Magnetic resonance (MR) images have great importance to assist doctors in diagnosing diseases, however, the long MR images scan duration remains the primary obstacle in clinical medicine. Compressed sensing reconstructed technique in MR imaging (CS-MRI) makes it possible to reconstruct a faithful MR image from very few measurements data, which helps to reduce the scan time. The purpose of this study is to improve the accuracy and efficiency of the CS-MRI. The authors propose a fast compressed sensing reconstruction model for MR images that can alleviate the aliasing artefacts that come from the reconstruction process by jointly minimising a total variation term, a fitting data term and a median filter term. Moreover, they accelerate the proposed algorithm by applying the split Bregman method to solve the proposed model. Then, the proposed model is applied to reconstruct a large number of MR images. They also compare the performance of the proposed model with another model. It can be observed from the experimental results that the proposed model has shown higher precision in reconstructing image quality and much more efficiency than the other one. Additionally, they also give a discussion on how to choose proper parameters in the proposed model to obtain more satisfactory results.

Dynamic texture (DT) classification has attracted extensive attention in the field of image sequence analysis. The probability distribution model, which has been used to analysis DT, can describe well the distribution property of signals. Here, the authors introduce the finite mixtures of Gumbel distributions (MoGD) and the corresponding parameter estimation method based on expectation–maximisation algorithm. Then, the authors propose the DT features based on MoGD model for DT classification. Specifically, after decomposing DTs with the dual-tree complex wavelet transform (DT-CWT), the median values of complex wavelet coefficient magnitudes of non-overlapping blocks in detail subbands are modelled with MoGDs. The model parameters are accumulated into a feature vector to describe DT. During the classification, a variational approximation version of the Kullback–Leibler divergence is used to measure the similarity between different DTs. The experimental evaluations on two popular benchmark DT data sets (UCLA and DynTex++) demonstrate the effectiveness of the proposed approach.

In the field of computer-aided recognition, edge feature is one of the key factors to determine recognition performance. Comparing to an optical image, since sonar image via acoustic wave is easily influenced by underwater environments such as particle density, temperature, and current, edge information should be boosted. Some image preprocessing techniques based on transform domain such as wavelet and curvelet may be good candidates but conventional methods show not only the possibility of enhancing edge features but also the limitation due to the absence of consideration to the edge direction. This study proposes an improved edge enhancement method based on curvelet transform (CVT), which is able to find out edge direction. The proposed method (PM) calculates the maximum value by ridgelet coefficients on each angular line, derived from the sub-step of the CVT, and the real edge direction is determined by local maxima selection after finding the azimuth of this value. In addition, selective sharpening is performed according to the feature information of edge. Experimental results have shown that the PM is comparable with conventional methods in terms of edge intensity, recognition rate, and peak signal-to-noise ratio.

Currently, the most successful model for image adaptive steganography is the framework of minimal distortion, in which a reasonable definition of costs can improve the security level. In the authors' previous work, they developed a rule for cost reassignment in spatial domain called the ‘controversial pixel prior (CPP)’ rule, which defines controversial pixels by utilizing the controversies among several comparable schemes. The CPP rule gives controversial pixels higher modification priorities. In this study, they investigate migrating the CPP rule from the spatial domain to the joint photographic experts group (JPEG) domain and name it the J-CPP rule. In JPEG images, the cover elements are discrete cosine transform (DCT) coefficients and variant factors mayinfluence the distortion definition includingquantisation step, inter-blocks correlation and block energy. However, there is no evidence to reveal which factor is of highest priority for promoting security. In this work, they investigate which factor is more helpful in promoting J-CPP rule, and they finally determine to set the spatial block residual as a penalty to perfect J-CPP rule. Through extensive experiments on different JPEG steganographic algorithms and steganalysis features, they demonstrate that the J-CPP rule can improve the security of JPEG adaptive steganography.

High-efficiency video coding (HEVC), which is the latest video coding standard, is expected to have a dominant position in the market in the near future. However, most video resources are now encoded using the H.264/AVC standard. Consequently, there is a growing need for fast H.264/AVC to HEVC transcoders to facilitate the migration to the updated standard. This paper proposes a fast H.264/AVC to HEVC transcoding scheme, which constructs a three-level classifier using an optimised tree-augmented Naive Bayesian approach to predict the HEVC coding unit depth. A feature selection method is then proposed to improve prediction accuracy. A motion vector (MV) calculation method is also proposed to reduce the complexity of MV prediction in HEVC by reusing MVs from H.264/AVC. Experimental results show that, compared with other state-of-the-art transcoding algorithms, the proposed algorithm considerably reduces coding complexity while causing only negligible rate-distortion degradation.

Spatially regularised discriminative correlation filters (SRDCFs) introduce spatial regularisation weights to mitigate the boundary effects caused by circular convolution which obtains superior performance. However, spatial regularisation is computationally expensive; this limits the real-time performance of SRDCF. This study proposes high-speed spatial constraint to DCFs (HSCDCFs) for tracking. Using a large area of the sample to learn a CF, then, the authors introduce the spatial constraint to penalise CF coefficients. Their method formulation allows the CFs to efficiently learn a mass of negative samples and high-quality positive samples. They perform experiments on two benchmark datasets: OTB-2013 and OTB-2015. Compared to SRDCF, they provide a slightly reduce of 2.7 and 3.1%, respectively, in mean overlap precision, their method obtains the real-time speed of 62.5 fps which is ten times faster than SRDCF.

Blind image deblurring of natural images still remains a demanding task. The traditional methods, pre-processes the uniform and non-uniform images with a deblurring algorithm and employs a low-rank prior algorithm. The rich textures do not possess enough similar patches in the deblurring process and this loss results in noisy images. Also, computational efficiency gets compromised during the performance of the succeeding process. In this study, the authors propose a novel method called, linearly uncorrelated principal component and deep convolution (LUPC-DC) for deblurring natural images. The natural images are first de-correlated with which good similar patches are extracted to generate a low-rank matrix by linearly uncorrelated principal component (PC) extraction. Then, the deep convolutional neural network model jointly extracts good similar patches and deblurs the first PCs. Eventually, good similar patches in the last PCs are suppressed using Hard Thresholding for computational efficiency. Analysis of concurrence performance of the algorithm confirms the viability of this method theoretically. In addition, simulation results and performance evaluations of image quality metrics are provided to assess the effectiveness of the proposed method. Moreover, the proposed method provides improvement in the peak-signal-to-noise ratio rate, success rate and reduction in the computation time for image deblurring.

It is difficult to recognise an image with affine transformation due to viewing angle and distance variations. Therefore, affine invariant feature extraction is a valuable technology in the field of image recognition. Inspired by bio-visual mechanism, an affine invariant for object recognition method based on a fusion feature framework is proposed in this study, which employs geometry descriptor and double biologically inspired transformation (DBIT). First, a shape feature of interest detector is adopted to detect contour features. Then, the area estimation of affine region detector is utilised to construct area ratio feature vectors. Second, an orientation edge detector is built to highlight the edges of different directions. On this basis, local space frequency detector is adopted to measure the spatial frequency at each direction and interval, which converts the output map into DBIT feature vectors. A weighted fusion strategy is performed based on Pearson correlation distance to fuse the geometry feature and DBIT feature. Some tests for Alphanumeric, Coil-100 MPEG-7, Mixed National Institute of Standards and Technology (MNIST) and Olivetti Research Laboratory face images database (ORL) database remain highly stable recognition accuracy, even when the shear factor is between −0.5 and  + 0.5. The experiment results show the authors’ proposed approach has a nice performance in feature invariance, selectivity and recognition accuracy.

Glaucoma is a class of eye disorder; it causes progressive deterioration of optic nerve fibres. Discrete wavelet transforms (DWTs) and empirical wavelet transforms (EWTs) are widely used methods in the literature for feature extraction using image decomposition. However, to increase the accuracy for measuring features of images a hybrid and concatenation approach has been presented in the proposed research work. DWT decomposes images into approximate and detail coefficients and EWT decomposes images into its sub band images. The concatenation approach employs the combination of all features obtained using DWT and EWT and their combination. Extracted features from each of DWT, EWT, DWTEWT and EWTDWT are concatenated. Concatenated features are normalised, ranked and fed to singular value decomposition to find robust features. Fourteen robust features are used by support vector machine classifier. The obtained accuracy, sensitivity and specificity are 83.57, 86.40 and 80.80%, respectively, for tenfold cross validation which outperforms the existing methods of glaucoma detection.

In the fusion process of medical computed tomography (CT) and magnetic resonance image (MRI), traditional multiscale methods often reduce the contrast of fused images. Although sparse representation (SR) methods overcome this shortcoming, they are often too smooth along the strong edges of the fusion image. To overcome these shortcomings, CT and MRI image fusion based on multiscale decomposition method and hybrid approach is proposed. There are three main steps. First, the cartoon parts and texture parts of CT and MRI are obtained by the improved image decomposition method using global sparse gradients. Second, the large structure cartoon parts are fused using the specific cartoon dictionary and the ‘L1-max norm’ principle. The textured parts are fused using non-subsampled contourlet transformation (NSCT) and the maximum energy rule. Finally, the final result is obtained by superimposing the fused cartoon part and the fused texture part. The experimental results demonstrate that the proposed method outperforms the state-of-the-art method SR and NSCT in terms of visual effect and objective quality.

In quantisation index modulation-based watermarking methods, each host coefficient is replaced by an adequate quantiser according to the embedded bit. However, after the attack, the distribution of the hosts does not remain the same, resulting in a failure to map coefficients to their correct bits. To maintain a good quality, many authors focused on improving the extraction, rather than playing on the embedding parameters. They used a reference message to define the parameters of the channel model after the attack. However, the available solutions were not effective, whereas the reference message cannot describe the channel alteration due to a shortage of samples, and the complexity of attacks modelling. To overcome these problems, the authors relied on a diversification technique, which allowed to enlarge and vary the samples. Then, based on quantisation embedding rules, they generated an accommodative detection model for each attack. At the extraction phase, the suitable detection model was identified by means of a reference message. To reinforce the security of their scheme, the watermark was initially shuffled, which prevents any unauthorised detection attempt. Both geometric and signal processing attacks were used in the validation of their method. The reduction of bit-error rate with respect to the conventional detection method reached up to 100%.

More global matching (MGM) overcomes the limitation of one-dimensional scanline optimisation in semi-global matching (SGM). Nevertheless, the possible weaknesses of the MGM algorithm are as follows: (i) only two directions are considered for each image traversal direction, which may lead to massive mismatches; (ii) disparity estimation around the object boundaries usually performs terrible since the smoothness term is designed independent of the image prior. In this research, the authors consider all of the four directions for each image traversal direction through a novel model. Besides utilising the prior of neighboured pixels' correlation, adaptive smoothness terms are modelled and augmented into the energy function. These contributions encourage ‘as-global-as-possible (AGAP)’. More importantly, different from the recent works in which the aggregated algorithms have been conducted as the data term of an energy function, conversely, the authors make the energy function as a part of cost aggregation framework. Performance evaluations on Middlebury v.2 and v.3 stereo data sets demonstrate that the proposed AGAP outperforms other four most challenging stereo matching algorithms, and also performs better on Microsoft i2i stereo videos. In addition, under various strategies of parallelisation, the presented AGAP shows a near real-time execution time.

The world is chasing towards the automation in the severity analysis and classification of the patients based on the severity of tuberculosis (TB). The automatic classification is very much useful for developing countries that are struggling to reduce the fatality rate of the persons suffering from TB as it is a top standing infectious disease. Thus, the automatic classification of the TB patients using the sputum images with the proposed fractional crow search-based support vector neural network is presented. The proposed classification method is the integration of the fractional theory in the crow search algorithm that increases the computational speed and reduces the cost and time spent on analysing the test samples. The importance of the proposed method is that it requires minimum manual power and hence, the inaccuracies are reduced. The experimentation performed using the Ziehl–Neelsen sputum smear microscopy image database proves that the proposed classifier is highly accurate and offered an improved performance in terms of accuracy rate, true positive rate, and false-positive rate.

This study presents the exact Legendre–Fourier moments and a novel arrangement of polar pixels that allows calculating orthogonal moments defined in a unit radius more accurately than traditional methods. This arrangement simplifies implementation and preserves the values of the pixels of the image during the calculation of the moments. Moreover, the exact Legendre–Fourier moments use the weighted substituted radial shifted Legendre polynomials as kernel, which has the ability to accurately calculate the circular moments. Finally, the author presents a comparative analysis of the reconstruction error with existing configurations and other families of circular moments. The results indicate that the mehod provides a significant advantage.

Chaotic mapping has been widely used in image encryption given the unpredictability, ergodicity, and sensitivity of parameters and initial values and the high correspondence with cryptography. The logistic map has the disadvantages of uneven distribution, low security, and small parameter space. In order to overcome these disadvantages, in this article, a new chaotic map based on a real-time variable logistic map with a randomly selected decimal is proposed. Furthermore, this chaotic mapping is applied to image encryption. Several simulation experiments show that the new encryption algorithm can obtain a safely encrypted image at a minimal time complexity.

In machine learning, the cost function is crucial because it measures how good or bad a system is. In image classification, well-known networks only consider modifying the network structures and applying cross-entropy loss at the end of the network. However, using only cross-entropy loss causes a network to stop updating weights when all training images are correctly classified. This is the problem of early saturation. This study proposes a novel cost function, called mixture separability loss (MSL), which updates the weights of the network even when most of the training images are accurately predicted. MSL consists of between-class and within-class loss. Between-class loss maximises the differences between inter-class images, whereas within-class loss minimises the similarities between intra-class images. They designed the proposed loss function to attach to different convolutional layers in the network in order to utilise intermediate feature maps. Experiments show that a network with MSL deepens the learning process and obtains promising results with some public datasets, such as Street View House Number, Canadian Institute for Advanced Research, and the authors’ self-collected Inha Computer Vision Lab gender dataset.

Interactive segmentation has recently become a hot topic for its wide application. The authors propose an efficacious appearance separation model for interactive binary segmentation, which incorporates the difference of foreground and background colour models and the difference of corresponding geodesic models into the popular densely connected conditional random field (Dense CRF) framework. The proposed method can adaptively set relevant parameter values in this framework according to the characteristics of target images in a per-image manner, therefore, it gets rid of the dependence on specific datasets. After accomplishing a mean-field inference, the authors are able to get satisfactory results without the time-consuming parameter learning process and multiple iterative optimisations. Overall, the proposed approach is highly efficient and mitigates the contradiction between accuracy and segmentation efficiency. In addition, the proposed approach reduces the efforts of scribble-style interaction from users. The experimental results on three famous datasets show that the proposed method is superior to the other five new algorithms released in recent years regarding accuracy, and is faster than or close to them in runtime.

Many effective image deformation methods have been proposed in recent years, but few of them use the contours of objects as the reference of deformation. However, the contour is an important factor, and contour-based deformation can guarantee that the edge of the object after deformation is smoother compared to point-based image deformation. This article presents an image deformation method based on contours using Moving Integral Least Squares (MILS) optimisation. First, the authors set the key points in the image to create control contours as required and then adjust the positions of these points to generate the desired contours. In order to warp these contours to their new positions, the image is deformed using MILS. They derive the affine, similarity, and rigid transformations in a general framework, and users can choose different curves according to their needs. The proposed method possesses two characteristics: (i) it is able to create detail-preserving and intuitive deformations; and (ii) the solution of the deformation function has a simple closed form. They compare their method to the state-of-the-art algorithm, which is modelled by rigid transformation. Experimental results show that their deformation is more vivid.

This study addresses two issues from batch clustering using K-means algorithm in colour image classification application. One of the major issues is the drifting phenomenon in the batch clustering due to the stochastic nature of the clustering procedure. Also in literature, the initial parameter is important to direct the clustering algorithm converge to the proper local solution. In this study, a new algorithm is proposed to address these two issues in application. Recently, a research found that the principal component analysis (PCA) result directly indicates the membership of the clusters in K-means algorithm. Hence using this, the first part of the proposed algorithm shows the possibility to estimate the initial parameters accurately for K-means with a hierarchical manner of PCA solution. In addition, a gradient descent approach is used for the global batch clustering to reduce the drifting and hence speed up convergence in the refining stage. All necessary proofs and justifications are also provided. The evaluation study has shown that the proposed algorithm performs better than the original K-means clustering algorithms with various initial parameter estimation processes.

The authors propose a tracking algorithm based on the reliability analysis of the convolutional neural network to avoid drift. In general, most tracking algorithms implemented with the deep network consist of a single network; they obtain the tracking results according to the confidence and perform updates with the samples, which are collected based on the previous target state. However, this kind of algorithm relies heavily on the accuracy of tracking results, and slight deviations can lead to improperly labelled training samples and degrade the network. Therefore, they design a verification network to guarantee the reliability of the tracking network by correcting the results and it can be connected to a tracking network by sharing convolutional layers. The reliability verification network estimates the accuracy of the results of the tracking network and discards ambiguous results to avoid accumulating errors. Specifically, the verification network can distinguish the target from the confused candidates more precisely because of the optimised training data. The training samples of the verification network consist of characteristics and labels, and they are optimised by feature selection and label enhancement, respectively. The experimental results illustrate the outstanding performance compared with several state-of-the-art methods on the challenging video sequences.

This study presents a multi-method fusion and optimisation framework that can optimally combine different existing methods to further enhance the segmentation performance. The framework, in which the original accumulating process is improved and a new combination process is added, is the extension of the previously developed ‘accumulated local fuzzy c-means with spatial information’ method. In the improved accumulating process, different segmentation methods are utilised in local windows to judge whether each pixel belongs to the object. In the new combination process, the accumulated results of different segmentation methods are weighted combined, where the weights of different methods are optimised by the genetic algorithm with the objective of minimising standard deviations of both the object and the background pixels. Typical images and all images in the Weizmann's Segmentation Evaluation Database are tested in the experiments. The results show that the authors’ method can perform better than some state-of-the-art methods, and combining more methods in the framework can bring better performance. Moreover, the proposed multi-method combination framework is parameterless, which increases its adaptability in various applications.

When utilising matrix factorisation to extract latent features for cross-media retrieval, semantic information may be lost in the process of factorisation. In addition, many presented approaches directly mapped different modalities into an isomorphic semantic space to conduct the similarity measurement of different modalities, which also resulted in the loss of crucial information. To address these problems, a semantic convex matrix factorisation subspace learning approach is proposed for cross-media retrieval between image and text. The proposed method can extract an intermediate-level feature representation for the high dimensional image modality in order to weaken the loss of information, in the meantime, learn a semantic feature representation with semantic information for the lower dimension text modality to strengthen the discriminated capability. After that, the intermediate-level feature representation of image is mapped into a latent semantic space by a projection matrix. Then the similarity of different modalities can be estimated in terms of uniform dimensional latent feature representations. Experimental results on three benchmark datasets demonstrate the superiority of the proposed approach over several state-of-the-art approaches.

Two-dimensional transforms are extensively used for speckle noise reduction (despeckling) in ultrasound images. This work proposes a double filter bank structure consisting of a discrete wavelet packet transform (DWPT) and directional filter bank (DFB) along with a fuzzy-based clustering technique to despeckle ultrasound images. Wavelet packet transform can efficiently reject noises based on grey scale relational thresholding and DFB can efficiently preserve edge information. In this study, instead of conventional thresholding methods, fuzzy-based clustering techniques are applied for noise rejection. The algorithm provides a consistent improvement over the competing state-of-the-art speckle reduction algorithms due to the improved ability to preserve geometric features while rejecting speckle noise adaptively. The authors claim is validated by applying a number of clinical images with performance indices such as peak signal to noise ratio, mean structural similarity, signal to mean square error, speckle signal to noise ratio, speckle suppression index and edge preservation index.

Thermal images have been widely used for detection, tracking, and classification of targets at night for military purpose. A thermal imaging sensor receives the radiation energy from the target and the background, so it has advantages in night vision. However, the thermal images have lower spatial resolution and more blurred edges than colour images, and edges can be contaminated by flames. Therefore, the extracted edges which can be used in automatic target classification may contain inaccurate edges that can decrease the accuracy of the classification. In this study, to overcome these problems, a novel background modelling method based on spatiotemporal parameter updating technique and foreground detection method using intensity moments analysis is proposed. To validate the proposed method, thermal images with military vehicle targets (tank and infantry vehicle) and public dataset in CDNET2014 (www.changedetection.net) are used in subjective and objective tests, respectively. The experiment results showed that the proposed method outperformed the state-of-the-art methods in objective measures and in subjective measure especially on detail-enhanced segmentation.

Detection of Betacam dropout defects that can occur in the digitisation process of old archived media has importance in the restoration of degraded data to a higher quality. Most of the existing methods rely on the temporal information of multiple consecutive frames to detect Betacam dropouts, which sometimes may not work well as several successive frames may contain a Betacam error at the same position. In this study, an automatic method of Betacam dropout error detection is proposed based on vertical patterns in a single frame. Hence, it is also applicable when temporal information-based detectors fail. The results of performance tests done in real working environments demonstrate that the proposed Betacam dropout detection method performs much better than the existing methods.