed. Conversely, the proposed method achieves segmentation through a set of learned feature detectors. In order to learn feature detectors, the proposed method exploits a committee of convolutional neural networks that perform multi-scale analysis on each band in order to derive individual probability maps on region boundaries. Probability maps are then inter-fused in order to produce a fused probability map. Further, the fused map is intra-fused using a morphological scheme into a hierarchical segmentation map. The proposed method is quantitatively compared to baseline techniques on a publiclyavailable dataset. Results, presented in this paper, highlight the improved accuracy of the proposed method.

IIEEE International Conference on Acoustics Speech and Signal Processing, 2002

ArXiv, 2021

Classification of images within the compressed domain offers significant benefits. These benefits... more Classification of images within the compressed domain offers significant benefits. These benefits include reduced memory and computational requirements of a classification system. This paper proposes two such methods as a proof of concept: The first classifies within the JPEG image transform domain (i.e. DCT transform data); the second classifies the JPEG compressed binary bitstream directly. These two methods are implemented using Residual Network CNNs and an adapted Vision Transformer. Top-1 accuracy of approximately 70% and 60% were achieved using these methods respectively when classifying the Caltech C101 database. Although these results are significantly behind the state of the art for classification for this database (9̃5%), it illustrates the first time direct bitstream image classification has been achieved. This work confirms that direct bitstream image classification is possible and could be utilised in a first pass database screening of a raw bitstream (within a wired or...

A significant number of researchers have recently applied deep learning methods to image fusion. ... more A significant number of researchers have recently applied deep learning methods to image fusion. However, most of these works either require a large amount of training data or depend on pre-trained models or frameworks. This inevitably encounters a shortage of training data or a mismatch between the framework and the actual problem. Recently, the publication of Deep Image Prior (DIP) method made it possible to do image restoration totally training-data-free. However, the original design of DIP is hard to be generalized to multi-image processing problems. This paper introduces a novel loss calculation structure, in the framework of DIP, while formulating image fusion as an inverse problem. This enables the extension of DIP to general multisensor/multifocus image fusion problems. Secondly, we propose a multi-channel approach to improve the effect of DIP. Finally, an evaluation is conducted using several commonly used image fusion assessment metrics. The results are compared with state...

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020

This article describes the application of machine learning techniques to develop state-of-the-art... more This article describes the application of machine learning techniques to develop state-of-the-art detection and prediction system for spatiotemporal events found within remote sensing data; specifically, harmful algal bloom (HAB) events. We propose HAB detection system based on a ground truth historical record of HAB events, a novel spatiotemporal datacube representation of each event (from MODIS and GEBCO bathymetry data), and a variety of machine learning architectures utilizing the state-of-the-art spatial and temporal analysis methods based on convolutional neural networks, long short-term memory components together with random forest, and support vector machine classification methods. This work has focused specifically on the case study of the detection of Karenia brevis algae (K. brevis) HAB events within the coastal waters of Florida (over 2850 events from 2003 to 2018; an order of magnitude larger than any previous machine learning detection study into HAB events). The development of multimodal spatiotemporal datacube data structures and associated novel machine learning methods give a unique architecture for the automatic detection of environmental events. Specifically, when applied to the detection of HAB events, it gives a maximum detection accuracy of 91% and a Kappa coefficient of 0.81 for the Florida data considered. A HAB forecast system was also developed where a temporal subset of each datacube was used to predict the presence of a HAB in the future. This system was not significantly less accurate than the detection system being able to predict with 86% accuracy up to 8 d in the future.

2014 IEEE 11th International Symposium on Biomedical Imaging (ISBI), 2014

ABSTRACT This paper describes a novel method for preprocessing of microscopy images by means of d... more ABSTRACT This paper describes a novel method for preprocessing of microscopy images by means of denoising and contrast enhancement in the wavelet domain. A non-linear enhancement function has been designed based on the local dispersion of the wavelet coefficients modelled as a bivariate Cauchy distribution. Within the same statistical framework, a simultaneous noise reduction in the image is performed by means of a shrinkage function, thus preventing noise amplification. The proposed method has been tested on a light microscopy image dataset and has been shown to greatly enhance the low-contrast noisy images while outperforming other state-of-art contrast enhancement methods.

IEEE Transactions on Circuits and Systems for Video Technology, 2006

Digital Signal Processing, 2013

This paper describes a new method for contrast enhancement in images of low-light or unevenly ill... more This paper describes a new method for contrast enhancement in images of low-light or unevenly illuminated scenes based on statistical modelling of wavelet coefficients of the image. A non-linear enhancement function has been designed based on the local dispersion of the wavelet coefficients modelled as a bivariate Cauchy distribution. Within the same statistical framework, a simultaneous noise reduction in the image is performed by means of a shrinkage function, thus preventing noise amplification. The proposed enhancement method has been shown to perform very well with insufficiently illuminated and noisy images, outperforming other conventional methods, in terms of contrast enhancement and noise reduction in the output image.

Signal Processing: Image Communication, 2015

Two undecimated forms of the Dual Tree Complex Wavelet Transform (DT-CWT) are introduced and thei... more Two undecimated forms of the Dual Tree Complex Wavelet Transform (DT-CWT) are introduced and their application to image denoising is described. These undecimated transforms extend the DT-CWT through the removal of downsampling of the filter outputs together with upsampling of the filters in a similar structure to the Undecimated Discrete Wavelet Transform (UDWT). Both the developed transforms offer exact translational invariance, improved scale-to-scale coefficient correlation together with the directional selectivity of the DT-CWT. Additionally, within each of these developed transforms, the subbands are of a consistent size. They therefore benefit from a direct one-to-one relationship between co-located coefficients at all scales. This is an important relationship that can be exploited within applications such as denoising, image fusion and segmentation. The enhanced properties of the transforms have been exploited within a bivariate shrinkage denoising application, demonstrating quantitative improvements in denoising results compared to the DT-CWT. The two novel transforms together with the DT-CWT offer a trade off between denoising performance, computational efficiency and memory requirements.

Proceedings 2001 International Conference on Image Processing, Feb 1, 2001

Simple wavelet and wavelet packet transforms have often been used for texture characterisation th... more Simple wavelet and wavelet packet transforms have often been used for texture characterisation through the analysis of spatial-frequenc y content. However, most previous methods make no use of any statistical analysis of the transforms' subbands. A novel method is now presented for modelling the multivariate distributions of subband coefficients by considering spatially related coefficients. The Bhattacharya and divergence metrics are then used to produce an improved texture classification method for the application to content based image retrieval.

Proceedings 2001 International Conference on Image Processing, Feb 1, 2001

Simple wavelet and wavelet packet transforms have often been used for texture characterisation th... more Simple wavelet and wavelet packet transforms have often been used for texture characterisation through the analysis of spatial-frequenc y content. However, most previous methods make no use of any statistical analysis of the transforms' subbands. A novel method is now presented for modelling the multivariate distributions of subband coefficients by considering spatially related coefficients. The Bhattacharya and divergence metrics are then used to produce an improved texture classification method for the application to content based image retrieval.

Extended Abstracts of the 2021 CHI Conference on Human Factors in Computing Systems, 2021

Live performances are immersive shared experiences, traditionally taking place in designated, car... more Live performances are immersive shared experiences, traditionally taking place in designated, carefully designed physical spaces such as theatres or concert halls. As it is becoming increasingly common for audiences to experience this type of content remotely using digital technology, it is crucial to reflect on the design of digital experiences and the technology used to deliver them. This research is guided by the question: How can the design of streaming technologies support artists in creating immersive and engaging audience experiences? A series of audience studies, which took place as cultural organisations were forced to adapt and deliver their content remotely due to the COVID19 global pandemic, highlighted problems with existing streaming solutions and informed a set of design recommendations for audience experience and research.

Time series of displacement are now routinely available from satellite InSAR and are used for fla... more Time series of displacement are now routinely available from satellite InSAR and are used for flagging anomalous ground motion, but not yet for forecasting. Here we test the capabilities of conventional time series analysis and forecasting methods such as SARIMA and supervised machine learning approaches such as Long Short Term Memory (LSTM) in comparison to simple function extrapolation methods. For our initial tests, we focus on forecasting periodic signals and begin by characterising the time-series using sinusoid fitting, seasonal decomposition and autocorrelation functions. We find that the three measures are broadly comparable but identify different types of seasonal characteristic. We use this to select a set of 310 points with highly seasonal characteristics and test the three chosen forecasting methods over prediction windows of 1-9 months. The lowest overall RMSE values are obtained for SARIMA when considering short term predictions ($<$1 month), whereas sinusoid extrap...

Signal Processing: Image Communication, 2010

Modern video codecs such as MPEG2, MPEG4-ASP and H.264 depend on sub-pixel motion estimation to o... more Modern video codecs such as MPEG2, MPEG4-ASP and H.264 depend on sub-pixel motion estimation to optimise rate-distortion efficiency. Sub-pixel motion estimation is implemented within these standards using interpolated values at 1/2 or 1/4 pixel accuracy. By using these interpolated values, the residual energy for each predicted macroblock is reduced. However this leads to a significant increase in complexity at the encoder, especially for H.264, where the cost of an exhaustive set of macroblock segmentations needs to be estimated for optimal mode selection. This paper presents a novel scheme for sub-pixel motion estimation based on the whole-pixel SAD distribution. Both half-pixel and quarter-pixel searches are guided by a model-free estimation of the SAD surface using a two dimensional kernel method. While giving an equivalent rate distortion performance, this approach approximately halves the number of quarter-pixel search positions giving an overall speed up of approximately 10% compared to the EPZS quarter-pixel method (the state of the art H.264 optimised subpixel motion estimator).

2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2016

A novel adaptation of the two dimensional Homomorphic filter is introduced using the Dual Tree Co... more A novel adaptation of the two dimensional Homomorphic filter is introduced using the Dual Tree Complex Wavelet Transform (DT-CWT) for improved illumination invariant processing. The Homomorphic filter is conventionally implemented within the log-Fourier domain using an isotropic high-pass filter based on the assumption that the illumination signal occupies low spatial frequencies. In this case however, low frequency structural reflectance content will be incorrectly attenuated. Our method implements the Homomorphic filter using the DT-CWT and exploits the property of cross scale persistence (for structural content) to generate a filter that retains cross scale content and therefore reduces incorrect attenuation of structural reflectance content.

Proceedings of the 16th Ieee International Conference on Image Processing, Nov 7, 2009

Modern video codecs such as MPEG2, MPEG4-ASP and H.264 depend on computationally complex sub-pixe... more Modern video codecs such as MPEG2, MPEG4-ASP and H.264 depend on computationally complex sub-pixel Motion Estimation (ME) to optimise rate-distortion efficiency. Sub-pixel ME is implemented within these standards using interpolated values at ½ or ¼ pixel accuracy. This paper presents a novel scheme for sub-pixel ME using a model free kernel method utilising the result of the whole-pixel SAD distribution applied to an H.264 encoder. While giving an equivalent rate distortion performance, this approach approximately halves the number of quarter-pixel search positions giving an overall speed up of approximately 10% compared to the EPZS quarter-pixel method (the state of the art H.264 optimised sub-pixel motion estimator).

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society, Jan 11, 2016

Accurate estimation of the contrast sensitivity of the human visual system is crucial for percept... more Accurate estimation of the contrast sensitivity of the human visual system is crucial for perceptually based image processing in applications such as compression, fusion and denoising. Conventional Contrast Sensitivity Functions (CSFs) have been obtained using fixed sized Gabor functions. However, the basis functions of multiresolution decompositions such as wavelets often resemble Gabor functions but are of variable size and shape. Therefore to use conventional contrast sensitivity functions in such cases is not appropriate. We have therefore conducted a set of psychophysical tests in order to obtain the contrast sensitivity function for a range of multiresolution transforms: the Discrete Wavelet Transform (DWT), the Steerable Pyramid, the Dual-Tree Complex Wavelet Transform (DT-CWT) and the Curvelet Transform. These measures were obtained using contrast variation of each transforms' basis functions in a 2AFC experiment combined with an adapted version of the QUEST psychometric...

Sensor Signal Processing for Defence (SSPD 2011), 2011

Abstract This paper introduces a novel system that is able to fuse two or more sets of multimodal... more Abstract This paper introduces a novel system that is able to fuse two or more sets of multimodal videos in the compressed domain. This is achieved without drift and produces an embedded bitstream that offers fine grain scalability. Previous attempts to fuse in the compressed video domain have been not been possible due to the complications of predictive loops within standard video encoding techniques.

Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205)

Content-based retrieval is ultimately dependent on the features used for the annotation of data a... more Content-based retrieval is ultimately dependent on the features used for the annotation of data and its efficiency is dependent on the invariance and robust properties of these features. For texture based features an important form of invariance is rotational invariance. In this paper novel rotationally invariant texture based features are introduced that are extracted from a Polar Fourier Transform (PFT). The PFT is similar to the Discrete Fourier Transform in two dimensions but uses transform parameters radius and angle rather than the Cartesian coordinates. The PFT is discretised appropriately across the angular and radial frequency space with the transform magnitudes forming the rotationally invariant features. These features although rotationally invariant, capture the angular distribution together with the radial distribution of frequency within texture. Preliminary results show the method to give better results than rotationally variant and invariant Gabor filter schemes.

Multimedia Tools and Applications, 2016

In this paper, we present a method for autonomously detecting and extracting region(s)-of-interes... more In this paper, we present a method for autonomously detecting and extracting region(s)-of-interest (ROI) from surveillance videos using trajectorybased analysis. Our approach, localizes ROI in a stochastic manner using correlated probability density functions that model motion dynamics of multiple moving targets. The motion dynamics model is built by analyzing trajectories of multiple moving targets and associating importance to regions in the scene. The importance of each region is estimated as a function of the total time spent by multiple targets, their instantaneous velocity and direction of movement whilst passing through that region. We systematically validate our model and benchmark our technique against competing baselines through extensive experimentation using public datasets such as CAVIAR, ViSOR, and CUHK as well as a scenario-specific in-house surveillance dataset. Results obtained have demonstrated the superiority of the proposed technique against a few popular existing state-of-the-art techniques.

International Journal of Remote Sensing, 2016

This paper presents a supervised hierarchical remote sensing image segmentation technique using a... more This paper presents a supervised hierarchical remote sensing image segmentation technique using a committee of multi-scale convolutional neural networks. With existing techniques, segmentation is achieved through fine-tuning a set of predefined feature detectors. However, such a solution is not robust since the introduction of new sensors or applications would require novel features and techniques to be developed. Conversely, the proposed method achieves segmentation through a set of learned feature detectors. In order to learn feature detectors, the proposed method exploits a committee of convolutional neural networks that perform multi-scale analysis on each band in order to derive individual probability maps on region boundaries. Probability maps are then inter-fused in order to produce a fused probability map. Further, the fused map is intra-fused using a morphological scheme into a hierarchical segmentation map. The proposed method is quantitatively compared to baseline techniques on a publiclyavailable dataset. Results, presented in this paper, highlight the improved accuracy of the proposed method.

IIEEE International Conference on Acoustics Speech and Signal Processing, 2002

ArXiv, 2021

Classification of images within the compressed domain offers significant benefits. These benefits... more Classification of images within the compressed domain offers significant benefits. These benefits include reduced memory and computational requirements of a classification system. This paper proposes two such methods as a proof of concept: The first classifies within the JPEG image transform domain (i.e. DCT transform data); the second classifies the JPEG compressed binary bitstream directly. These two methods are implemented using Residual Network CNNs and an adapted Vision Transformer. Top-1 accuracy of approximately 70% and 60% were achieved using these methods respectively when classifying the Caltech C101 database. Although these results are significantly behind the state of the art for classification for this database (9̃5%), it illustrates the first time direct bitstream image classification has been achieved. This work confirms that direct bitstream image classification is possible and could be utilised in a first pass database screening of a raw bitstream (within a wired or...

A significant number of researchers have recently applied deep learning methods to image fusion. ... more A significant number of researchers have recently applied deep learning methods to image fusion. However, most of these works either require a large amount of training data or depend on pre-trained models or frameworks. This inevitably encounters a shortage of training data or a mismatch between the framework and the actual problem. Recently, the publication of Deep Image Prior (DIP) method made it possible to do image restoration totally training-data-free. However, the original design of DIP is hard to be generalized to multi-image processing problems. This paper introduces a novel loss calculation structure, in the framework of DIP, while formulating image fusion as an inverse problem. This enables the extension of DIP to general multisensor/multifocus image fusion problems. Secondly, we propose a multi-channel approach to improve the effect of DIP. Finally, an evaluation is conducted using several commonly used image fusion assessment metrics. The results are compared with state...

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020

This article describes the application of machine learning techniques to develop state-of-the-art... more This article describes the application of machine learning techniques to develop state-of-the-art detection and prediction system for spatiotemporal events found within remote sensing data; specifically, harmful algal bloom (HAB) events. We propose HAB detection system based on a ground truth historical record of HAB events, a novel spatiotemporal datacube representation of each event (from MODIS and GEBCO bathymetry data), and a variety of machine learning architectures utilizing the state-of-the-art spatial and temporal analysis methods based on convolutional neural networks, long short-term memory components together with random forest, and support vector machine classification methods. This work has focused specifically on the case study of the detection of Karenia brevis algae (K. brevis) HAB events within the coastal waters of Florida (over 2850 events from 2003 to 2018; an order of magnitude larger than any previous machine learning detection study into HAB events). The development of multimodal spatiotemporal datacube data structures and associated novel machine learning methods give a unique architecture for the automatic detection of environmental events. Specifically, when applied to the detection of HAB events, it gives a maximum detection accuracy of 91% and a Kappa coefficient of 0.81 for the Florida data considered. A HAB forecast system was also developed where a temporal subset of each datacube was used to predict the presence of a HAB in the future. This system was not significantly less accurate than the detection system being able to predict with 86% accuracy up to 8 d in the future.

2014 IEEE 11th International Symposium on Biomedical Imaging (ISBI), 2014

ABSTRACT This paper describes a novel method for preprocessing of microscopy images by means of d... more ABSTRACT This paper describes a novel method for preprocessing of microscopy images by means of denoising and contrast enhancement in the wavelet domain. A non-linear enhancement function has been designed based on the local dispersion of the wavelet coefficients modelled as a bivariate Cauchy distribution. Within the same statistical framework, a simultaneous noise reduction in the image is performed by means of a shrinkage function, thus preventing noise amplification. The proposed method has been tested on a light microscopy image dataset and has been shown to greatly enhance the low-contrast noisy images while outperforming other state-of-art contrast enhancement methods.

IEEE Transactions on Circuits and Systems for Video Technology, 2006

Digital Signal Processing, 2013

This paper describes a new method for contrast enhancement in images of low-light or unevenly ill... more This paper describes a new method for contrast enhancement in images of low-light or unevenly illuminated scenes based on statistical modelling of wavelet coefficients of the image. A non-linear enhancement function has been designed based on the local dispersion of the wavelet coefficients modelled as a bivariate Cauchy distribution. Within the same statistical framework, a simultaneous noise reduction in the image is performed by means of a shrinkage function, thus preventing noise amplification. The proposed enhancement method has been shown to perform very well with insufficiently illuminated and noisy images, outperforming other conventional methods, in terms of contrast enhancement and noise reduction in the output image.

Signal Processing: Image Communication, 2015

Two undecimated forms of the Dual Tree Complex Wavelet Transform (DT-CWT) are introduced and thei... more Two undecimated forms of the Dual Tree Complex Wavelet Transform (DT-CWT) are introduced and their application to image denoising is described. These undecimated transforms extend the DT-CWT through the removal of downsampling of the filter outputs together with upsampling of the filters in a similar structure to the Undecimated Discrete Wavelet Transform (UDWT). Both the developed transforms offer exact translational invariance, improved scale-to-scale coefficient correlation together with the directional selectivity of the DT-CWT. Additionally, within each of these developed transforms, the subbands are of a consistent size. They therefore benefit from a direct one-to-one relationship between co-located coefficients at all scales. This is an important relationship that can be exploited within applications such as denoising, image fusion and segmentation. The enhanced properties of the transforms have been exploited within a bivariate shrinkage denoising application, demonstrating quantitative improvements in denoising results compared to the DT-CWT. The two novel transforms together with the DT-CWT offer a trade off between denoising performance, computational efficiency and memory requirements.

Proceedings 2001 International Conference on Image Processing, Feb 1, 2001

Simple wavelet and wavelet packet transforms have often been used for texture characterisation th... more Simple wavelet and wavelet packet transforms have often been used for texture characterisation through the analysis of spatial-frequenc y content. However, most previous methods make no use of any statistical analysis of the transforms' subbands. A novel method is now presented for modelling the multivariate distributions of subband coefficients by considering spatially related coefficients. The Bhattacharya and divergence metrics are then used to produce an improved texture classification method for the application to content based image retrieval.

Proceedings 2001 International Conference on Image Processing, Feb 1, 2001

Simple wavelet and wavelet packet transforms have often been used for texture characterisation th... more Simple wavelet and wavelet packet transforms have often been used for texture characterisation through the analysis of spatial-frequenc y content. However, most previous methods make no use of any statistical analysis of the transforms' subbands. A novel method is now presented for modelling the multivariate distributions of subband coefficients by considering spatially related coefficients. The Bhattacharya and divergence metrics are then used to produce an improved texture classification method for the application to content based image retrieval.

Extended Abstracts of the 2021 CHI Conference on Human Factors in Computing Systems, 2021

Live performances are immersive shared experiences, traditionally taking place in designated, car... more Live performances are immersive shared experiences, traditionally taking place in designated, carefully designed physical spaces such as theatres or concert halls. As it is becoming increasingly common for audiences to experience this type of content remotely using digital technology, it is crucial to reflect on the design of digital experiences and the technology used to deliver them. This research is guided by the question: How can the design of streaming technologies support artists in creating immersive and engaging audience experiences? A series of audience studies, which took place as cultural organisations were forced to adapt and deliver their content remotely due to the COVID19 global pandemic, highlighted problems with existing streaming solutions and informed a set of design recommendations for audience experience and research.

Time series of displacement are now routinely available from satellite InSAR and are used for fla... more Time series of displacement are now routinely available from satellite InSAR and are used for flagging anomalous ground motion, but not yet for forecasting. Here we test the capabilities of conventional time series analysis and forecasting methods such as SARIMA and supervised machine learning approaches such as Long Short Term Memory (LSTM) in comparison to simple function extrapolation methods. For our initial tests, we focus on forecasting periodic signals and begin by characterising the time-series using sinusoid fitting, seasonal decomposition and autocorrelation functions. We find that the three measures are broadly comparable but identify different types of seasonal characteristic. We use this to select a set of 310 points with highly seasonal characteristics and test the three chosen forecasting methods over prediction windows of 1-9 months. The lowest overall RMSE values are obtained for SARIMA when considering short term predictions ($<$1 month), whereas sinusoid extrap...

Signal Processing: Image Communication, 2010

Modern video codecs such as MPEG2, MPEG4-ASP and H.264 depend on sub-pixel motion estimation to o... more Modern video codecs such as MPEG2, MPEG4-ASP and H.264 depend on sub-pixel motion estimation to optimise rate-distortion efficiency. Sub-pixel motion estimation is implemented within these standards using interpolated values at 1/2 or 1/4 pixel accuracy. By using these interpolated values, the residual energy for each predicted macroblock is reduced. However this leads to a significant increase in complexity at the encoder, especially for H.264, where the cost of an exhaustive set of macroblock segmentations needs to be estimated for optimal mode selection. This paper presents a novel scheme for sub-pixel motion estimation based on the whole-pixel SAD distribution. Both half-pixel and quarter-pixel searches are guided by a model-free estimation of the SAD surface using a two dimensional kernel method. While giving an equivalent rate distortion performance, this approach approximately halves the number of quarter-pixel search positions giving an overall speed up of approximately 10% compared to the EPZS quarter-pixel method (the state of the art H.264 optimised subpixel motion estimator).

2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2016

A novel adaptation of the two dimensional Homomorphic filter is introduced using the Dual Tree Co... more A novel adaptation of the two dimensional Homomorphic filter is introduced using the Dual Tree Complex Wavelet Transform (DT-CWT) for improved illumination invariant processing. The Homomorphic filter is conventionally implemented within the log-Fourier domain using an isotropic high-pass filter based on the assumption that the illumination signal occupies low spatial frequencies. In this case however, low frequency structural reflectance content will be incorrectly attenuated. Our method implements the Homomorphic filter using the DT-CWT and exploits the property of cross scale persistence (for structural content) to generate a filter that retains cross scale content and therefore reduces incorrect attenuation of structural reflectance content.

Proceedings of the 16th Ieee International Conference on Image Processing, Nov 7, 2009

Modern video codecs such as MPEG2, MPEG4-ASP and H.264 depend on computationally complex sub-pixe... more Modern video codecs such as MPEG2, MPEG4-ASP and H.264 depend on computationally complex sub-pixel Motion Estimation (ME) to optimise rate-distortion efficiency. Sub-pixel ME is implemented within these standards using interpolated values at ½ or ¼ pixel accuracy. This paper presents a novel scheme for sub-pixel ME using a model free kernel method utilising the result of the whole-pixel SAD distribution applied to an H.264 encoder. While giving an equivalent rate distortion performance, this approach approximately halves the number of quarter-pixel search positions giving an overall speed up of approximately 10% compared to the EPZS quarter-pixel method (the state of the art H.264 optimised sub-pixel motion estimator).

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society, Jan 11, 2016

Accurate estimation of the contrast sensitivity of the human visual system is crucial for percept... more Accurate estimation of the contrast sensitivity of the human visual system is crucial for perceptually based image processing in applications such as compression, fusion and denoising. Conventional Contrast Sensitivity Functions (CSFs) have been obtained using fixed sized Gabor functions. However, the basis functions of multiresolution decompositions such as wavelets often resemble Gabor functions but are of variable size and shape. Therefore to use conventional contrast sensitivity functions in such cases is not appropriate. We have therefore conducted a set of psychophysical tests in order to obtain the contrast sensitivity function for a range of multiresolution transforms: the Discrete Wavelet Transform (DWT), the Steerable Pyramid, the Dual-Tree Complex Wavelet Transform (DT-CWT) and the Curvelet Transform. These measures were obtained using contrast variation of each transforms' basis functions in a 2AFC experiment combined with an adapted version of the QUEST psychometric...

Sensor Signal Processing for Defence (SSPD 2011), 2011

Abstract This paper introduces a novel system that is able to fuse two or more sets of multimodal... more Abstract This paper introduces a novel system that is able to fuse two or more sets of multimodal videos in the compressed domain. This is achieved without drift and produces an embedded bitstream that offers fine grain scalability. Previous attempts to fuse in the compressed video domain have been not been possible due to the complications of predictive loops within standard video encoding techniques.

Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205)

Content-based retrieval is ultimately dependent on the features used for the annotation of data a... more Content-based retrieval is ultimately dependent on the features used for the annotation of data and its efficiency is dependent on the invariance and robust properties of these features. For texture based features an important form of invariance is rotational invariance. In this paper novel rotationally invariant texture based features are introduced that are extracted from a Polar Fourier Transform (PFT). The PFT is similar to the Discrete Fourier Transform in two dimensions but uses transform parameters radius and angle rather than the Cartesian coordinates. The PFT is discretised appropriately across the angular and radial frequency space with the transform magnitudes forming the rotationally invariant features. These features although rotationally invariant, capture the angular distribution together with the radial distribution of frequency within texture. Preliminary results show the method to give better results than rotationally variant and invariant Gabor filter schemes.