As the physical size of single pixels in digital cameras grows smaller, the captured images are increasingly affected by defocused blurring and consequently valuable details are lost. Different aperture patterns have already been proposed to mitigate this problem based on presumed conditions, which maybe violated in practise. Sensor characteristics and current photometric scene properties have been largely ignored in the design of aperture patterns in the literature. In this study, a number of perceptually optimised coded apertures are introduced for defocused deblurring. These apertures are specifically designed considering illumination conditions, sensor specifications and human visual system characteristics. The designed patterns are compared with circular apertures of equal throughput and pinhole aperture. Experiments show signal-to-noise ratio (SNR) gains of up to 0.35 and 2 dB over circular and pinhole apertures, respectively. To study the trade-off between diffraction and deblurring gains, the proposed binary masks are enhanced by smoothing and morphological operations, which can yield non-binary and rounded binary patterns. The results of the authors’ study show that rounded binary patterns improve diffraction behaviour while maintaining the desired SNR level.

Inspired by the facts that visual saliency captures more attention and spectral residual (SR) can indicate the saliency of the image, a novel reduced-reference image quality assessment metric is proposed based on the regularity of the SR. The orientation and frequency components of an image are first extracted in wavelet domain. Then SR is obtained to represent the saliency of the component. Next fractal dimension is adopted to encode SR and concatenated as the image features. Finally, the feature differences between reference image and distorted one are pooled as the quality score. The proposed metric is evaluated on four largest image databases (TID2013, TID2008, CSIQ, and LIVE databases), and experimental results confirm that the proposed metric has a good performance.

Uterine fibroids segmentation in ultrasound images is of great importance in the definition of intra-operative planning of ultrasound-guided high-intensity focused ultrasound (HIFU) therapy. However, it is challenging to obtain accurate, robust and efficient uterine fibroid segmentation due to low quality of ultrasound images. In this study, the authors propose a novel adaptive localised region and edge-based active contour model using shape constraint and sub-global information to accurately and efficiently segment the uterine fibroids in ultrasound images with robustness against initial contour. The authors first define adaptive local radius for the localised region-based model and combine it with the edge-based model to accurately and efficiently capture image's heterogeneous features and edge features. Then, they incorporate a shape constraint to reduce boundary leakage or excessive contraction to obtain more accurate segmentation. To overcome the initialisation sensitivity, they introduce the sub-global information to prevent the curve from trapping into the local minima and obtain robust results. Furthermore, the authors optimise computation by adaptively sharing local region and employing the multi-scale segmentation method to achieve efficient segmentation. The proposed method is validated by uterine fibroid ultrasound images in HIFU therapy and the results demonstrate that it can achieve accurate, robust and efficient segmentation.

In this study, a new approach for pansharpening of multispectral and panchromatic (PAN) images is proposed. The proposed technique is based on recently developed signal decomposition technique known as Hilbert vibration decomposition (HVD). In the proposed method, the histogram equalised PAN image is decomposed into many instantaneous amplitude (IA) and frequency components in the decreasing order of energy using the HVD. The IA of the first component (having highest energy) in the decomposition of the PAN image is used to generate the pansharpened image using appropriate pansharpening model. The tuning factor associated with the pansharpening model is optimised by single-objective particle swarm optimization algorithm. This method is also extended for the hyperspectral images. Experimental results of the proposed technique are compared with existing pansharpening methods in terms of both visual perception and objective metrics. It is observed that the proposed pansharpening scheme has improved spectral and spatial qualities as compared with the existing schemes. The effects of aliasing and misregistration errors in the proposed method are also investigated and it is observed that the proposed method is robust against aliasing and misregistration errors as compared with other existing methods.

An improved dehazing algorithm based on dark channel theory is proposed, in order to solve the problems of colour distortion and halo effect which still exists in dark channel prior algorithm. The dark channel prior theory may lead to colour distortion in sky region. Firstly, the guided filter is used to refine the segmentation of the sky region, and the atmospheric light is estimated accurately. Then, the median filter is used to obtain the detailed edge information. So a more clear transmission can be gotten which effectively suppress the halo problem. Finally, the gamma correction is applied to enhance image lightness with an empirically selected gamma parameter. The experimental results show that the proposed algorithm can effectively remove the haze. It can correct the colour distortion of the sky area and eliminate the halo effect at the edge of the scene.

Feature selection and fusion is of crucial importance in multi-feature visual tracking. This study proposes a multi-task kernel-based sparse learning method for multi-feature visual tracking. The proposed sparse learning method can discriminate the reliable and unreliable features for optimal multi-feature fusion through using a Fisher discrimination criterion-based multi-objective model to adaptively train the kernel weights of different features such as pixel intensity, edge and texture. To guarantee a robustness of the sparse representation method, a mixed norm is employed in the sparse leaning method to adaptively select correlated particle observations for multi-task sparse reconstruction. Experimental results show that the proposed sparse learning method can achieve a better tracking performance than state-of-the-art tracking methods do.

Face recognition has become a popular topic due to its applications in security, surveillance and so on. Current local methods such as the local binary pattern (LBP) or local derivative pattern (LDP) perform better than holistic methods since they are more stable on local changes such as misalignment, expression or occlusion, but their high computational complexity limit their applications. While LBP is a good feature method, the scale invariant feature transform (SIFT) is widely accepted as one of the best features to capture edge or local shape information. However, SIFT-based schemes are sensitive to illumination variation. Thus, the authors propose an LBP edge-mapped descriptor that uses maxima of gradient magnitude points. It accurately illustrates facial contours and has low computational complexity. Under variable lighting, experimental results show that the authors' method has a 16.5% higher recognition rate and requires 9.06 times less execution time than SIFT under FERET fc. Besides, when applied to the Extended Yale Face Database B, the authors' method outperformed SIFT-based approaches as well as saving about 70.9% in execution time. In uncontrolled conditions, their method has a 0.82% higher recognition rate than LDP histogram sequences in the Unconstrained Facial Images database.

As a particular case of the finite mixture model, Rayleigh mixture model (RMM) is considered as a useful tool for medical ultrasound (US) image segmentation. However, conventional RMM relies on intensity distribution only and does not take any spatial information into account that leads to misclassification on boundaries and inhomogeneous regions. The authors proposed an improved RMM with neighbour (RMMN) information to solve this problem by introducing neighbourhood information through a mean template. The incorporation of the spatial information made RMMN more robust to noise on the boundaries. The size of the window which incorporates neighbour information was resized adaptively according to the local gradient distribution. They evaluated their model on experiments on synthetic data and real US images used by high-intensity focused ultrasound therapy. On this data, they demonstrated that the proposed model outperforms several state-of-the-art methods in terms of both segmentation accuracy and computation time.

Block matching 3D denoising (BM3D) is an excellent single-image denoising method. However, it still needs to be improved for solving practical problems. In this study, the authors attempt to improve the method of BM3D. First, one of the problems of BM3D is that some of its references cannot perform self-adaption when the noise intensity of the images is changed. Therefore, they propose a method using total variation (TV) to calculate the image noise intensity and make the references perform self-adaption. Second, finding similar blocks in the BM3D method is a time-consuming procedure. To solve this problem, they analyse the relationship between the numbers of similar blocks and denoising effect, improve the process of searching for similar blocks, and reduce the running time. Third, through the experiment they find that the denoising effect of BM3D method in the domain of complex texture is unsatisfactory. Thus, they proposed a hybrid denoising method for the complex texture area, using the new TV model and BM3D method together to restore the image. Their experimental results show that the improved BM3D method performs better than the original BM3D method.

A new texture classification method based on wavelet transform is presented. The elements of the signature vector, FDBC, of an image are the fractal dimensions and barycentric coordinates of the bit planes of the wavelet coefficients in both the three-level high-frequency domains and the third low-frequency domain. The pretreatment is done with SVD decomposition and reconstruction by dropping half singular values. The one-nearest-neighbour classifier (1NN) with distance is used to make the classification. Furthermore, to improve classification result, the classifier 1NN is strengthened with weighted distance. The proposed method is tested on five subsets from Brodatz database and UMD database and is experimentally proved more efficient and more promising.

The vision-based lane detection is an important component of advanced driver assistance systems and it is essential for lane departure warning, lane keeping, and vehicle localisation. However, it is a challenging problem to improve the robustness of multi-lane detection due to factors, such as perspective effect, possible low visibility of lanes, and partial occlusions. To deal with these issues, the authors propose an improved lane hypothesis generation method using a reliable binary blob analysis. Most existing top-view based methods focused on the lane model fitting, but they neglected the reliability of hypothesis generation and the effectiveness in challenging conditions. To cope with these shortcomings, the authors carried out vanishing point detection and inverse perspective mapping to remove the perspective effect from the road images. Then two-stage binary blob filtering and blob verification techniques using classification are introduced to improve the robustness of lane hypothesis generation for lane detection. The experimental results show that the average detection accuracy on a new challenging multi-lane dataset is 97.7%. The performance of the proposed method outperforms that of the state-of-the-art method by 1.6% in detection accuracy on the Caltech lane benchmark dataset.

Image segmentation has many applications which range from machine learning to medical diagnosis. In this study, the authors propose a framework for the segmentation of images based on super-pixels and algorithms for community identification in graphs. The super-pixel pre-segmentation step reduces the number of nodes in the graph, rendering the method the ability to process large images. Moreover, community detection algorithms provide more accurate segmentation than traditional approaches based on spectral graph partition. The authors also compared their method with two algorithms: (i) the graph-based approach by Felzenszwalb and Huttenlocher and (ii) the contour-based method by Arbelaez. Results have shown that their method provides more precise segmentation and is faster than both of them.

Image self-similarity property is important to super-resolution reconstruction. However, how to effectively exploit the self-similarity information to reconstruct an underlying high-resolution image is still a challenging problem. The authors propose a novel model for solving the single image upsampling problem with the self-similarity property. First, the authors construct a statistical prior that requires maximising the similarity between the low- and high-resolution image pairs. Then, the authors develop an alternative Gaussian approximation solver based on the Gaussian mixture model to find the optimal high-resolution output. To obtain a better performance, the authors summarise some refined implementation skills to raise the reconstruction quality. For demonstration, a series of objective and subjective measurements are used to evaluate the performance of the model.

This study presents a model to effectively recognise image noise of different types and levels: impulse, Gaussian, Speckle and Poisson noise, and a mixture of multiple types of the noise. To classify image noise type, the convolutional neural network (CNN) method with backpropagation algorithm and stochastic gradient descent optimisation techniques are implemented. In order to reduce the training time and computational cost of the algorithm, the principal components analysis (PCA) filters generating strategy is deployed to obtain data adaptive filter banks. The authors validated their designed CNN with PCA for noise types recognition model with degraded images containing noise of single and combination of multiple types, with a total of 11,000 and 1650 datasets for training and testing purposes, respectively. The variety and complexity of data have never been addressed before in any other research work. The capability of their intelligent system in handling images degraded under this complicated environment has surpassed human-eye performance in noise types recognition. The authors’ experiments have proven the reliability of the proposed noise types recognition model by having achieved an overall average accuracy of 99.3% while recognising eight classes of noise.

In this study, a new framework for multimodal image registration is proposed based on the expectation–maximisation (EM) methodology. This framework allows to address simultaneously parametric and elastic registrations independently on the modality of the target and source images without making any assumptions about their intensity relationship. The EM formulation for the image registration problem leads to a regularised quadratic optimisation scheme to compute the displacement vector field (DVF) that aligns the images and depends on their joint intensity distribution. At the first stage, a parametric transformation is assumed for the DVF, where the resulting quadratic optimisation is computed recursively to calculate its optimal parameters. Next, a general unknown deformation models the elastic part of the DVF, which is represented by an additive structure. The resulting optimisation process by the EM formulation results in a cost function that involves data and regularisation terms, which is also solved recursively. A comprehensive evaluation of the parametric and elastic proposals is carried out by comparing to state-of-the-art algorithms and images from different application fields, where an advantage is visualised by the authors’ proposal in terms of a compromise between accuracy and robustness.

Image super-resolution (SR) reconstruction, which gains high-pixel and multi-detail image from single or several low-pixel images, has attracted increasing interest in recent years. This study proposes a new SR method based on sparse representation, which made good use of the non-local (NL) structure similarity and edge sharpness dictionary. Firstly, all the training patches are classified into different clusters according to diverse edge sharpness of patches. Secondly, different dictionaries are trained for different training patches in each cluster. Thirdly, the NL structure similarity is added into the constraint of NL structure similarity model, and the suitable dictionary is selected for current patch to achieve the coefficients according to the value of edge sharpness of patch. Finally, the high-resolution (HR) image is obtained by integrating HR patches obtained by the product of HR dictionaries and coefficients. Moreover, by calculating edge sharpness, the different dictionaries which adapt to patches with different structure are obtained, and the NL similarity is well utilised and more details are added to HR patch. Compared to some classical and common methods, the proposed method possesses better reconstruction effects in numerical and visual aspects.

This study is concerned with the surface mount component positioning problem and an improved chamfer matching method based on iterative position and rotation angle estimation approach is proposed. Instead of performing matching of the image with different pre-specified angle templates pixel by pixel, as does in traditional chamfer matching methods, the gradient information of the distance transform image is incorporated into the chamfer matching method and reduced computational cost of the method is achieved. The iterative formulas to update the translation and rotation angle are derived by reference to the characteristic of rigid body motion. The criterions of search region establishment, seed point selection and terminal conditions are given. The effectiveness of the proposed method is verified by applying the method to component positioning with actual captured images.

Mobile display technology has been developed fast and low-power design for the display is more important than ever. Hardware techniques for low-power design are not flexible and the existing software techniques are not considering the characteristics of the display panels. In this study, a histogram-based contrast-aware power control method for active-matrix organic light-emitting diode display is proposed. The proposed power control method solves an optimisation problem to achieve power control with maximum contrast available. A new histogram representation is also proposed to calculate the power consumption of an image efficiently in the optimisation process. Simulation results and hardware experiment results show that the proposed method provides both precise power control and better contrast of the output image.

A high-fidelity colour transfer should align the colour distributions between images and meanwhile avoid the damage to the original structure. However, the traditional methods often fail to yield high-fidelity transfer results due to some existing tone and structure artefacts. In this study, the authors propose a new framework to effectively integrate the tone and structure refinements of colour transfer. They develop the ideas of image decomposition and gradient guidance to perform tone reconstruction while protecting original structure. Its overall flow includes the five key steps: tone clustering, structure extraction, structure optimisation, gradient-guided tone reconstruction, and structure restoration. Moreover, they propose an evaluation metric to measure the differences of tone and structure between images. They demonstrate the performance of the proposed method through a number of experiments in visual comparison and objective evaluation.

Most of the recent leading multiple magnetic resonance imaging (MRI) super-resolution techniques for brain are limited to rigid motion. In this study, the authors aim to develop a super-resolution technique with diffeomorphism mainly for longitudinal brain MRI data. For the images from different time slots, unpredicted deformation may occur. In previous studies, sole rigid registration or traditional non-rigid registration has been frequently used to achieve multi-plane super-resolution. However, non-rigid motion of two brains from different time slots is difficult to model, since brain contains a wealth of complex structure such as the cerebral cortex. In order to address such problem, rigid and large diffeomorphic registration has been embedded into their super-resolution framework. In addition, many previous researchers use norm to achieve super-resolution framework. In this work, norm minimisation and regularisation based on a bilateral prior are adopted. These operations ensure its robustness to the assumed model of data and noise. Their approach is evaluated using Alzheimer datasets from seven different resolutions. Results show that their reconstructions have advantages over rigid and conventional non-rigid registration-based super-resolution, in terms of the root-mean-square error and structure similarity. Furthermore, their reconstruction results improve the precision of brain automatic segmentation.

Triplet Markov fields (TMF) model is widely used to deal with non-stationary synthetic aperture radar (SAR) images. However, its ability to capture global information remains limited due to the non-causal property. A hierarchical TMF model is proposed in this study based on the non-linear diffusion (ND) strategy, which is denoted as ND-hierarchical TMF (HTMF). ND is adopted to generate multiscale decomposition according to local image content, and that is superior to traditional wavelet decomposition in reflecting hierarchical nature of image structure and detailed features. The auxiliary field in ND-HTMF is redefined and initialised on the finest scale to characterise edge information and that enhances the prior modelling ability for non-stationary local image features. The multiscale likelihood and multiscale causal prior energy functions are then defined respectively in bottom-up and top-down procedures to capture local and global information for performing segmentation. Segmentation experiments on simulated and real SAR images demonstrate the effectiveness of ND-HTMF in both edge characterisation accuracy and robustness against speckle noise.

Classifying hyperspectral data within high dimensionality is a challenging task. To cope with this issue, this study implements a semi-supervised multi-kernel class consistency regulariser graph-based spatial–spectral feature learning framework. For feature learning process, establishing the neighbouring relationship between the distinct samples from the high-dimensional space is the key to a favourable outcome for classification. The proposed method implements two kernels and a class consistency regulariser. The first kernel constructs simple edges where every single vertex represents one particular sample and the edge weight encodes the initial similarity between distinct samples. Later the obtained relation is fed into the second kernel to obtain the final features for classification where the semi-supervised learning is conducted to estimate the grouping relations among different samples according to their similarity, class, and spatial information. To validate the performance of proposed framework, the authors conduct several experiments on three publically available hyperspectral datasets. The proposed work equates favourably with state-of-the-art works with an overall classification accuracy of 98.54, 97.83, and 98.38% for Pavia University, Salinas-A, and Indian Pines datasets, respectively.

The high performance of high dynamic range (HDR) image and the limitations of sensors impel to acquire HDR images from low dynamic range (LDR) ones. The most popular methods use multiple LDR images with different exposures of the same scene as input. However, it is difficult to have multiple exposures for most of LDR images. The authors propose a single image-based method to generate HDR image based on camera response function (CRF) reconstruction. The method first estimates the CRF according to the empirical model and the input LDR image. Then, the empirical model of the inverse CRF is constructed according to the relationship between the derivatives of CRF and its inverse function; the optimal solution of inverse CRF is obtained depending on the imaging properties of the edge pixels. Finally, the HDR image is generated by performing the inverse CRF on the original LDR image. The imitation of imaging procedure inherently makes the final HDR image high quality. The experimental results indicate that the proposed approach expand image from the dark and bright regions simultaneously. The resulting metric images also illustrate that their proposed method causes lower total contrast error compared with other single image-based methods.

The authors introduce an edge detection and recovery framework based on open active contour models (snakelets) to mitigate the problem of noisy or broken edges produced by classical edge detection algorithms, like Canny. The idea is to utilise the local continuity and smoothness cues provided by strong edges and grow them to recover the missing edges. This way, the strong edges are used to recover weak or missing edges by considering the local edge structures, instead of blindly linking edge pixels based on a threshold. The authors initialise short snakelets on the gradient magnitudes or binary edges automatically and then deform and grow them under the influence of gradient vector flow. The output snakelets are able to recover most of the breaks or weak edges and provide a smooth edge representation of the image; they can also be used for higher-level analysis, like contour segmentation.