Search

Graph regularized nonnegative matrix factorization (GNMF) has been widely used in data representation due to its excellent dimensionality reduction. When it comes to clustering polluted data, GNMF inevitably learns inaccurate representations, leading to models that are unusually sensitive to outliers in the data. For example, in a face dataset, obscured by items such as a mask or glasses, there is a high probability that the graph regularization term incorrectly describes the association relationship for that sample, resulting in an incorrect elicitation in the matrix factorization process. In this article, a novel self-initiated unsupervised subspace learning method named robust nonnegative matrix factorization with self-initiated multigraph contrastive fusion (RNMF-SMGF) is proposed. RNMF-SMGF is capable of creating samples with different angles and learning different graph structures based on these different angles in a self-initiated method without changing the original data. In the process of subspace learning guided by graph regularization, these different graph structures are fused into a more accurate graph structure, along with entropy regularization, L 2,1/2 -norm constraints to facilitate the robust learning of the proposed model and the formation of different clusters in the low-dimensional space. To demonstrate the effectiveness of the proposed model in robust clustering, we have conducted extensive experiments on several benchmark datasets and demonstrated the effectiveness of the proposed method. The source code is available at: https://github.com/LstinWh/RNMF-SMGF/.

We aimed to investigate the dysregulation of the microRNAs(miRNAs) in cholangiocarcinoma (CCA), including its impact on the homeostasis of the transcriptome and cellular behavior. MiRNAs serve as potent epigenetic regulators of transcriptional output, targeting various signaling pathways. This study aimed to investigate the expression level, epigenetic mechanism and function of miR-125a-3 in CCA. The study data showed that the expression level of miR125a-3p was decreased in CCA tissue samples and cell lines, and it was closely related to lymph node metastasis, tissue differentiation and TNM stage. The data demonstrate a strong association between decreased miR-125a-3p expression and poorer prognosis in cholangiocarcinoma patients. miR-125a-3p acts as a tumor suppressor by inhibiting the viability, migration and invasion of CCA cells. There are CpG islands in the promoter region of miR-125a-3p gene, and the methylation of the promoter region of miR-125a-3p gene leads to the transcriptional repression of miR-125a-3p. In addition, miR125a-3p can target and regulate CAC1 mRNA and protein expression in the downstream mechanism, and the high expression of CAC1 can promote the proliferation, migration and invasion of cholangiocarcinoma cells. These data demonstrate that miR-125a-3p promoter methylation leads to silencing of its expression. Mechanically, miR-125a-3p acts as a tumor suppressor and participates in the occurrence and development of CCA through targeting CAC1 gene expression. Therefore, miR-125a-3p may serve as a new target for the diagnosis, prognostic assessment or molecular therapy of CCA., Competing Interests: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (© 2024 The Author(s).)

Objective: Observational studies have reported that mental disorders are comorbid with temporomandibular joint disorder (TMD). However, the causal relationship remains uncertain. To clarify the causal relationship between three common mental illnesses and TMD, we conduct this Mendelian Randomization (MR) study., Methods: The large-scale genome-wide association studies data of major depression, bipolar disorder and schizophrenia were retrieved from the Psychiatric Genomics Consortium. The summary data of TMD was obtained from the Finn-Gen consortium, including 211,023 subjects of European descent (5,668 cases and 205,355 controls). The main approach utilized was inverse variance weighting (IVW) to evaluate the causal association between the three mental disorders and TMD. Five sensitivity analyses including MR-Egger, Maximum Likelihood, Weighted median, MR. RAPS and MR-PRESSO were used as supplements. We conducted heterogeneity tests and pleiotropic tests to ensure the robustness., Results: As shown by the IVW method, genetically determined major depression was associated with a 1.65-fold risk of TMD (95% CI = 1.10-2.47, p < 0.05). The direction and effect size remained consistent with sensitivity analyses. The odds ratios (ORs) were 1.51 (95% CI = 0.24-9.41, p > 0.05) for MR-Egger, 1.60 (95% CI = 0.98-2.61, p > 0.05) for Weighted median, 1.68 (95% CI = 1.19-2.38, p < 0.05) for Maximum likelihood, 1.56 (95% CI = 1.05-2.33, p < 0.05) for MR. RAPS, and 1.65 (95% CI = 1.10-2.47, p < 0.05) for MR-PRESSO, respectively. No pleiotropy was observed (both P for MR-Egger intercept and Global test >0.05). In addition, the IVW method identified no significant correlation between bipolar disorder, schizophrenia and TMD., Conclusion: Genetic evidence supports a causal relationship between major depression and TMD, instead of bipolar disorder and schizophrenia. These findings emphasize the importance of assessing a patient's depressive status in clinical settings., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Wu, Chen, Zhao, He, Yin, Yao, Liu and Yan.)

Gaze-based implicit intention inference provides a new human-robot interaction for people with disabilities to accomplish activities of daily living independently. Existing gaze-based intention inference is mainly implemented by the data-driven method without prior object information in intention expression, which yields low inference accuracy. Aiming to improve the inference accuracy, we propose a gaze-based hybrid method by integrating model-driven and data-driven intention inference tailored to disability applications. Specifically, intention is considered as the combination of verbs and nouns. The objects corresponding to the nouns are regarded as intention-interpreting objects and served as prior knowledge, i.e., punished factors. The punished factor considers the object information, i.e., the priority in object selection. Class-specific attribute weighted naïve Bayes model learned through training data is presented to represent the relationship among intentions and objects. An intention inference engine is developed by combining the human prior knowledge, and the data-driven class-specific attribute weighted naïve Bayes model. Computer simulations: (i) verify the contribution of each critical component of the proposed model, (ii) evaluate the inference accuracy of the proposed model, and (iii) show that the proposed method is superior to state-of-the-art intention inference methods in terms of accuracy.

Background and Objective: Left atrial enlargement (LAE) is an anatomical variation of the left atrium and the result of the long-term increase of left atrial pressure. Most of the increase in stress or volume is due to potential cardiovascular disease. Studies have shown that LAE can independently predict the development of clinically significant cardiovascular disease and heart failure. If the left atrial volume is accurately measured, it will be an essential indicator of human health and an essential means for doctors to find patients' potential diseases. We can analyze the dynamic changes in the left atrial structure and analyze left atrial dilation. However, manual segmentation was inefficient and error-prone before the 3D reconstruction of the left atrium. In order to solve this problem, a convolution neural network (CNN) method based on cardiac magnetic resonance image (MRI) is proposed to automatically segment the left atrial region., Methodology: In this paper, we have proposed and developed a novel U-Net with Gaussian blur and channel weight neural network (GCW-UNet) to automatically segment the left atrial region in the MRI of a patient with LAE. After Gaussian blur, different resolutions of the MRI are obtained. High-resolution MRI clearly shows the detailed features of the left atrium, while low-resolution MRI clearly shows the overall outline of the left atrium, which can solve the problem of more minor MRI features. Adaptive channel weights can enhance the atrial segmentation capability of the network., Results: Compared with the state-of-the-art left atrial segmentation methods, our CNN-based technique results in the segmentation of the left atrium being closer to the manual segmentation by an experienced radiologist. On the test datasets, the mean Dice similarity coefficient reaches 93.57%., Conclusion: Firstly, MRI has a small number of imaging artifacts, which results in low segmentation accuracy. Our method successfully solves the problem. Secondly, due to the high similarity between the background (the area outside the left atrium) and the foreground (the left atrium) in MRI, traditional neural networks misclassify the background as the foreground. Our GCW-Unit can address the imbalanced number of pixels between the foreground and background. Finally, after segmenting the left atrium in the MRI by GCW-Unit, we reconstructed the left atrium to model a three-dimensional heart of a patient suffering from LAE. Based on the different time frames of one heartbeat, we could present the dynamics of the left atrial structure during a cardiac cycle. This can better assist in the evaluation of LAE in heart patients., Competing Interests: Declaration of Competing Interest None., (Copyright © 2022. Published by Elsevier B.V.)

Objective: To present and validate a method for automated identification of the Lagrangian vortices and Eulerian vortices for analyzing flow within the right atrium (RA), from phase contrast magnetic resonance imaging (PC-MRI) data., Methodology: Our proposed algorithm characterizes the trajectory integral associated with vorticity deviation and the spatial mean of vortex rings, for the Lagrangian averaged vorticity deviation (LAVD) based identification and tracking of vortex rings within the heart chamber. For this purpose, the optical flow concept was adopted to interpolate the time frames between larger discrete frames, to minimize the error caused by constructing a continuous velocity field for the integral process of LAVD. Then the Hough transform was used to automatically extract the vortex regions of interest. The computed flow data within the RA of the participants' hearts was then used to validate the performance of our proposed method., Results: In the paper, illustrations are provided for derived evolution of Euler vortices and Lagrangian vortices of a healthy subject. The visualization results have shown that our proposed method can accurately identify the Euler vortices and Lagrangian vortices, in the context of measuring the vorticity and vortex volume of the vortices within the RA chamber. Then the employment of Hough transform-based automated vortex extraction has improved the robustness and scalability of the LAVD in identifying cardiac vortices. The analytical results have demonstrated that the introduction of the Horn-Schunck optical flow can more accurately synthesize the intermediate PC-MRI to construct a continuous velocity field, compared with other interpolation methods., Conclusion: A novel analytical framework has been developed to accurately identify the flow vortices in the RA chamber based on Horn-Schunck optical flow and Hough transform. From the obtained analytical study results, the development and changes of dominant vortices within this cardiac chamber during the cardiac cycle can be acquired. This can provide to cardiologists a deeper understanding of the hemodynamics within the heart chambers., Competing Interests: Declaration of competing interest The authors declare that there is no conflict of interests., (Copyright © 2022 Elsevier B.V. All rights reserved.)

In phase-measurement profilometry (PMP), the gamma effect can cause severe nonlinear distortion of the phase pattern (i.e., water ripples on the surface profile). Gamma correction is an effective method to eliminate the gamma effects of commercial projectors. However, projector defocusing on the suppression of higher harmonics inevitably results in an estimated gamma deviation from the true value. In this study, gamma mapping is constructed using the duty ratio (DR) to code the chord distribution of the simulated distorted phase while considering projector defocusing. With the known gamma mapping, the accurate gamma is calculated by DR coding of the actual distorted phase under projector defocusing. Simulated experiments verified that the relative errors of the gamma calculated by the proposed method under different degrees of defocus were less than 3.5%. Furthermore, the experimental results demonstrate that the proposed gamma calculation method is robust to the defocus effect of the projector and that a smoother surface can be reconstructed after gamma correction.

Pose estimation is a particularly important link in the task of robotic bin-picking. Its purpose is to obtain the 6D pose (3D position and 3D posture) of the target object. In real bin-picking scenarios, noise, overlap, and occlusion affect accuracy of pose estimation and lead to failure in robot grasping. In this paper, a new point-pair feature (PPF) descriptor is proposed, in which curvature information of point-pairs is introduced to strengthen feature description, and improves the point cloud matching rate. The proposed method also introduces an effective point cloud preprocessing, which extracts candidate targets in complex scenarios, and, thus, improves the overall computational efficiency. By combining with the curvature distribution, a weighted voting scheme is presented to further improve the accuracy of pose estimation. The experimental results performed on public data set and real scenarios show that the accuracy of the proposed method is much higher than that of the existing PPF method, and it is more efficient than the PPF method. The proposed method can be used for robotic bin-picking in real industrial scenarios.

Model-based single image dehazing algorithms restore haze-free images with sharp edges and rich details for real-world hazy images at the expense of low PSNR and SSIM values for synthetic hazy images. Data-driven ones restore haze-free images with high PSNR and SSIM values for synthetic hazy images but with low contrast, and even some remaining haze for real-world hazy images. In this paper, a novel single image dehazing algorithm is introduced by combining model-based and data-driven approaches. Both transmission map and atmospheric light are first estimated by the model-based methods, and then refined by dual-scale generative adversarial networks (GANs) based approaches. The resultant algorithm forms a neural augmentation which converges very fast while the corresponding data-driven approach might not converge. Haze-free images are restored by using the estimated transmission map and atmospheric light as well as the Koschmieder's law. Experimental results indicate that the proposed algorithm can remove haze well from real-world and synthetic hazy images.

Objective: The measurement of cardiac blood flow vortex characteristics can help to facilitate the analysis of blood flow dynamics that regulates heart function. However, the complexity of cardiac flow along with other physical limitations makes it difficult to adequately identify the dominant vortices in a heart chamber, which play a significant role in regulating the heart function. Although the existing vortex quantification methods can achieve this goal, there are still some shortcomings: such as low precision, and ignoring the center of the vortex without the description of vortex deformation processes. To address these problems, an optical flow Lagrangian averaged vorticity deviation (Optical flow-LAVD) method is proposed. Methodology: We examined the flow within the right atrium (RA) of the participants' hearts, by using a single set of scans pertaining to a slice at two-chamber short-axis orientation. Toward adequate extraction of the vortex ring characteristics, a novel approach driven by the Lagrangian averaged vorticity deviation (LAVD) was implemented and applied to characterize the trajectory integral associated with vorticity deviation and the spatial mean of rings, by using phase-contrast magnetic resonance imaging (PC-MRI) datasets as a case study. To interpolate the time frames between every larger discrete frame and minimize the error caused by constructing a continuous velocity field for the integral process of LAVD, we implemented the optical flow as an interpolator and introduced the backward warping as an intermediate frame synthesis basis, which is then used to generate higher quality continuous velocity fields. Results: Our analytical study results showed that the proposed Optical flow-LAVD method can accurately identify vortex ring and continuous velocity fields, based on optical flow information, for yielding high reconstruction outcomes. Compared with the linear interpolation and phased-based frame interpolation methods, our proposed algorithm can generate more accurate synthesized PC-MRI. Conclusion: This study has developed a novel Optical flow-LAVD model to accurately identify cardiac vortex rings, and minimize the associated errors caused by the construction of a continuous velocity field. Our paper presents a superior vortex characteristics detection method that may potentially aid the understanding of medical experts on the dynamics of blood flow within the heart., Competing Interests: DG was employed by company University 2020 Foundation, Inc. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2021 Yang, Wu, Samuel, Zhang, Ghista, Yang and Wong.)

Achieving high visibility and high signal-to-noise ratio (SNR) from a single-shot image captured in low-light environments is an under-constrained problem. To cope with this issue, the intrinsic relationship between the image domain and the radiance domain is first established based on the human visual model, the atmospheric scattering model, and the camera imaging model, and the ideal exposure is derived. Using the illumination-reflection-noise prior, a new convex optimization by employed gradient constraint and the Krisch operator is then presented to estimate the noise-reduced illumination and reflection components. A high SNR image in the optimal exposure is generated in radiance domain, which is finally inversely mapped to obtain a high SNR image in image domain. Experimental results in subjective and objective tests show that the proposed algorithm has a high SNR and pleasant perception in comparison with the state-of-the-art methods.

The formation of vortex rings in the left ventricular (LV) blood flow is a mechanism for optimized blood transport from the mitral valve inlet to aortic valve outlet, and the vorticity is an important measure of a well-functioning LV. However, due to lack of quantitative methods, the process of defining the boundary of a vortex in the LV and identifying the dominant vortex components has not been studied previously. The Lagrangian-averaged vorticity deviation (LAVD) can enable us to compute the trajectory integral of the normed difference of the vorticity from its spatial mean. Therefore, in this work, we have employed LAVD to identify the Lagrangian vortices and Eulerian vortices for measuring the vortex volume and vorticity in the LV blood flow. We found that during the LV ejection period, the positive (counterclockwise) and negative (clockwise) vorticity of patients are consistently stronger than those of the healthy groups, and the counterclockwise vortex volume of healthy groups (0.84+0.26 ml) is greater than that of patients (0.55+0.28 ml) during the pre-ejection period. Then, during the middle ejection phase, the counterclockwise vortex ring volume of patients (1.89+0.36 ml) exceeds that of healthy groups (1.38+0.43 ml). Finally, during the end-ejection period, the counterclockwise vortex ring volume of healthy subjects (0.61+0.17 ml) is the same as that of patients (0.60+0.19 ml). The results presented in this paper can provide new insights into the blood flow patterns within the LV. It can accurately indicate the role of vortices and vorticity values in intra-LV flow, and portray how cardiomyopathy (and its distorted contractile mechanism) can affect intra-LV flow patterns and mitigate adequate LV outflow., (© 2021. International Federation for Medical and Biological Engineering.)

Phase measuring deflectometry (PMD) is an effective technique for three-dimensional measurement of specular surfaces. However, the ambiguity of monoscopic PMD and the time-consuming searching process of stereoscopic PMD are challenges for specular surface reconstruction. To solve it, we propose an iterative reconstruction algorithm for the stereoscopic phase measuring deflectometry system free of the time-consuming searching process for each pixel. An arbitrary seed point on the specular surface is accurately obtained via a coarse-to-fine optimization means without any other expensive and complicate auxiliaries. Then, a plane with the height of seed point is set as the initial surface form for the iteration, in which the pinhole model is used to find the linear relation to update the surface form. The converging height is the output as the final result. Simulations and experiments verify the feasibility and efficiency of our proposed method based on the stereoscopic phase measuring deflectometry system. The accuracy and robustness are comprehensively evaluated as well.

In cardiology, ultrasound is often used to diagnose heart disease associated with myocardial infarction. This study aims to develop robust segmentation techniques for segmenting the left ventricle (LV) in ultrasound images to check myocardium movement during heartbeat. The proposed technique utilizes machine learning (ML) techniques such as the active contour (AC) and convolutional neural networks (CNNs) for segmentation. Medical experts determine the consistency between the proposed ML approach, which is a state-of-the-art deep learning method, and the manual segmentation approach. These methods are compared in terms of performance indicators such as the ventricular area (VA), ventricular maximum diameter (VMXD), ventricular minimum diameter (VMID), and ventricular long axis angle (AVLA) measurements. Furthermore, the Dice similarity coefficient, Jaccard index, and Hausdorff distance are measured to estimate the agreement of the LV segmented results between the automatic and visual approaches. The obtained results indicate that the proposed techniques for LV segmentation are useful and practical. There is no significant difference between the use of AC and CNN in image segmentation; however, the AC method could obtain comparable accuracy as the CNN method using less training data and less run-time., Competing Interests: Declaration of Competing Interest We declare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our work, there is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as influencing the position presented in, or the review of, this manuscript. We have followed the author guidelines in preparing the manuscript, and I confirm that the paper is not under consideration for publication elsewhere., (Copyright © 2020. Published by Elsevier B.V.)

Feature point detection is the basis of computer vision, and the detection methods with geometric invariance and illumination invariance are the key and difficult problem in the field of feature detection. This paper proposes an illumination-invariant feature point detection method based on neighborhood information. The method can be summarized into two steps. Firstly, the feature points are divided into eight types according to the number of connected neighbors. Secondly, each type of feature points is classified again according to the position distribution of neighboring pixels. The theoretical deduction proves that the proposed method has lower computational complexity than other methods. The experimental results indicate that, when the photometric variation of the two images is very large, the feature-based detection methods are usually inferior, while the learning-based detection methods performs better. However, our method performs better than the learning-based detection method in terms of the number of feature points, the number of matching points, and the repeatability rate stability. The experimental results demonstrate that the proposed method has the best illumination robustness among state-of-the-art feature detection methods.

Capturing high-quality images via mobile devices in low-light or backlighting conditions is very challenging. In this paper, a new, single image brightening algorithm is proposed to enhance an image captured in low-light conditions. Two virtual images with larger exposure times are generated to increase brightness and enhance fine details of the underexposed regions. In order to reduce the brightness change, the virtual images are generated via intensity mapping functions (IMFs) which are computed using available camera response functions (CRFs). To avoid possible color distortion in the virtual image due to one-to-many mapping, a least square minimization problem is formulated to determine brightening factors for all pixels in the underexposed regions. In addition, an edge-preserving smoothing technique is adopted to avoid noise in the underexposed regions from being amplified in the virtual images. The final brightened image is obtained by fusing the original image and two virtual images via a gradient domain guided image filtering (GGIF) based multiscale exposure fusion (MEF) with properly defined weights for all the images. Experimental results show that the relative brightness and color are preserved better by the proposed algorithm. The details in bright regions are also preserved well in the final image. The proposed algorithm is expected to be useful for computational photography on smart phones.

Geometric model fitting is a fundamental issue in computer vision, and the fitting accuracy is affected by outliers. In order to eliminate the impact of the outliers, the inlier threshold or scale estimator is usually adopted. However, a single inlier threshold cannot satisfy multiple models in the data, and scale estimators with a certain noise distribution model work poorly in geometric model fitting. It can be observed that the residuals of outliers are big for all true models in the data, which makes the consensus of the outliers. Based on this observation, we propose a preference analysis method based on residual histograms to study the outlier consensus for outlier detection in this paper. We have found that the outlier consensus makes the outliers gather away from the inliers on the designed residual histogram preference space, which is quite convenient to separate outliers from inliers through linkage clustering. After the outliers are detected and removed, a linkage clustering with permutation preference is introduced to segment the inliers. In addition, in order to make the linkage clustering process stable and robust, an alternative sampling and clustering framework is proposed in both the outlier detection and inlier segmentation processes. The experimental results also show that the outlier detection scheme based on residual histogram preference can detect most of the outliers in the data sets, and the fitting results are better than most of the state-of-the-art methods in geometric multi-model fitting.

Calibration is a critical step for the phase measuring deflectometry system. Existing calibration methods are mainly optimizing the calibration parameters with respect to the 2D re-projection error criterion. However, such a procedure cannot reduce metric errors in the practical application. Therefore, an accurate and practical calibration method is proposed. In which, conventional calibration means is first applied for the primary calibration. Then, a precise square planar mirror is used for the optimization of system calibration parameters. All the intrinsic and extrinsic parameters are considered as a global multi-objective optimization problem. Three metric error criteria are introduced to evaluate the 3D reconstruction accuracy of the reference mirror. Compared with classical calibration means, which apply the parameter optimization in 2D image space to minimize the re-projection errors, the proposed optimization approach is executed in 3D space directly. An experiment and comparison are conducted to verify that the proposed optimal calibration approach can effectively reduce the system deviation and to improve the system measurement accuracy., Competing Interests: The authors declare no conflict of interest.

Estimating the Direction of Arrival (DOA) is a basic and crucial problem in array signal processing. The existing DOA methods fail to obtain reliable and accurate results when noise and reverberation occur in real applications. In this paper, an accurate and robust estimation method for estimating the DOA of sources signal is proposed. Incorporating the Estimating Signal Parameters via Rotational Invariance Techniques (ESPRIT) algorithm with the RANdom SAmple Consensus (RANSAC) algorithm gives rise to the RAN-ESPRIT method, which removes outliers automatically in noise-corrupted environments. In this work, a uniform circular array (UCA) is converted into a virtual uniform linear array (ULA) to begin with. Then, the covariance matrix of the received signals of the virtual linear array is reconstructed, and the ESPRIT algorithm is deployed to estimate initial DOA of the source signal. Finally, the modified RANSAC method with automatically selected thresholds is used to fit the source signal to obtain accurate DOA. The proposed method can remove the unreliable DOA feature data and leads to more accuracy of DOA estimation of source signals in reverberation environments. Experimental results demonstrate that the proposed method is more robust and efficient compared to the traditional methods (i.e., ESPRIT, TLS-ESPRIT)., Competing Interests: The authors declare no conflict of interest.

It is challenging to convert a hazy color image into a gray-scale image because the color contrast field of a hazy image is distorted. In this paper, a novel decolorization algorithm is proposed to transfer a hazy image into a distortionrecovered gray-scale image. To recover the color contrast field, the relationship between the restored color contrast and its distorted input is presented in CIELab color space. Based on this restoration, a nonlinear optimization problem is formulated to construct the resultant gray-scale image. A new differentiable approximation solution is introduced to solve this problem with an extension of the Huber loss function. Experimental results show that the proposed algorithm effectively preserves the global luminance consistency while represents the original color contrast in gray-scales, which is very close to the corresponding ground truth gray-scale one.

Decolorization is to convert a color image into a gray scale image while preserve image features like salient structure and chrominance contrast. The sign of the color contrast is crucial for the decolorization algorithm and is usually determined in existing works by giving a strict defined color order or twomode weak order. In this paper, a fast computation on color order is achieved via a simple global mapping which is introduced in a linear parametric model using an extended structure transfer filter. The values of the parameters are obtained via an elegant approximation method. A local decolorization algorithm is finally designed on basis of the global linear mapping so that both color and spatial information are preserved robustly and accurately. Experimental results show that the proposed decolorization algorithms obtain a good performance among existing quality metrics for the decolorization. In addition, the proposed global decolorization algorithm is friendly to mobile devices with limited computational resource.

Brain tissue segmentation from magnetic resonance (MR) images is an importance task for clinical use. The segmentation process becomes more challenging in the presence of noise, grayscale inhomogeneity, and other image artifacts. In this paper, we propose a robust kernelized local information fuzzy C-means clustering algorithm (RKLIFCM). It incorporates local information into the segmentation process (both grayscale and spatial) for more homogeneous segmentation. In addition, the Gaussian radial basis kernel function is adopted as a distance metric to replace the standard Euclidean distance. The main advantages of the new algorithm are: efficient utilization of local grayscale and spatial information, robustness to noise, ability to preserve image details, free from any parameter initialization, and with high speed as it runs on image histogram. We compared the proposed algorithm with 7 soft clustering algorithms that run on both image histogram and image pixels to segment brain MR images. Experimental results demonstrate that the proposed RKLIFCM algorithm is able to overcome the influence of noise and achieve higher segmentation accuracy with low computational complexity.

Many recent applications require text segmentation for born-digital compound images. To this end, we propose a coarse-to-fine framework for segmenting texts of arbitrary scales and orientations in born-digital compound images. In the coarse stage, the local image activity measure is designed based upon the variation distribution of characters, to highlight the difference between textual and pictorial regions. This stage outputs a coarse textual layer including textual regions as well as a few pictorial regions with high activity. In the fine stage, a textual connected component (TCC) based refinement is proposed to eliminate the survived pictorial regions. In particular, a scale and orientation invariant grouping algorithm is proposed to adaptively generate TCCs with uniform statistical features. The minimum average distance and morphological operations are employed to assist the formation of candidate TCCs. Then, three string-level features (i.e., shapeness, color similarity, and mean activity level) are designed to distinguish the true TCCs from the false positive ones that are formed by connecting the high activity pictorial components. Extensive experiments show that the proposed framework can segment textual regions precisely from born-digital compound images, while preserving the integrity of texts with varied scales and orientations, and avoiding over-connection of textual regions.

It is known that local filtering-based edge preserving smoothing techniques suffer from halo artifacts. In this paper, a weighted guided image filter (WGIF) is introduced by incorporating an edge-aware weighting into an existing guided image filter (GIF) to address the problem. The WGIF inherits advantages of both global and local smoothing filters in the sense that: 1) the complexity of the WGIF is O(N) for an image with N pixels, which is same as the GIF and 2) the WGIF can avoid halo artifacts like the existing global smoothing filters. The WGIF is applied for single image detail enhancement, single image haze removal, and fusion of differently exposed images. Experimental results show that the resultant algorithms produce images with better visual quality and at the same time halo artifacts can be reduced/avoided from appearing in the final images with negligible increment on running times.

In this paper, we introduce an exposure fusion scheme for differently exposed images with moving objects. The proposed scheme comprises a ghost removal algorithm in a low dynamic range domain and a selectively detail-enhanced exposure fusion algorithm. The proposed ghost removal algorithm includes a bidirectional normalization-based method for the detection of nonconsistent pixels and a two-round hybrid method for the correction of nonconsistent pixels. Our detail-enhanced exposure fusion algorithm includes a content adaptive bilateral filter, which extracts fine details from all the corrected images simultaneously in gradient domain. The final image is synthesized by selectively adding the extracted fine details to an intermediate image that is generated by fusing all the corrected images via an existing multiscale algorithm. The proposed exposure fusion algorithm allows fine details to be exaggerated while existing exposure fusion algorithms do not provide such an option. The proposed scheme usually outperforms existing exposure fusion schemes when there are moving objects in real scenes. In addition, the proposed ghost removal algorithm is simpler than existing ghost removal algorithms and is suitable for mobile devices with limited computational resource.

It is very challenging to synthesize a high dynamic range (HDR) image from multiple differently exposed low dynamic range images when there are moving objects in the images. This is due to the fact that the moving objects will cause ghosting artifacts to appear in the synthesized HDR image. To prevent such artifacts, a patching algorithm is required to correct motion regions such that all the moving objects are synchronized in the differently exposed images. In this paper, a new optimization problem is formulated to correct the motion regions of the multiple differently exposed images by considering both spatial and temporal consistencies. The resultant scheme is a hybrid patching scheme composed of a correction method which is an intensity mapping function at pixel level, and a hole-filling method that uses block-level template matching. The proposed patching scheme is not only robust to large intensity changes in these input images, but also at regions that are over- or underexposed. Experimental results show that the proposed method is able to prevent ghosting artifacts from appearing in the final synthesized HDR image.

In this paper, an efficient method for high-speed face recognition based on the discrete cosine transform (DCT), the Fisher's linear discriminant (FLD) and radial basis function (RBF) neural networks is presented. First, the dimensionality of the original face image is reduced by using the DCT and the large area illumination variations are alleviated by discarding the first few low-frequency DCT coefficients. Next, the truncated DCT coefficient vectors are clustered using the proposed clustering algorithm. This process makes the subsequent FLD more efficient. After implementing the FLD, the most discriminating and invariant facial features are maintained and the training samples are clustered well. As a consequence, further parameter estimation for the RBF neural networks is fulfilled easily which facilitates fast training in the RBF neural networks. Simulation results show that the proposed system achieves excellent performance with high training and recognition speed, high recognition rate as well as very good illumination robustness.

A general and efficient design approach using a radial basis function (RBF) neural classifier to cope with small training sets of high dimension, which is a problem frequently encountered in face recognition, is presented. In order to avoid overfitting and reduce the computational burden, face features are first extracted by the principal component analysis (PCA) method. Then, the resulting features are further processed by the Fisher's linear discriminant (FLD) technique to acquire lower-dimensional discriminant patterns. A novel paradigm is proposed whereby data information is encapsulated in determining the structure and initial parameters of the RBF neural classifier before learning takes place. A hybrid learning algorithm is used to train the RBF neural networks so that the dimension of the search space is drastically reduced in the gradient paradigm. Simulation results conducted on the ORL database show that the system achieves excellent performance both in terms of error rates of classification and learning efficiency.