Search

• We would like to submit the manuscript entitled "High-order relational generative adversarial network for video super-resolution", which we wish to be considered as a research article for publication in "Pattern Recognition". Because our contributed research covers the area" Image processing and representation", this manuscript belongs in this journal. No conflict of interest exits in the submission of this manuscript, and manuscript is approved by all authors for publication. • Both the instability of motion compensation and feature aggregation often undermine the SR results of videos. Only the global spatio-temporal dependencies of same scale features are considered as one-order relations, with difficulty in handling the complex motions and context variations. To deal with these issues, we argue that high-order relations should be modelled and applied in the generative strategies to exploit feature alignments and long-range information fusion. In fact, the high-order relations mean that the dependencies of features among pixel positions are mined not only from global feature representations but also through local patch interactions. Moreover, the stronger relations can be further revealed across different scales, by which the aggregation weight matrices with high-order statistics are adaptively determined. The videos are regarded as the complementary fusion of motion and context. Thus, simultaneously capturing the underlying relations of the motion and context is the key to video SR. In this paper, we propose a high-order relational generative adversarial network (HOR-GAN) for accurate video SR, which has strong capacity of alleviating the erroneous motion compensation and merging the useful information among the consecutive frames. • In summary, we highlight the main contributions of this paper as follows: We propose a HOR-GAN framework for highly accurate and realistic video SR. By exploiting high-order relations of feature patches under the construction of the pyramid graphs, the proposed HOR-GAN works well to produce a better feature alignment and fuse more spatio-temporal information. We adopt the dual discriminators to provide spatially coherent feedback to the generator, thus making the generator focus more on fine-grained features. The effectiveness of the proposed method has been justified through the extensive experiments. • We design a motion-aware relation module to accurately align neighboring frames with the reference ones. In this module, a patch-wise matching strategy is first used to build cross-scale correspondences of similar patches. The graph attention layers then adaptively aggregate the local patch features to further decrease the alignment error. We develop a context-aware relation module to make full use of high-order dependencies among all warped feature patches. We introduce the multi-scale graph convolution layers to mine contextual interaction relations for better fusing the spatio-temporal features, in which the position information is encoded to augment the detail recovery. Finally, each pixel is enhanced via a global self-attention. • We deeply appreciate your consideration of our manuscript, and we look forward to receiving comments from the reviewers. If you have any queries, please don't hesitate to contact me. Yours sincerely, Rui Chen Video super-resolution can reconstruct a sequence of high-resolution frames with temporally consistent contents from their corresponding low-resolution sequences. The key challenge for this task is how to effectively utilize both inter-frame temporal relations and intra-frame spatial relations. The existing methods for super-resolving the videos commonly estimate optical flows to align the features of multiple frames based on temporal correlations. However, motion estimation is often error-prone and hence largely hinders the recovery of plausible details. Moreover, high-order contextual dependencies in the feature space are rarely exploited for further enhancing the spatio-temporal information fusion. To this end, we propose a novel generative adversarial network to super-resolve low-resolution videos, which makes full use of patch embeddings and is effective in exploring high-order spatio-temporal relations of the feature patches. Specifically, a motion-aware relation module is designed to handle the alignment between neighboring frames and reference ones. Depending on a patch-matching strategy for adaptive selection of multiple most similar patches, the cross-scale graph is constructed to reliably aggregate these patches using a feature pyramid. Based on the structure of multi-scale graph, a context-aware relation module is developed to capture high-order dependencies among resulting warped patches for better leveraging long-range complementary contexts. To further enhance reconstruction ability, the temporal position information of video sequences is also encoded into this module. Dual discriminators with cycle consistent constraints are adopted to provide more informative feedback to the generator while maintaining the global coherence. Extensive experiments have demonstrated the effectiveness of the proposed method in terms of quantitative and qualitative evaluation metrics. [ABSTRACT FROM AUTHOR]

The convergent methods available for convolutional dictionary learning (CDL) are the proximal gradient method (PGM) and the inertial proximal gradient method (IPGM). However, it is not trivial and heuristic for IPGM to set the step size and inertia parameter, and IPGM produces local minima and oscillating solutions. To address these issues, in this paper, we introduce a dry friction, which has an oscillation-alleviating property. Specifically, the proposed IPGM with dry friction (IPGM-DF) generates the composite proximal mappings, whose construction and optimization solver are two challenges. An auxiliary function is designed to construct the composite proximal mappings, whose optimization problem is solved by the alternating direction method of multipliers. Fortunately, IPGM-DF obtains the formulas of the step size and inertia parameter. The finite convergence of IPGM-DF is proved. Experimental results of image reconstruction, separation, and fusion demonstrate the superiority of IPGM-DF over IPGM and the state-of-the-art methods. For image reconstruction, the objective function value of IPGM-DF is reduced by about 38.708% than that of IPGM. Throughout alleviating oscillations, IPGM-DF obtains a much lower value of the objective function than IPGM, which indicates that IPGM-DF jumps out of the local minima of IPGM. In addition, the average PSNR of IPGM-DF is about 3.5 dB larger than that of IPGM and the state-of-the-art methods. The code is available. • An inertial proximal gradient method with dry friction (IPGM-DF) is proposed for convolutional dictionary learning. • IPGM-DF can automatically calculate the step size and inertia parameter. • IPGM-DF can alleviate oscillations and jump out of the local minimum of the proximal gradient method. • IPGM-DF is applied to the image reconstruction, separation, and fusion task. • The experimental results of image processing demonstrate that IPGM-DF is superior to SOTA methods. [ABSTRACT FROM AUTHOR]

Highlights • This paper proposes a unified framework to perform classification and low-level modeling jointly. • Robustness is improved by considering a possibly badly labeled training set. • The proposed model allows a very rich interpretation of the modeled data structure. • Performance is assessed on synthetic and real data in the specific context of hyperspectral image interpretation. • The proposed model is generic enough to incorporate any kind of low-level modeling. Abstract Within a supervised classification framework, labeled data are used to learn classifier parameters. Prior to that, it is generally required to perform dimensionality reduction via feature extraction. These preprocessing steps have motivated numerous research works aiming at recovering latent variables in an unsupervised context. This paper proposes a unified framework to perform classification and low-level modeling jointly. The main objective is to use the estimated latent variables as features for classification and to incorporate simultaneously supervised information to help latent variable extraction. The proposed hierarchical Bayesian model is divided into three stages: a first low-level modeling stage to estimate latent variables, a second stage clustering these features into statistically homogeneous groups and a last classification stage exploiting the (possibly badly) labeled data. Performance of the model is assessed in the specific context of hyperspectral image interpretation, unifying two standard analysis techniques, namely unmixing and classification. [ABSTRACT FROM AUTHOR]

Highlights • A new attention based CNN-LSTM network is proposed for script identification. • It is based on local and global feature extraction and dynamically weighting them. • Attention used twice to give priority to significant features and to focus on more relevant patches. • The method has been tested on SIW-13, CVSI-2015, ICDAR-17 and MLe2e datasets. Abstract Script identification plays a significant role in analysing documents and videos. In this paper, we focus on the problem of script identification in scene text images and video scripts. Because of low image quality, complex background and similar layout of characters shared by some scripts like Greek, Latin, etc., text recognition in those cases become challenging. In this paper, we propose a novel method that involves extraction of local and global features using CNN-LSTM framework and weighting them dynamically for script identification. First, we convert the images into patches and feed them into a CNN-LSTM framework. Attention-based patch weights are calculated applying softmax layer after LSTM. Next, we do patch-wise multiplication of these weights with corresponding CNN to yield local features. Global features are also extracted from last cell state of LSTM. We employ a fusion technique which dynamically weights the local and global features for an individual patch. Experiments have been done in four public script identification datasets: SIW-13, CVSI2015, ICDAR-17 and MLe2e. The proposed framework achieves superior results in comparison to conventional methods. [ABSTRACT FROM AUTHOR]

Lossless image compression is an essential aspect of image processing, particularly in many fields that require high information fidelity. In recent years, learned lossless image compression methods have shown promising results. However, many of these methods do not make optimal use of available information, leading to sub-optimal performance. This paper proposes a multi-prior entropy model for lossless image compression, which effectively leverages available information to achieve better compression performance. The proposed multi-prior comprises a cross-channel prior, hybrid local context, and hyperprior, allowing it to effectively utilize all available information. To remove redundancy across color channels, the original image is first losslessly transformed into YUV color space. The network then learns priors from the original image, the prior-coding channels, and the local context, which are fused to form the multi-prior used for GMM parameters estimation. Moreover, to capture the features of different images, a hybrid local context is abstracted using different kernel sizes of mask convolutions in a local context. The experimental results on several datasets demonstrate that our algorithm outperforms several existing learning-based image compression methods and traditional methods, such as JPEG2000, WebP, and FLIF. • We focus on how to get the utmost feasible information of images for entropy coding and propose a hybrid-context-based multi-prior based lossless image compression. This algorithm encodes the original image in a channel-by-channel manner. • We adopt the YUV color space for coding. Experimental results show that transforming the original image into YUV space can improve the lossless compression performance. • To accommodate the contextual information of different regions of the image, we use a hybrid multi-scale model to explore the local contextual information. [ABSTRACT FROM AUTHOR]

Poor light conditions constrain the high pursuit of clarity and visible quality of photography especially smartphone devices. Admittedly, existing specific image processing methods, whether super-resolution methods or low-light enhancement methods, can hardly simultaneously enhance the resolution and brightness of low-light images at the same time. This paper dedicates a specialized enhancer with a dual-path modulated-interactive structure to recover high-quality sharp images in conditions of near absence of light, dubbed CollaBA , which learns the direct mapping from low-resolution dark-light images to their high-resolution normal sharp version. Specifically, we construct the generative modulation prior, serving as illuminance attention information, to regulate the exposure level of the neighborhood range. In addition, we construct an interactive degradation removal branch that progressively embeds the generated intrinsic prior to recover high-frequency detail and contrast at the feature level. We also introduce a multi-substrate up-scaler to integrate multi-scale sampling features, effectively addressing artifact-related problems. Rather than adopting the naive time-consuming learning strategy, we design a novel bi-level implicit adversarial learning mechanism as our fast training strategy. Extensive experiments on benchmark datasets — demonstrate our model's wide-ranging applicability in various ultra-low-light scenarios, across 8 key performance metrics with significant improvements, notably achieving a 35.8% improvement in LPIPS and a 23.1% increase in RMSE. The code will be available at https://github.com/moriyaya/CollaBA. • We present CollaBA, a specialized dual-path modulated-aggregated enhancer, as a fresh approach to tackling the intricate challenge of collaborating amplification and brightening images taken in extremely low-light conditions. • Our CollaBA gains remarkable performance by imposing generative modulation priors to guide exposure regulation, progressively integrating them into the multi-scale degradation removal branch through spatial feature transformation. • Instead of naive time-consuming adversarial learning strategy, a novel bi-level implicit adversarial learning mechanism is designed, effectively improving the stability of training and the quality of visual perception. • Extensive experiments were conducted to thoroughly validate that our method surpasses existing state-of-the-art approaches on real-world benchmark datasets, particularly in extremely low-light conditions, achieving a 35.8% improvement in LPIPS and a 23.1% increase in RMSE. [ABSTRACT FROM AUTHOR]

Local ternary pattern (LTP) is a simple, yet high-discriminative texture feature model. In order to further promote the performance of LTP in noisy texture classification, this paper proposes a multi-scale median LTP (MLTP) framework. It firstly designs a non-overlapped median sampling scheme to resist against noise, and then re-defines central descriptor, radial descriptor and magnitude descriptor for MLTP. Moreover, it also proposes the pattern mapping solution named rotation-invariant uniform three (riu3) so as to project original MLTP codes from high-dimensional patterns to low-dimensional ones. In our multi-scale algorithm, every image is sampled and encoded separately by three MLTP descriptors, and then original patterns are mapped by a pre-stored riu3 pattern lookup table. Finally, the frequency histogram of joint distribution on three mapped MLTP code images is computed to generate the feature vector of given sampling scale. The different vectors under different scales are concatenated together to produce the objective vectors for training classifier or texture classification. The available classifiers for our algorithm include nearest neighbor classifier (NNC), and support vector machine (SVM), etc. The experiments on five publicly-available texture databases show that the multi-scale MLTP method proposed in this paper could obtain an average accuracy of 99.48% on dataset OTC12, 96.97% on UIUC, 98.71% on KTH-TIPS2b, 98.55% on Brodatz and 97.49% on ALOT. Compared with other state-of-the-art approaches, under the low-intensity conditions of Gaussian, Salt & Pepper and random pixel corruption noise, multi-scale MLTP + SVM could still keep higher accuracy and have better noise robustness than others. [ABSTRACT FROM AUTHOR]

In this paper we further develop the recent concept of multi-component shapes, which is applicable to image processing and image analysis tasks. The domain of multi-component shapes is very diverse and includes shapes that correspond to a group of objects that act together (e.g. a fish shoal), natural components of a segmented object (e.g. cells in embryonic tissues), a set of shapes corresponding to the same object appearing at different times (e.g. human gait in an image sequence), and many more. So far, there are few methods for numerically evaluating multi-component shapes. In this paper we introduce one such method: a disconnectedness measure, that naturally corresponds to multi-component shapes, and has no analogue in single-component shape measures. The new measure depends on the number of shape components, the whole shape but also the shape of its components, on the relative size of the shape’s components and their mutual position. All these are natural requirements for a “disconnectedness” multi-component shape measure. In addition, the new measure is invariant with respect to translation, rotation and scaling transformations. The measure is simple and fast to compute. The disconnectedness measure introduced here is a generic image analysis tool. It has not been developed for a specific application. As such, it can be applied to a variety of applications. Several of them are provided in the paper, as well as synthetic examples that support a better understanding of the behavior of the new measure. [ABSTRACT FROM AUTHOR]

The term “iris” refers to the highly textured annular portion of the human eye that is externally visible. An iris recognition system exploits the richness of these textural patterns to distinguish individuals. Iris recognition systems are being used in a number of human recognition applications such as access control, national ID schemes, border control, etc. To capture the rich textural information of the iris pattern regardless of the eye color, traditional iris recognition systems utilize near-infrared (NIR) sensors to acquire images of the iris. This, however, restricts the iris image acquisition distance to close quarters (less than 1 m). Over the last several years, there have been numerous attempts to design and implement iris recognition systems that operate at longer standoff distances ranging from 1 m to 60 m. Such long range iris acquisition and recognition systems can provide high user convenience and improved throughput. This paper reviews the state-of-the-art design and implementation of iris-recognition-at-a-distance (IAAD) systems. In this regard, the design of such a system from both the image acquisition (hardware) and image processing (algorithms) perspectives are presented. The major contributions of this paper include: (1) discussing the significance and applications of IAAD systems in the context of human recognition, (2) providing a review of existing IAAD systems, (3) presenting a complete solution to the design problem of an IAAD system, from both hardware and algorithmic perspectives, (4) discussing the use of additional ocular information, along with iris, for improving IAAD accuracy, and (5) discussing the current research challenges and providing recommendations for future research in IAAD. [ABSTRACT FROM AUTHOR]

Improving the safety of transportation systems attracts lots of attention. Researchers introduced their methods to detect and analyze the vehicle and the pedestrian on the road to accomplish this goal. However, the bicycle is also a significant factor of the safety on a road. In this paper, a bicycle detector for side-view image is proposed based on the observation that a bicycle consists of two wheels in the form of ellipse shapes and a frame in the form of two triangles. Through the proposed triangle detection algorithm, the bicycle model and the geometric constraints on the relationship between the triangles and ellipses, the computation is fast according to the sample implementation and the evaluation of the reduced data amount. Besides, the training process is unnecessary and only single image is required for our algorithm. The experimental results are also given in this paper to show the practicability and the performance of the proposed bicycle model and bicycle detection algorithm. [ABSTRACT FROM AUTHOR]

Image registration is one of the major technologies of the medical image processing and analysis. According to the nature of the transformation, image registration can be categorized into two main classes: Rigid Registration and Non-rigid Registration. In this paper, a novel learning-based dissimilarity metric and its approximation are proposed and respectively adopted to the rigid and non-rigid registration of medical images. This novel metric utilizes Bhattacharyya Distances to measure the dissimilarity of the testing image pairs by incorporating the expected intensity distributions which is learnt from the registered training image pairs. In order to test the robustness and accuracy of the proposed metric in the multi-modal rigid registration, one thousand eight hundred randomized CT–T1 registrations were performed and evaluated by the Retrospective Image Registration Evaluation (RIRE) project. For the non-rigid registration, the approximation of the proposed dissimilarity metric is adopted with the Markov Random Field (MRF) modeled non-rigid registration approach. Non-rigid registration experiments of 2D synthetic images with artificial deformation fields and 3D images from the Simulated Brain Database were performed and evaluated respectively in this paper. The experimental results of both rigid and non-rigid medical image registrations demonstrated that the proposed dissimilarity metric outperforms the state-of-the-art approach (Mutual Information) and the close related approach (Kullback–Leibler Distances). [ABSTRACT FROM AUTHOR]

With a wide range of shadows in many collected images, shadow removal has aroused increasing attention since uncontaminated images are of vital importance for many computer vision tasks. Current methods consider the same convolution operations for both shadow and non-shadow regions while ignoring the large gap between the color mappings for the shadow region and the non-shadow region, leading to poor quality of reconstructed images and a heavy computation burden. To solve this problem, this paper introduces a novel plug-and-play Shadow-Aware Dynamic Convolution (SADC) module to decouple the interdependence between the shadow region and the non-shadow region. Inspired by the fact that the color mapping of the non-shadow region is easier to learn, our SADC processes the non-shadow region with a lightweight convolution module in a computationally cheap manner and recovers the shadow region with a more complicated convolution module to ensure the quality of image reconstruction. Given that the non-shadow region often contains more background color information, we further develop a novel intra-convolution distillation loss to strengthen the information flow from the non-shadow region to the shadow region. Extensive experiments on the ISTD and SRD datasets show our method achieves better performance in shadow removal over many state-of-the-art methods. Codes have been made available at https://github.com/xuyimin0926/SADC. • We propose a plug-and-play shadow-aware dynamic convolution module to handle the shadow regions and non-shadow regions separately. • We design a specific intra-convolution distillation loss to enhance the shadow removal performance based on the non-shadow convolution results. • Extensive experimental results demonstrate the superior performance of the proposed SADC module. existing state-of-the-art methods. [ABSTRACT FROM AUTHOR]

• We propose a methodology for the hierarchical representation of multimodal images. • The hierarchical representation relies on the novel concept of braids of partitions. • We introduce a braid-based method to conduct multimodal image segmentation. • The segmentation method is based on an energy minimization procedure. • The proposed approach is validated on different multimodal images. Hierarchical data representations are powerful tools to analyze images and have found numerous applications in image processing. When it comes to multimodal images however, the fusion of multiple hierarchies remains an open question. Recently, the concept of braids of partitions has been proposed as a theoretical tool and possible solution to this issue. In this paper, we demonstrate the relevance of the braid structure for the hierarchical representation of multimodal images. We first propose a fully operable procedure to build a braid of partitions from two hierarchical representations. We then derive a framework for multimodal image segmentation, relying on an energetic minimization scheme conducted on the braid structure. The proposed approach is investigated on different multimodal images scenarios, and the obtained results confirm its ability to efficiently handle the multimodal information to produce more accurate segmentation outputs. [ABSTRACT FROM AUTHOR]

• We proposed a novel NAS method to model the inter-layer dependency. With learnable transition matrices and attention vectors, the weights of inner edges can be derived from the weights of their predecessor edges sequentially rather than being learned independently. • We utilized a new transition-induced iterative edge pruning strategy that can progressively prune edges in the context of the architecture transition paradigm. • Extensive experiments on five benchmarks demonstrate that the proposed ITNAS can search better architectures than other methods, validating that modeling the inter-layer dependency matters in architecture search. Neural architecture search (NAS) attracts much research attention, contributing to its ability to identify better architectures than manually-designed ones. Recently, differential neural architecture search methods have been widely used due to their impressive effectiveness and performance. They represent the network architecture as a repetitive proxy-directed acyclic graph (DAG) and optimize the network weights and architecture weights alternatively in a differential manner. However, existing methods model the architecture weights on each edge (i.e., a layer in the network) as statistically independent variables, ignoring the dependency between edges in DAG induced by their directed topological connections. In this paper, we make the first attempt to investigate such a dependency or relationship by proposing a novel inter-layer transition NAS method. It casts the architecture optimization into a sequential decision process where the dependency between the architecture weights of connected edges is explicitly modeled. Specifically, edges are divided into inner and outer groups according to whether or not their predecessor edges are in the same cell. While the architecture weights of outer edges are optimized independently, those of inner edges are derived sequentially based on the architecture weights of their predecessor edges and the learnable transition matrices in an attentive probability transition manner. Experiments on five benchmark classification datasets, four searching spaces, and NAS-Bench-201 confirm the value of modeling inter-layer dependency and demonstrate the proposed method outperforms other methods. [ABSTRACT FROM AUTHOR]

The present paper focuses on non-linear pattern matching based on the Logarithmic Image Processing (LIP) Model. Our contribution consists first of using the scalar multiplication defined in the LIP context to extend the little-known Asplünd׳s metric to gray level images. Such a metric is explainable as a novel technique of double-sided image probing and presents the decisive advantage of being physically justified in the field of transmitted light acquisition. Moreover, thanks to the consistency of the LIP context with human vision, Asplünd׳s metric is also applicable to images acquired in reflected light: in fact, plenty of image processing algorithms aim at extracting information as a human eye would do. Finally, the proposed approach is particularly efficient in the presence of lighting variations or lighting drift. In the paper, we also suggest a solution to overcome the main drawback of probing techniques, which resides in a high sensitivity to noise. Various examples are presented to highlight the efficiency of the method. [ABSTRACT FROM AUTHOR]

Abstract: This paper proposes a new approach for multi-object 3D scene modeling. Scenes with multiple objects are characterized by object occlusions under several views, complex illumination conditions due to multiple reflections and shadows, as well as a variety of object shapes and surface properties. These factors raise huge challenges when attempting to model real 3D multi-object scene by using existing approaches which are designed mainly for single object modeling. The proposed method relies on the initialization provided by a rough 3D model of the scene estimated from the given set of multi-view images. The contributions described in this paper consists of two new methods for identifying and correcting errors in the reconstructed 3D scene. The first approach corrects the location of 3D patches from the scene after detecting the disparity between pairs of their projections into images. The second approach is called shape-from-contours and identifies discrepancies between projections of 3D objects and their corresponding contours, segmented from images. Both unsupervised and supervised segmentations are used to define the contours of objects. [Copyright &y& Elsevier]

Abstract: The paper studies a 3D fingerprint reconstruction technique based on multi-view touchless fingerprint images. This technique offers a solution for 3D fingerprint image generation and application when only multi-view 2D images are available. However, the difficulties and stresses of 3D fingerprint reconstruction are the establishment of feature correspondences based on 2D touchless fingerprint images and the estimation of the finger shape model. In this paper, several popular used features, such as scale invariant feature transformation (SIFT) feature, ridge feature and minutiae, are employed for correspondences establishment. To extract these fingerprint features accurately, an improved fingerprint enhancement method has been proposed by polishing orientation and ridge frequency maps according to the characteristics of 2D touchless fingerprint images. Therefore, correspondences can be established by adopting hierarchical fingerprint matching approaches. Through an analysis of 440 3D point cloud finger data (220 fingers, 2 pictures each) collected by a 3D scanning technique, i.e., the structured light illumination (SLI) method, the finger shape model is estimated. It is found that the binary quadratic function is more suitable for the finger shape model than the other mixed model tested in this paper. In our experiments, the reconstruction accuracy is illustrated by constructing a cylinder. Furthermore, results obtained from different fingerprint feature correspondences are analyzed and compared to show which features are more suitable for 3D fingerprint images generation. [Copyright &y& Elsevier]

Abstract: Rules of physics in many real-life problems force some constraints to be satisfied. This paper deals with nonlinear pattern recognition under non-negativity constraints. While kernel principal component analysis can be applied for feature extraction or data denoising, in a feature space associated to the considered kernel function, a pre-image technique is required to go back to the input space, e.g., representing a feature in the space of input signals. The main purpose of this paper is to study a constrained pre-image problem with non-negativity constraints. We provide new theoretical results on the pre-image problem, including the weighted combination form of the pre-image, and demonstrate sufficient conditions for the convexity of the problem. The constrained problem is considered with the non-negativity, either on the pre-image itself or on the weights. We propose a simple iterative scheme to incorporate both constraints. A fortuitous side-effect of our method is the sparsity in the representation, a property investigated in this paper. Experimental results are conducted on artificial and real datasets, where many properties are investigated including the sparsity property, and compared to other methods from the literature. The relevance of the proposed method is demonstrated with experimentations on artificial data and on two types of real datasets in signal and image processing. [Copyright &y& Elsevier]

Abstract: Ping et al. [Z. Ping, H. Ren, J. Zou, Y. Sheng, and W. Bo, Generic orthogonal moments: Jacobi–Fourier moments for invariant image description, Pattern Recognition 40 (4) (2007) 1245–1254] made a landmark contribution to the theory of two-dimensional orthogonal moments confined to the unit disk by unifying the radial kernels of existing polynomial-based circular orthogonal moments under the roof of shifted Jacobi polynomials. However, the work contains some errata that result mainly from the confusion between the two slightly different definitions of shifted Jacobi polynomials in the literature. Taking into account the great importance and the high impact of the work in the pattern recognition community, this paper points out the confusing points, corrects the errors, and gives some other relevant comments. The corrections developed in this paper are illustrated by some experimental evidence. [Copyright &y& Elsevier]

Abstract: Accurate local region description is a keypoint in many applications and has been the topic of lots of recent papers. Starting from the very accurate SIFT, most of the approaches exploit the local gradient information that suffers from several drawbacks. First it is unstable in case of severe geometry distortions, second it cannot be easily summarized in a compact way and third it is not designed to account vectorial color information. In this paper, we propose an alternative by designing compact descriptors that account both the colors present in the region and their spatial distribution. Each pixel being characterized by five coordinates, two in the image space and three in the color space, we try to evaluate affine transforms that allow to go from the spatial coordinates to the color coordinates and inversely. Obviously such kind of transform does not exist but we show that after applying it to the original coordinates, the resulted positions are both discriminative and invariant to many acquisition conditions. Hence, depending on the original space (image or color) and the destination space (color or image), we design different complementary descriptors. Their discriminative power and invariance properties are assessed and compared with the best color descriptors in the context of region matching and object classification. [Copyright &y& Elsevier]

Abstract: In computer vision, camera calibration is a necessary process when the retrieval of information such as angles and distances is required. This paper addresses the multi-camera calibration problem with a single dimension calibration pattern under general motions. Currently, the known algorithms for solving this problem are based on the estimation of vanishing points. However, this estimate is very susceptible to noise, making the methods unsuitable for practical applications. Instead, this paper presents a new calibration algorithm, where the cameras are divided into binocular sets. The fundamental matrix of each binocular set is then estimated, allowing to perform a projective calibration of each camera. Then, the calibration is updated for the Euclidean space, ending the process. The calibration is possible without imposing any restrictions on the movement of the pattern and without any prior information about the cameras or motion. Experiments on synthetic and real images validate the new method and show that its accuracy makes it suitable also for practical applications. [Copyright &y& Elsevier]

Abstract: The problem of image reconstruction from its statistical moments is particularly interesting to researchers in the domain of image processing and pattern recognition. Compared to geometric moments, the orthogonal moments offer the ability to recover much more easily the image due to their orthogonality, which allows reducing greatly the complexity of computation in the phase of reconstruction. Since the 1980s, various orthogonal moments, such as Legendre moments, Zernike moments and discrete Tchebichef moments have been introduced early or late to image reconstruction. In this paper, another set of orthonormal moments, the Gaussian–Hermite moments, based on Hermite polynomials modulated by a Gaussian envelope, is proposed to be used for image reconstruction. Especially, the paper''s focus is on the determination of the optimal scale parameter and the improvement of the reconstruction result by a post-processing which make Gaussian–Hermite moments be useful and comparable with other moments for image reconstruction. The algorithms for computing the values of the basis functions, moment computation and image reconstruction are also given in the paper, as well as a brief discussion on the computational complexity. The experimental results and error analysis by comparison with other moments show a good performance of this new approach. [Copyright &y& Elsevier]

Highlights • A new color image descriptor, namely multi-channel orthogonal radial substituted Chebyshev moments (MORSCMs), is introduced. • A new method for highly accurate, fast and stable computation of MORSCMs is proposed. • Rotation, Scaling and Translation invariants of the MORSCMs are derived. • The proposed moments archives high recognition rates in presence of geometric and noise attacks. • Comparison with existing methods clearly show the superiority of the new MORSCMs. Abstract Orthogonal moments and their invariants to similarity transformations for monochrome and gray-scale images are widely used in many pattern recognition and image processing applications. Quaternion orthogonal moments are used with color images. Recently, the multi-channel framework is proposed as a successful alternative of the quaternion orthogonal moments in representation and recognition of the color images. In this paper, a new set of multi-channel orthogonal moments and their invariants to rotation, scaling and translation (RST) is proposed for color image representation and recognition. The proposed multi-channel moments are based on the orthogonal radial substituted Chebyshev functions. The multi-channel orthogonal radial substituted Chebyshev moments (MORSCMs) are defined in polar coordinates over a unit circle. An accurate kernel-based method is utilized for accurate computation of the MORSCMs. A series of experiments is performed to validate this new set of multi-channel moments and compare its performance with the existing quaternion and multi-channel orthogonal moments. The obtained results ensure the superiority of the proposed MORSCMs over all existing moments in representation and recognition of the color images. [ABSTRACT FROM AUTHOR]

Highlights • We proposed a weighted heterogeneous distance metric (WHDM). • We presented WHDM and Dempster–Shafer theory based k NN algorithm. • We proposed a multimodal perturbation method (RRSB) for k NN ensemble. • The effectiveness of our algorithms was shown on multiple UCI data sets and a KDD data set. Abstract In this paper, to address the issue that ensembling k -nearest neighbor (k NN) classifiers with resampling approaches cannot generate component classifiers with a large diversity, we consider ensembling k NN through a multimodal perturbation-based method. Since k NN is sensitive to the input attributes, we propose a weighted heterogeneous distance Metric (WHDM). By using a WHDM and evidence theory, a progressive k NN classifier is developed. Based on a progressive k NN, the random subspace method, attribute reduction, and Bagging, a novel algorithm termed RRSB (reduced random subspace-based Bagging) is proposed for construct ensemble classifier, which can increase the diversity of component classifiers without damaging the accuracy of the component classifiers. In detail, RRSB adopts the perturbation on the learning parameter with a weighted heterogeneous distance metric, the perturbation on the input space with random subspace and attribute reduction, the perturbation on the training data with Bagging, and the perturbation on the output target of k neighbors with evidence theory. In the experimental stage, the value of k , the different perturbations on RRSB and the ensemble size are analyzed. In addition, RRSB is compared with other multimodal perturbation-based ensemble algorithms on multiple UCI data sets and a KDD data set. The results from the experiments demonstrate the effectiveness of RRSB for k NN ensembling. [ABSTRACT FROM AUTHOR]

Highlights • Real-time geometric fitting for cross sections of surface of revolution (SOR). • Real-time pose and structure recovery for SOR based on geometric fitting. • SOR shape reconstruction based on ellipse fitting. • Real-time 3D tracking of SOR with the real size of cross sections. • Geometric fitting for imaged parallel circles. Abstract This paper presents a novel ellipse fitting method to simultaneously estimate the Euclidean pose and structure of a surface of revolution (SOR) by minimizing the geometric reprojection error of the visible cross sections in image space. This geometric error function and its Jacobian matrix are explicitly derived to enable Levenberg-Marquardt (LM) optimization. With the obtained pose and structure, the Euclidean shape of a SOR can be reconstructed by generating the ellipse tangency to the apparent contour of the SOR. Given the real size of several visible cross sections, this approach can be extended to perform a real-time 3D tracking of the SOR. Additionally, this technique can be also generalized to fitting for imaged parallel circles. Sufficient experiments validate the accuracy and the real-time performance of the proposed method. [ABSTRACT FROM AUTHOR]

Highlights • A new dermoscopic segmentation method based on learning class-wise deep features. • A class-specific model helps balance the contribution of melanoma and non-melanoma data. • A step-wise refinement approach iteratively maximises the segmentation map agreement. • A probability based integration approach combines the relevant complementary segmentation results. • We achieve higher accuracy compared to the state-of-the-art methods on 3 datasets. Abstract The segmentation of abnormal regions on dermoscopic images is an important step for automated computer aided diagnosis (CAD) of skin lesions. Recent methods based on fully convolutional networks (FCN) have been very successful for dermoscopic image segmentation. However, they tend to overfit to the visual characteristics that are present in the dominant non-melanoma studies and therefore, perform poorly on the complex visual characteristics exhibited by melanoma studies, which usually consists of fuzzy boundaries and heterogeneous textures. In this paper, we propose a new method for automated skin lesion segmentation that overcomes these limitations via a novel deep class-specific learning approach which learns the important visual characteristics of the skin lesions of each individual class (melanoma vs. non-melanoma) on an individual basis. We also introduce a new probability-based, step-wise integration to combine complementary segmentation results derived from individual class-specific learning models. We achieved an average Dice coefficient of 85.66% on the ISBI 2017 Skin Lesion Challenge (SLC), 91.77% on the ISBI 2016 SLC and 92.10% on the PH2 datasets with corresponding Jaccard indices of 77.73%, 85.92% and 85.90%, respectively, for the same datasets. Our experiments on three well-established public benchmark datasets demonstrate that our method is more effective than other state-of-the-art methods for skin lesion segmentation. [ABSTRACT FROM AUTHOR]

Abstract: This paper gives an end-to-end overview of 3D video and free viewpoint video, which can be regarded as advanced functionalities that expand the capabilities of a 2D video. Free viewpoint video can be understood as the functionality to freely navigate within real world visual scenes, as it is known for instance from virtual worlds in computer graphics. 3D video shall be understood as the functionality that provides the user with a 3D depth impression of the observed scene, which is also known as stereo video. In that sense as functionalities, 3D video and free viewpoint video are not mutually exclusive but can very well be combined in a single system. Research in this area combines computer graphics, computer vision and visual communications. It spans the whole media processing chain from capture to display and the design of systems has to take all parts into account, which is outlined in different sections of this paper giving an end-to-end view and mapping of this broad area. The conclusion is that the necessary technology including standard media formats for 3D video and free viewpoint video is available or will be available in the future, and that there is a clear demand from industry and user for such advanced types of visual media. As a consequence we are witnessing these days how such technology enters our everyday life [Copyright &y& Elsevier]

Abstract: Human hand shape features extraction from image frame sequences is one of the key steps in human hand 2D/3D tracking system and human hand shape recognition system. In order to satisfy the need of human hand tracking in real time, a fast and accurate method for acquirement of edge features from human hand images without consideration of hand over face is put forward in this paper. The proposed approach is composed of two steps, the coarse location phase (CLP) and the refined location phase (RLP) from coarseness to refinement. In the phase of CLP, the hand contour is approximately described by a polygon with concave and convex, an approach to obtaining hand shape polygon using locating points and locating lines is meticulously discussed. Then, a coarse location (CL) algorithm for extraction of interested hand shape features, such as contour, fingertips, roots of fingers, joints and the intersection of knuckle on different fingers, is proposed. In the phase of RLP, a multi-scale approach is introduced into our study to refine the features obtained by the CL algorithm. By means of defining the response strength of different types of features, a refined location (RL) algorithm is proposed. The major contribution of this paper is that the novel detection operators for features of hand images are presented in the above two steps, which have been successfully applied to our 3D hand shape tracking system and 2D hand shape recognition system. A number of comparative studies with real images and online videos demonstrate that the proposed method can extract the three defined human hand image features with high accuracy and high speed. [ABSTRACT FROM AUTHOR]

Abstract: This paper concerns color image restoration aiming at objective quality improvement of compressed color images in general rather than merely artifact reduction. In compressed color images, colors are usually represented by luminance and chrominance components. Considering characteristics of human vision system, chrominance components are generally represented more coarsely than luminance component. To recover such chrominance components, we previously proposed a model-based chrominance restoration algorithm where color images are modeled by a Markov random field. This paper presents a color image restoration algorithm derived by the MAP estimation, where all components are totally estimated. Experimental results show that the proposed restoration algorithm is more effective than the previous one. [Copyright &y& Elsevier]

Low-textured image stitching remains a challenging problem. It is difficult to achieve good alignment of images and it is easy to break images structures are often broken due to insufficient and unreliable point correspondences. Moreover, because of the viewpoint variations between multiple images, the stitched images suffer from projective distortions. To solve these problems, this paper presents a line-guided local warping method with a global similarity constraint for image stitching. Line features which serve well for geometric descriptions and scene constraints, are employed to guide image stitching accurately. On one hand, the line features are integrated into a local warping model through a designed weight function. On the other hand, line features are adopted to impose strong geometric constraints, including line correspondence and line colinearity, to improve the stitching performance through mesh optimization. To mitigate projective distortions, we adopt a global similarity constraint, which is integrated with the projective warps via a designed weight strategy. This constraint causes the final warp to slowly change from a projective to a similarity transformation across the image. Finally, the images undergo a two-stage alignment scheme that provides accurate alignment and reduces projective distortion. We evaluate our method on a series of images and compare it with several other methods. The experimental results demonstrate that the proposed method provides a convincing stitching performance and that it outperforms other state-of-the-art methods. [ABSTRACT FROM AUTHOR]

In this work we describe a fast and stable algorithm for the computation of the orthogonal moments of an image. Indeed, orthogonal moments are characterized by a high discriminative power, but some of their possible formulations are characterized by a large computational complexity, which limits their real-time application. This paper describes in detail an approach based on recurrence relations, and proposes an optimized Matlab implementation of the corresponding computational procedure, aiming to solve the above limitations and put at the community’s disposal an efficient and easy to use software. In our experiments we evaluate the effectiveness of the recurrence formulation, as well as its performance for the reconstruction task, in comparison to the closed form representation, often used in the literature. The results show a sensible reduction in the computational complexity, together with a greater accuracy in reconstruction. In order to assess and compare the accuracy of the computed moments in texture analysis, we perform classification experiments on six well-known databases of texture images. Again, the recurrence formulation performs better in classification than the closed form representation. More importantly, if computed from the GLCM of the image using the proposed stable procedure, the orthogonal moments outperform in some situations some of the most diffused state-of-the-art descriptors for texture classification. [ABSTRACT FROM AUTHOR]

Traditional image clustering methods take a two-step approach, feature learning and clustering, sequentially. However, recent research results demonstrated that combining the separated phases in a unified framework and training them jointly can achieve a better performance. In this paper, we first introduce fully convolutional auto-encoders for image feature learning and then propose a unified clustering framework to learn image representations and cluster centers jointly based on a fully convolutional auto-encoder and soft k -means scores. At initial stages of the learning procedure, the representations extracted from the auto-encoder may not be very discriminative for latter clustering. We address this issue by adopting a boosted discriminative distribution, where high score assignments are highlighted and low score ones are de-emphasized. With the gradually boosted discrimination, clustering assignment scores are discriminated and cluster purities are enlarged. Experiments on several vision benchmark datasets show that our methods can achieve a state-of-the-art performance. [ABSTRACT FROM AUTHOR]

Due to the factors like rapidly fast motion, cluttered backgrounds, arbitrary object appearance variation and shape deformation, an effective target representation plays a key role in robust visual tracking. Existing methods often employ bounding boxes for target representations, which are easily polluted by noisy clutter backgrounds that may cause drifting problem when the target undergoes large-scale non-rigid or articulated motions. To address this issue, in this paper, motivated by the spatio-temporal nonlocality of target appearance reoccurrence in a video, we explore the nonlocal information to accurately represent and segment the target, yielding an object likelihood map to regularize a correlation filter (CF) for visual tracking. Specifically, given a set of tracked target bounding boxes, we first generate a set of superpixels to represent the foreground and background, and then update the appearance of each superpixel with its long-term spatio-temporally nonlocal counterparts. Then, with the updated appearances, we formulate a spatio-temporally graphical model comprised of the superpixel label consistency potentials. Afterwards, we generate segmentation by optimizing the graphical model via iteratively updating the appearance model and estimating the labels. Finally, with the segmentation mask, we obtain an object likelihood map that is employed to adaptively regularize the CF learning by suppressing the clutter background noises while making full use of the long-term stable target appearance information. Extensive evaluations on the OTB50, SegTrack, Youtube-Objects datasets demonstrate the effectiveness of the proposed method, which performs favorably against some state-of-art methods. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents a novel face recognition method by means of fusing color, local spatial and global frequency information. Specifically, the proposed method fuses the multiple features derived from a hybrid color space, the Gabor image representation, the local binary patterns (LBP), and the discrete cosine transform (DCT) of the input image. The novelty of this paper is threefold. First, a hybrid color space, the RC r Q color space, is constructed by combining the R component image of the RGB color space and the chromatic component images, C r and Q, of the YC b C r and YIQ color spaces, respectively. The RC r Q hybrid color space, whose component images possess complementary characteristics, enhances the discriminating power for face recognition. Second, three effective image encoding methods are proposed for the component images in the RC r Q hybrid color space to extract features: (i) a patch-based Gabor image representation for the R component image, (ii) a multi-resolution LBP feature fusion scheme for the C r component image, and (iii) a component-based DCT multiple face encoding for the Q component image. Finally, at the decision level, the similarity matrices generated using the three component images in the RC r Q hybrid color space are fused using a weighted sum rule. Experiments on the Face Recognition Grand Challenge (FRGC) version 2 Experiment 4 show that the proposed method improves face recognition performance significantly. In particular, the proposed method achieves the face verification rate (ROC III curve) of 92.43%, at the false accept rate of 0.1%, compared to the FRGC baseline performance of 11.86% face verification rate at the same false accept rate. [Copyright &y& Elsevier]

Abstract: The Dichromatic Reflection Model (DRM) introduced by Shafer in 1985 is surely the most referenced physics-based model of image formation. A preliminary analysis of this model derives in the conclusion that colour channels remain coupled by the reflectance of objects surface material. Taking this observation as a basis, this paper shows that this coupling manifests itself at the signal level in the form of a set of properties only satisfied in uniform reflectance areas. Such properties are deeply analysed and formalized throughout the paper, and, eventually, they are stated geometrically. As will be seen, a compatibility relationship stems from this geometric interpretation, which allows checking at a very low cost whether two pixels correspond to the same scene reflectance or not. Finally, an edge detector based on the use of the previous relationship is presented as an example of application. This edge detector inherits all the properties of the compatibility relationship: simplicity, low computational power requirements, sensor noise adaptivity and insensitivity to image intensity changes due to scene objects curvature. [Copyright &y& Elsevier]

Abstract: This paper deals with recursive reconstruction of a scene model from unknown motion of a two-camera system capturing the images of the scene. Single camera systems with a relatively small field of view have limited accuracy because of the inherent confusion between translation and rotation. Estimation results from the stereo camera systems are also compromised due to this confusion if the systems require the fields of view to intersect for stereo correspondence. The cameras constituting the two-camera system considered in this paper are arranged so that there is a small intersection of the fields of view. This configuration of divergent view improves the accuracy of the structure and motion estimation because the ambiguity mentioned above decreases due to a large field of view. In this paper, a recursive algorithm is proposed for fast scene model reconstruction using a two-camera system of divergent view. Using inversely inferred stereo correspondences in the intersection of the fields of view is also proposed to remove degeneracy of scale factor determination and to acquire more accurate results from the information redundancy. The results of the experiments with long term real image sequences are presented to demonstrate the feasibility of the proposed system. [Copyright &y& Elsevier]

Abstract: Creating successful machine vision systems often begins a process of developing customised reliable image segmentation algorithms for the detection, and possibly categorisation of regions of interest within images. This can require significant investment of time from both the image processing and the domain experts to set up. Frequently this process is mediated via interviews, or language-based systems which may not fully capture the visual decision-making process of the domain experts. The resulting algorithms can also often be “brittle” in the sense of being highly specialised to the task for which they are tuned, and are consequently sensitive to changes in operating conditions or image specifications. One approach is to use interactive evolution for developing rapidly reconfigurable systems in which the users’ tacit knowledge and requirements can be elicited and used for finding the appropriate parameters to achieve the required segmentation without any need for specialised knowledge of the underlying machine vision systems. This paper presents an interactive tool that can be used to quickly and easily evolve optimal image segmentation parameters from scratch. Building on previous work, the new algorithm reported here incorporates user-guided local search and makes the fitness function more flexible to facilitate the underlying multi-objective decision-making process. One of the key requirements for any interactive system is a high level of usability, both in terms of effectiveness—being able to build accurate models that meet end-user requirements—and efficiency—being able to achieve the required results within a minimal amount of time and undue effort. The system described in this paper has been designed with these considerations in mind to ensure a high level of user-experience of the interaction process. We present results from a series of experiments with a range of users to analyse the effect of the improvements that have been made over the previous system. The efficiency of the tool is also tested with “novice users”, and its usability by “novice users” is analysed. [Copyright &y& Elsevier]

Abstract: A new touching cells splitting algorithm based on concave points and ellipse fitting is proposed in this paper. The algorithm includes two parts: contour pre-processing and ellipse processing. The purpose of contour pre-processing is to smooth fluctuations of the contour, find concave points of the contour and divide the contour into different segments via the concave points. The purpose of ellipse processing is to process the different segments of the contour into possible single cells by using the properties of the fitted ellipses. Because concave points divide the whole contour of touching cells into different segments and different segments of one single cell have similar properties, the ellipse processing can separate the touching cells through ellipse fitting. This paper demonstrates a new way of using ellipse fitting to split the binary contour of touching cells. Experimental results show that our algorithm is efficient. [Copyright &y& Elsevier]

Abstract: Recently, the visual secret sharing (VSS) technique by random grids (RG), proposed by Kafri and Keren in 1987, has drawn attention in academia again to remove the abovementioned drawbacks. However, Kafri and Keren''s scheme is a 2-out-of-2 VSS scheme but neither -out-of- nor 2-out-of-. In this paper, novel -out-of- (Method 1) and 2-out-of- (Method 2) secret image sharing schemes based on RG are proposed to encrypt the secret into cipher-grids without pixel expansion and additional codebook required. In the decryption phase, while participants superimpose all (Method 1) or at least two (Method 2) cipher-grids without any extra computation, the secret is recognizable by the human visual system. To our best knowledge, this paper is the first attempt in the literature to develop new RG-based VSS schemes by means of extending the basic 2-out-of-2 scheme to the -out-of- as well as 2-out-of- ones. To demonstrate the correctness of the proposed schemes, the formal analysis is given while the experimental results show the proposed schemes do work well. [Copyright &y& Elsevier]

Abstract: The hit-or-miss transform (HMT) is a fundamental operation on binary images, widely used since 40 years. As it is not increasing, its extension to grey-level images is not straightforward, and very few authors have considered it. Moreover, despite its potential usefulness, very few applications of the grey-level HMT have been proposed until now. Part I of this paper [B. Naegel, N. Passat, C. Ronse, Grey-level hit-or-miss transforms—part I: unified theory. Pattern Recogn., in press doi:10.1016/j.patcog.2006.06.004] was devoted to the description of a theory enabling to unify the main definitions of the grey-level HMT, mainly proposed by Ronse and Soille, respectively. Part II of this paper, developed hereafter, deals with the applicative potential of the grey-level HMT, illustrated by its use for vessel segmentation from 3D angiographic data. Different HMT-based segmentation methods are then described and analysed, leading to concrete analysis techniques for brain and liver vessels, but also providing algorithmic strategies which could further be used for many other kinds of image processing applications. [Copyright &y& Elsevier]

Abstract: Zernike moments which are superior to geometric moments because of their special properties of image reconstruction and immunity to noise, suffer from several discretization errors. These errors lead to poor quality of reconstructed image and wide variations in the numerical values of the moments. The predominant factor, as observed in this paper, is due to the discrete integer implementation of the steps involved in moment calculation. It is shown in this paper that by modifying the algorithms to include discrete float implementation, the quality of the reconstructed image improves significantly and the first-order moment becomes zero. Low-order Zernike moments have been found to be stable under linear transformations while the high-order moments have large variations. The large variations in high-order moments, however, do not greatly affect the quality of the reconstructed image, implying that they should be ignored when numerical values of moments are used as features. The 11 functions based on geometric moments have also been found to be stable under linear transformations and thus these can be used as features. Pixel level analysis of the images has been carried out to strengthen the results. [Copyright &y& Elsevier]

Abstract: A new method for detecting and tracking multiple moving objects based on discrete wavelet transform and identifying the moving objects by their color and spatial information is proposed in this paper. Many tracking algorithms have better performance under static background but get worse results under background with fake motions. Therefore, most of the tracking algorithms are used indoors instead of outdoor environment. Since discrete wavelet transform has a nice property that it can divide a frame into four different frequency bands without loss of the spatial information, it is adopted to solve this problem due to the fact that most of the fake motions in the background can be decomposed into the high frequency wavelet sub-band. In tracking multiple moving objects, many applications have problems when objects pass across each other. Color and spatial information are used in this paper to solve this problem. The experimental results prove the feasibility and usefulness of the proposed method. [Copyright &y& Elsevier]

Differentiating hole from component is an important issue in digital topology. In a recent paper, Lee, Poston, and Rosenfeld proposed a method to distinguish external and internal boundaries in 2D and 3D images relying on the property of normal vector and winding number. The method uses a smoothing function to replace digital lattice for calculating normal vector on image boundary. In this paper, we show that normal vector and winding number can be defined directly in 2D digital images and used for hole detection without resorting to any smoothing operation. We analyze first the discontinuity of Freeman codes of contour and prove its properties. We define then outward normal vector in 2D images and demonstrate also its discontinuity properties. The difficulty of counting the transition of normal vector in a given direction is analyzed and a solution is proposed. Based on the theoretic properties of edge code and normal vector, we propound the first algorithm to count the transitions of normal vector in a given direction, and consequently holes and external contours can be distinguished easily. We further define winding number directly in digital images, show its properties, and propose a second algorithm implementing the idea of winding number which is conceptually simpler and easier than the first one. A proof of correctness of our both algorithms is given and computation results are presented. [Copyright &y& Elsevier]

This paper presents a new algorithm for estimation-based range image segmentation. Aiming at surface-primitive extraction from range data, we focus on the reliability of the primitive representation in the process of region estimation. We introduce an optimal description of surface primitives, by which the uncertainty of a region estimate is explicitly represented with a covariance matrix. Then the reliability of an estimate is interpreted in terms of “measure of uncertainty”. The segmentation approach follows the region-growing scheme, in which the regions are estimated in an iterative way. With the probabilistic model proposed in this paper, surface homogeneity is defined and tested by an optimal criterion. A notable feature of the algorithm is that the order of merging is organized to lead the growth towards the most reliable representation of the merged region. Concerned with man-made objects in the scene, we restrict the class of surface primitives to be quadric or planar. The proposed algorithm has been applied to real data and synthetic data and demonstrated with experimental results. [Copyright &y& Elsevier]

This paper describes an effort towards the development of Arabic cheque databases for research in the recognition of hand-written Arabic cheques. Databases of real-life Arabic legal amounts, Arabic sub-words, courtesy amounts, Indian digits, and Arabic cheques are described. This paper highlights some characteristics of the Arabic language and presents the various steps that have been completed to build these databases including segmentation, binarization and data tagging. It also describes a solid validation procedure including grammars and algorithms used to verify the correctness of the tagging process. Detailed descriptions of the database organization and class distribution are included. These databases aim to facilitate experimental comparisons between various recognition methods, and will be provided to all interested researchers upon request to CENPARMI.11CENPARMI, Concordia University, GM-606, Montreal, Quebec H3E 1M8, Canada. [Copyright &y& Elsevier]

• We propose a new surround function to effectively estimate the illumination map. • We propose an Adaptive Retinex Block that can be regarded as an extension of Single Scale Retinex in features space. • We propose a low-exposure denoiser to smooth the image before enhancement. • We proposed a lightweight network that achieves competitive performance against several state-of-the-art methods with less than 150K parameters. Although Convolution Neural Networks (CNNs) have made substantial progress in the low-light image enhancement task, one critical problem of CNNs is the paradox of model complexity and performance. This paper presents a novel SurroundNet that only involves less than 150 K parameters (about 80–98 percent size reduction compared to SOTAs) and achieves very competitive performance. The proposed network comprises several Adaptive Retinex Blocks (ARBlock), which can be viewed as a novel extension of Single Scale Retinex in feature space. The core of our ARBlock is an efficient illumination estimation function called Adaptive Surround Function (ASF). It can be regarded as a general form of surround functions and be implemented by convolution layers. In addition, we also introduce a Low-Exposure Denoiser (LED) to smooth the low-light image before the enhancement. We evaluate the proposed method on two real-world low-light datasets. Experimental results demonstrate the superiority of our submitted SurroundNet in both performance and network parameters against State-of-the-Art low-light image enhancement methods. The code is available at https://github.com/ouc-ocean-group/SurroundNet. [ABSTRACT FROM AUTHOR]

This paper addresses one major aspect of a project aimed at the design of an image database query engine, where the images are searched at the 3D object-level. This approach is a novelty since image databases are usually searched by comparing colours, textures and 2D shapes of regions in the images. The main topic and contribution of this paper is a new method for object part segmentation, from constant-curvature contour primitives, that is suited to process images of objects in real scenes. The proposed method groups constant-curvature primitives using intermediate-level geometric relationships and object outline information. A detailed description of the method is presented along with validating experiments. [Copyright &y& Elsevier]

In unsupervised domain adaptation, a key research problem is joint distribution alignment across the source and target domains. However, direct alignment of the source and target joint distributions is infeasible, because the target conditional distribution cannot be known without target labels. Instead of estimating target labels for target conditional distribution approximation, this paper proposes a new criterion of domain-shared group sparsity that is an equivalent condition for conditional distribution alignment. Together with marginal distribution alignment, we develop a domain-shared group-sparse dictionary learning model to learn domain-shared representations with aligned joint distributions. A cross-domain label propagation method is then proposed to train a classifier for the target domain using the domain-shared group-sparse representations and the target-specific information from the target data. The proposed method outperforms eight state-of-the-art unsupervised domain adaptation algorithms for cross-domain face recognition and cross-dataset object recognition with hand-drafted and deep features. Experimental results across multiple sub-domains show that the proposed method also performs well across datasets with large variance. Our results are quantitatively and qualitatively analyzed, and experiments of parameter sensitivity and convergence analysis are performed to show the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

In large-scale visual recognition tasks, researchers are usually faced with some challenging problems, such as the extreme imbalance in the number of training data between classes or the lack of annotated data for some classes. In this paper, we propose a novel neural network architecture that automatically synthesizes pseudo feature representations for the classes in lack of annotated images. With the supply of semantic attributes for classes, the proposed Attribute-Based Synthetic Network (ABS-Net) can be applied to zero-shot learning (ZSL) scenario and conventional supervised learning (CSL) scenario as well. For ZSL tasks, the pseudo feature representations can be viewed as annotated feature-level instances for novel concepts, which facilitates the construction of unseen class predictor. For CSL tasks, the pseudo feature representations can be viewed as products of data augmentation on training set, which enriches the interpretation capacity of CSL systems. We demonstrate the effectiveness of the proposed ABS-Net in ZSL and CSL settings on a synthetic colored MNIST dataset (C-MNIST). For several popular ZSL benchmark datasets, our architecture also shows competitive results on zero-shot recognition task, especially leading to tremendous improvement to state-of-the-art mAP on zero-shot retrieval task. [ABSTRACT FROM AUTHOR]

In recent years, image annotation has attracted extensive attention due to the explosive growth of image data. With the capability of describing images at the semantic level, image annotation has many applications not only in image analysis and understanding but also in some relative disciplines, such as urban management and biomedical engineering. Because of the inherent weaknesses of manual image annotation, Automatic Image Annotation (AIA) has been raised since the late 1990s. In this paper, a deep review of state-of-the-art AIA methods is presented by synthesizing 138 literatures published during the past two decades. We classify AIA methods into five categories: 1) Generative model-based image annotation, 2) Nearest neighbor-based image annotation, 3) Discriminative model-based image annotation, and 4) Tag completion-based image annotation, 5) Deep Learning-based image annotation. Comparisons of the five types of AIA methods are made on the basis of the underlying idea, main contribution, model framework, computational complexity, computation time, and annotation accuracy. We also give an overview of five publicly available image datasets and four standard evaluation metrics commonly used as benchmarks for evaluating AIA methods. Then the performance of some typical or well-behaved models is assessed based on benchmark dataset and standard evaluation metrics. Finally, we share our viewpoints on the open issues and challenges in AIA as well as research trends in the future. [ABSTRACT FROM AUTHOR]