Quantitative characterisation of blood vessels from images (e.g. morphometric analysis) is important to a variety of biomedical problems such as disease diagnosis and staging or assessment of angiogenesis. However, the accuracy of such characterisation depends heavily on the outcome of image preprocessing algorithms. Therefore, more efficient algorithms for vessel image segmentation or extraction have emerged within the past few years. Nevertheless, such methods may perform poorly or fail entirely for images with large noise, even after a careful tuning of parameters. Moreover, none of these methods intentionally considers the removal of structural noise (such as spots that obscure and/or are brighter than vessels). To address these issues, the authors propose a novel thresholding algorithm for capillary images by detecting the polarity in the circular profiles (PCPs) of image pixels. This can robustly distinguish tube-like objects from both cloud-like contaminations and structural noise. Extensive simulation studies based on multiple evaluation criteria suggest that the PCP algorithm typically has a superior performance over other representative approaches. Finally, they also demonstrate the satisfactory performance of the PCP method on real image data.

Histogram equalisation (HE), which redefines the distribution of grey-levels in image, is an important step in image processing to enhance the image quality. Until now, numerous HE techniques have been proposed, among which major numbers have focused on solving the problem of how the grey-levels in the histogram of an input image should be properly partitioned so that the image produced by collecting all equalisation results for the partitioned sub-histograms leads to the quality enhancement of image. However, the partition-based equalisation methods have an inherent limitation that it is not able to equalise a sub-histogram crossing a partition boundary, which is the main cause of image distortion. In this study, the authors propose a new HE method to overcome this limitation. Precisely, rather than constraining disjoint mapping ranges of the grey-levels among the partitions, they devise two enabling techniques: (i) a mapping range for each grey-level with no range-disjoint constraint and (ii) a mapping distance between two adjacent grey-levels to make a full exploitation of mapping flexibility of grey-levels. They embody the image's global intensity distribution of grey-levels in the first technique while they embody the image's context in the second one.

In this study, the authors propose a restoration algorithm for blurred and noisy continuous image sequences. The proposed approach treats an image sequence as a space-time volume and employs a spatio-temporal mean curvature regularisation which is a novel regularisation proposed in the study to enhance the smoothness of the solution. An augmented Lagrangian method with splitting techniques is used to handle the problem, iteratively finding solutions to the subproblems. Experiments show that the proposed approach can produce higher quality results and more natural images comparing with other space-time volume based methods on image sequence denoising and deblurring problems.

High-efficiency video coding (HEVC) is the newest video coding standard developed by the joint video team, consisting of ITU-T video coding experts group and ISO/IEC Moving Picture Experts Group. The HEVC standard has aggregated an exhaustive algorithm for mode decision based on a recursive quad-tree structured coding tree block. Moreover, several specific features have been incorporated into the motion estimation (ME) process to improve its coding efficiency. However, they resulted in very high computational complexity. To accelerate the encoding process, fast mode decision algorithms for the partitioning module and also for the ME module were proposed in this study. These algorithms are based on early zero block detection technique. To improve the efficiency of these algorithms, an overall algorithm which combines the two techniques has been implemented. The performance of the proposed algorithm was checked through a comparative analysis in terms of encoding time and compression rate. Compared to HEVC test model 10.0, the authors’ proposed algorithms bring a great reduction of the HEVC complexity encoder with a saving time, which can reach 25% in average for different tested videos and a slight coding loss in terms of image quality and compression rate.

Object tracking as a branch of computer vision plays a key role in the intelligent video surveillance. In recent years, the mean shift algorithm due to its simplicity and robustness has attracted great attention for tracking the object by using a colour model, while only using the colour causes the error in some cases of tracking such as illumination variations and so on. Consequently, the authors proposed an enhanced mean shift tracking algorithm. First, they presented a new texture-based target representation by a modified version of the interlaced derivative pattern, which considers spatial dependencies between pixels. Second, an improvement for the mean shift tracking algorithm based on this representation is suggested. In addition, a parameter to resize the window around the object is considered, adaptively. Experimental results on some benchmark video in comparison with other state-of-art methods show the efficiency and utility of the proposed algorithm in many complex conditions.

When dealing with salient object that contains several regions with different appearances, salient object detection can be a difficult task as often only parts of the salient object are highlighted and consistency between the salient regions is poor. This study tackles this problem by introducing objectness to assist the salient object detection. Rather than treating objectness in the same manner as other low-level cues (e.g. uniqueness, location etc.) for the determination of regional saliency values, the authors emphasise that objectness should also play a significant role in tuning the consistency between salient regions. The authors integrate objectness, uniqueness and centre bias to find potential salient regions and then enforce consistency between these regions using a full-connected Gaussian Markov random field with the weights determined by the objectness score. Experimental results on public benchmark datasets indicate that the authors’ method performs well on many images which cannot be well detected traditionally.

Image super-resolution (SR) for a single low-resolution image is an important and challenging task in image processing. In this study, the authors propose a novel non-local feature back-projection method for image SR, which can effectively reduce jaggy and ringing artefacts common, in general, iterative back-projection (IBP) method. In their method, the objective high-resolution (HR) image is obtained by projecting reconstructed errors back to HR image iteratively. To optimise the initial HR image and constrain anisotropic errors propagation during IBP process, an efficient non-local feature interpolation algorithm is designed. Specially, edge information is used as constraints to make the interpolation surface preserve better shape. Furthermore, as post-processing, non-local similarities are utilised to remove noise and irregularities induced by errors propagation. Experimental results show that their method achieves better performance than state-of-the-art methods in terms of both quantitative metrics and visual qualities.

In this study, the authors propose a new method to enhance image information, based on wavelet decomposition and original partial reconstruction of image. This reconstruction called ‘asynchronous reconstruction’ is not carried out in the same way as the usual sequential one. It is based on rank order coding. In fact, while sequential reconstruction is to sum all or a part of the responses obtained for each scale of ‘coarse to fine’ decomposition, asynchronous reconstruction tries to be closer to human brain which uses a limited number of frequency channels. Actually, after wavelet decomposition, responses are sorted from top down for each pixel of the image. Final asynchronous reconstruction for each pixel is obtained by adding a chosen number of wavelet responses, beginning by the maximum response. So, at a given level of reconstruction, the pixel values do not come from the same frequency channels. The interest of this method has been tested on a face verification task using the IV2 biometric database. Stopping criterion for reconstruction can be a constant number of wavelet responses to use, but an adaptive process has been also investigated. Three criteria are explored: standard deviation, entropy and lost edges ratio.

Similarity measure plays a significant task in intensity-based image registration. Nowadays, mutual information (MI) has been used as an efficient similarity measure for multimodal image registration. MI reflects the quantitative aspects of the information as it considers the probabilities of the voxels. However, different voxels have distinct efficiency towards the gratification of the elementary target, which may be self-reliant of their probability of occurrence. Therefore, both intensity distributions and effectiveness are essential to characterise a voxel. In this study, a novel similarity measure has been proposed by integrating the effectiveness of each voxel along with the intensity distributions for computing the enhanced MI using joint histogram of the two images. Penalised spline interpolation is incorporated to the joint histogram of the similarity measure, where each grid point is penalised with a weighted factor to avoid the local extrema and to achieve better registration accuracy as compared with existing methods with efficient computational runtime. To demonstrate the proposed method, the authors have used a challenging medical image dataset consisting of pre- and post-operative brain magnetic resonance imaging. The registration accuracy for the dataset improves the clinical diagnosis, and detection of growth of tumour in post-operative image.