Concatenation Operation Research Articles

SCAR: Spatial-/channel-wise attention regression networks for crowd counting

Recently, crowd counting is a hot topic in crowd analysis. Many CNN-based counting algorithms attain good performance. However, these methods only focus on the local appearance features of crowd scenes but ignore the large-range pixel-wise contextual and crowd attention information. To remedy the above problems, in this paper, we introduce the Spatial-/Channel-wise Attention Models into the traditional Regression CNN to estimate the density map, which is named as “SCAR”. It consists of two modules, namely Spatial-wise Attention Model (SAM) and Channel-wise Attention Model (CAM). The former can encode the pixel-wise context of the entire image to more accurately predict density maps at the pixel level. The latter attempts to extract more discriminative features among different channels, which aids model to pay attention to the head region, the core of crowd scenes. Intuitively, CAM alleviates the mistaken estimation for background regions. Finally, two types of attention information and traditional CNN’s feature maps are integrated by a concatenation operation. Furthermore, the extensive experiments are conducted on four popular datasets, Shanghai Tech Part A/B, GCC, and UCF_CC_50 Dataset. The results show that the proposed method achieves state-of-the-art results.

Deep Dual-Channel Neural Network for Image-Based Smoke Detection

Smoke detection plays an important role in industrial safety warning systems and fire prevention. Due to the complicated changes in the shape, texture, and color of smoke, identifying the smoke from a given image still remains a substantial challenge, and this has accordingly aroused a considerable amount of research attention recently. To address the problem, we devise a new deep dual-channel neural network (DCNN) for smoke detection. In contrast to popular deep convolutional networks (e.g., Alex-Net, VGG-Net, Res-Net, and Dense-Net and the DNCNN that is specifically devoted to detecting smoke), our proposed end-to-end network is mainly composed of dual channels of deep subnetworks. In the first subnetwork, we sequentially connect multiple convolutional layers and max-pooling layers. Then, we selectively append the batch normalization layer to each convolutional layer for overfitting reduction and training acceleration. The first subnetwork is shown to be good at extracting the detailed information of smoke, such as texture. In the second subnetwork, in addition to the convolutional, batch normalization, and max-pooling layers, we further introduce two important components. One is the skip connection for avoiding the vanishing gradient and improving the feature propagation. The other is the global average pooling for reducing the number of parameters and mitigating the overfitting issue. The second subnetwork can capture the base information of smoke, such as contours. We finally deploy a concatenation operation to combine the aforementioned two deep subnetworks to complement each other. Based on the augmented data obtained by rotating the training images, our proposed DCNN can promptly and stably converge to the perfect performance. Experimental results conducted on the publicly available smoke detection database verify that the proposed DCNN has attained a very high detection rate that exceeds 99.5% on average, superior to state-of-the-art relevant competitors. Furthermore, our DCNN only employs approximately one-third of the parameters needed by the comparatively tested deep neural networks. The source code of DCNN will be released at https://kegu.netlify.com/ .

Semantic Sentence Matching with Densely-Connected Recurrent and Co-Attentive Information

Sentence matching is widely used in various natural language tasks such as natural language inference, paraphrase identification, and question answering. For these tasks, understanding logical and semantic relationship between two sentences is required but it is yet challenging. Although attention mechanism is useful to capture the semantic relationship and to properly align the elements of two sentences, previous methods of attention mechanism simply use a summation operation which does not retain original features enough. Inspired by DenseNet, a densely connected convolutional network, we propose a densely-connected co-attentive recurrent neural network, each layer of which uses concatenated information of attentive features as well as hidden features of all the preceding recurrent layers. It enables preserving the original and the co-attentive feature information from the bottommost word embedding layer to the uppermost recurrent layer. To alleviate the problem of an ever-increasing size of feature vectors due to dense concatenation operations, we also propose to use an autoencoder after dense concatenation. We evaluate our proposed architecture on highly competitive benchmark datasets related to sentence matching. Experimental results show that our architecture, which retains recurrent and attentive features, achieves state-of-the-art performances for most of the tasks.

A deep one-shot network for query-based logo retrieval

Logo detection in real-world scene images is an important problem with applications in advertisement and marketing. Existing general-purpose object detection methods require large training data with annotations for every logo class. These methods do not satisfy the incremental demand of logo classes necessary for practical deployment since it is practically impossible to have such annotated data for new unseen logo. In this work, we develop an easy-to-implement query-based logo detection and localization system by employing a one-shot learning technique using off the shelf neural network components. Given an image of a query logo, our model searches for logo within a given target image and predicts the possible location of the logo by estimating a binary segmentation mask. The proposed model consists of a conditional branch and a segmentation branch. The former gives a conditional latent representation of the given query logo which is combined with feature maps of the segmentation branch at multiple scales in order to obtain the matching location of the query logo in a target image. Feature matching between the latent query representation and multi-scale feature maps of segmentation branch using simple concatenation operation followed by 1 × 1 convolution layer makes our model scale-invariant. Despite its simplicity, our query-based logo retrieval framework achieved superior performance in FlickrLogos-32 and TopLogos-10 dataset over different existing baseline methods.

A Novel Effectively Optimized One-Stage Network for Object Detection in Remote Sensing Imagery

With great significance in military and civilian applications, the topic of detecting small and densely arranged objects in wide-scale remote sensing imagery is still challenging nowadays. To solve this problem, we propose a novel effectively optimized one-stage network (NEOON). As a fully convolutional network, NEOON consists of four parts: Feature extraction, feature fusion, feature enhancement, and multi-scale detection. To extract effective features, the first part has implemented bottom-up and top-down coherent processing by taking successive down-sampling and up-sampling operations in conjunction with residual modules. The second part consolidates high-level and low-level features by adopting concatenation operations with subsequent convolutional operations to explicitly yield strong feature representation and semantic information. The third part is implemented by constructing a receptive field enhancement (RFE) module and incorporating it into the fore part of the network where the information of small objects exists. The final part is achieved by four detectors with different sensitivities accessing the fused features, all four parallel, to enable the network to make full use of information of objects in different scales. Besides, the Focal Loss is set to enable the cross entropy for classification to solve the tough problem of class imbalance in one-stage methods. In addition, we introduce the Soft-NMS to preserve accurate bounding boxes in the post-processing stage especially for densely arranged objects. Note that the split and merge strategy and multi-scale training strategy are employed in training. Thorough experiments are performed on ACS datasets constructed by us and NWPU VHR-10 datasets to evaluate the performance of NEOON. Specifically, 4.77% and 5.50% improvements in mAP and recall, respectively, on the ACS dataset as compared to YOLOv3 powerfully prove that NEOON can effectually improve the detection accuracy of small objects in remote sensing imagery. In addition, extensive experiments and comprehensive evaluations on the NWPU VHR-10 dataset with 10 classes have illustrated the superiority of NEOON in the extraction of spatial information of high-resolution remote sensing images.

State complexity of unambiguous operations on finite automata

The paper determines the number of states in finite automata necessary to represent “unambiguous” variants of the union, concatenation, and Kleene star operations on formal languages. For the disjoint union of languages represented by an m-state and an n-state deterministic finite automata (DFA), the state complexity is mn−1; for the unambiguous concatenation, it is known to be m2n−1−2n−2 (Daley et al., “Orthogonal concatenation: language equations and state complexity”, J. UCS, 2010), and this paper shows that this number of states is necessary already over a binary alphabet; for the unambiguous star, the state complexity function is determined to be 382n+1. In the case of a unary alphabet, disjoint union requires up to 12mn states in a DFA, unambiguous concatenation has state complexity m+n−2, and unambiguous star requires n−2 states in the worst case. For nondeterministic finite automata, as well as for unambiguous finite automata, the complexity for disjoint union is m+n, for unambiguous concatenation, square, and star, the resulting complexities are m+n, 2n, and n+1, respectively, and all of them are witnessed by binary languages. In the unary nondeterministic or unambiguous case, the corresponding complexities are m+n for disjoint union, m+n−1 and 2n−1 for unambiguous concatenation and square, respectively, and n−1 for unambiguous star.

Multi‐scale Cross‐path Concatenation Residual Network for Poisson denoising

The signal degradation due to the Poisson noise is a common problem in the low-light imaging field. Recently, deep learning employing the convolution neural network for image denoising has drawn considerable attention owing to its favourable denoising performance. On the basis of the fact that the reconstruction of corrupted pixels can be facilitated by the context information in image denoising, the authors propose a deep multi-scale cross-path concatenation residual network (MC 2 RNet) which incorporates cross-path concatenation modules for Poisson denoising. Multiple paths are achieved by the cross-path concatenation operation and the skip connection. As a consequence, multi-scale context representations of images under different receptive fields can be learnt by MC 2 RNet. With the residual learning strategy, MC 2 RNet learns the residual between the noisy image and the latent clean image rather than the direct mapping to facilitate model training. Specially, unlike existing discriminative Poisson denoising algorithms that train a model only for the specific noise level, they aim to train a single model for handling Poisson noise with different levels, i.e. blind Poisson denoising. Quantitative experiments demonstrate that the proposed model is superior over the state-of-the-art Poisson denoising approaches in terms of peak signal-to-noise ratio and visual effect.

FPGA Implementation of a Novel Construction of Optical Zero-Correlation Zone Codes for OCDMA Systems

Abstract This paper proposes a novel method for constructing optical zero-correlation zone (OZCZ) codes based on the Mutually Orthogonal Complementary Sets (MOCS) matrix and the concatenation operation. The most important property of the obtained codes is that the periodic correlation function of the proposed codes is zero for the phase shifts within the zero-correlation zone. The procedure of construction is simple, flexible and offers increasing number of users M = 2 n + 1 \left( {M = {2^{n + 1}}} \right) . In the other hand, we have presented a fully reconfigurable hardware implementation based on very high description language of the proposed OZCZ codes for optical code-division multiple-access encoder system. The implementation is performed on an Altera field-programmable gate arrays device using ROM memory and an up counter. The simulation results have been found satisfactory and are in conformity with the theoretical observations.

A Two-tier Security Solution for Storing Data Across Public Cloud

Background: Data integrity protection in Cloud Computing becomes very challenging since the user no longer has the possession of their own data. This makes cloud data storage security of critical importance. The users can resort to legal action against the cloud provider if the provider fails to maintain the integrity of the shared data, but it also raises a need to secure users' private data across the public cloud. Methods: In this paper, we propose a novel end-to-end solution to ensure the security of data stored over a public cloud. It is a two-tier approach where the data is stored in an encrypted format and only the owner of the data can have access to the original data shared across the cloud. The algorithm OwnData: Encryption and Decryption is based on AES file encryption, which has the flexibility to be implemented across different cloud platforms. Results: The proposed OwnData model to provide privacy and confidentiality to the user data successfully secures data in an encrypted format. The users can gain full access control over the accessibility of their data. The application has been improvised to minimize page load time which enables it to achieve improvements in scalability. Algorithm and concatenation operators (dot) give minimum computation load during uploading of data to the cloud platform or downloading the same. Conclusion: The algorithm is robust, scalable and secure and It gives the user complete authorization and control over the data even when data is being stored remotely or in any other cloud premises.

State-Space Representations of Deep Neural Networks.

This letter deals with neural networks as dynamical systems governed by finite difference equations. It shows that the introduction of -many skip connections into network architectures, such as residual networks and additive dense networks, defines th order dynamical equations on the layer-wise transformations. Closed-form solutions for the state-space representations of general th order additive dense networks, where the concatenation operation is replaced by addition, as well as th order smooth networks, are found. The developed provision endows deep neural networks with an algebraic structure. Furthermore, it is shown that imposing th order smoothness on network architectures with -many nodes per layer increases the state-space dimension by a multiple of , and so the effective embedding dimension of the data manifold by the neural network is -many dimensions. It follows that network architectures of these types reduce the number of parameters needed to maintain the same embedding dimension by a factor of when compared to an equivalent first-order, residual network. Numerical simulations and experiments on CIFAR10, SVHN, and MNIST have been conducted to help understand the developed theory and efficacy of the proposed concepts.

An Efficient Essential Secret Image Sharing Scheme Using Derivative Polynomial

As a popular technology in information security, secret image sharing is a method to guarantee the secret image’s security. Usually, the dealer would decompose the secret image into a series of shadows and then assign them to a number of participants, and only a quorum of participants could recover the secret image. Generally, it is assumed that every participant is equal. Actually, due to their position in many practical applications, some participants are given special privileges. Therefore, it is desirable to give an approach to generate shadows with different priorities shadows. In this paper, an efficient essential secret image sharing scheme using a derivative polynomial is proposed. Compared with existing related works, our proposed scheme can not only create the same-sized shadows with smaller size but also removes the concatenation operation in the sharing phase. Theoretical analysis and simulations confirm the security and effectiveness of the proposed scheme.

Building complex events

We examine the Doubly Inflected Construction of Sicilian (DIC; Cardinaletti and Giusti 2001, 2003; Cruschina 2013), in which a motion verb V1 from a restricted set is followed by an event verb V2 and both verbs are inflected for the same person and tense features. The interpretation of DIC involves a complex event which behaves as a single, integrated event by linguistic tests. Based on data drawn from different sources, we argue that DIC is an asymmetrical serial verb construction (Aikhenvald 2006). We propose an analysis of DIC in which V1 and V2 enter the semantic composition as lexical verbs, with V1 contributing a motion event and projecting a theme and a goal argument which are identified, respectively, with an agent and a location argument projected by V2. A morphosyntactic mechanism of feature-spread requires that the person and tense features be realized both on V1 and on V2, while, semantically, these features are interpreted only once, in a position from which they take scope over the complex predicate resulting from the combination of V1 and V2. The semantic analysis is based on an operation of event concatenation, defined over spatio-temporally contiguous events which share specific participants, and is implemented in a neo-Davidsonian framework (Parsons 1990).

Nondeterministic complexity in subclasses of convex languages

We study the nondeterministic state complexity of basic regular operations on the classes of prefix-, suffix-, factor-, and subword-free, -closed, and -convex regular languages and on the classes of right, left, two-sided, and all-sided ideal regular languages. For the operations of concatenation, intersection, union, reversal, star, and complementation, we get tight upper bounds for all considered classes except for complementation on factor- and subword-convex languages. Most of our witnesses are described over optimal alphabets. The description of a proper suffix-convex language over a five-letter alphabet meeting the upper bound 2n for complementation, and obtaining an asymptotically tight bound Θ(n) for complementation of unary prefix-free languages are among the most interesting results of this paper.

An out-of-band mobile authenticating mechanism for controlling access to data outsourced in the mobile cloud environment

'Mobile Cloud Computing' (MCC), a cloud environment, formed by mobile users at the client-side and cloud servers at the back-end enables users to store and pervasively access a huge amount of data via different mobile devices in a distributive manner. This increases the risk of data abuse, since data is outsourced on the distrusted cloud servers. The drawbacks of the current frameworks, such as overloaded computations in key distribution, reduced flexibility, and scalability, are unable to achieve fine-graininess and confidentiality. Thus, we propose an 'access control mechanism' which is lightweight with minimal computational overhead and provides fine-grained access control for sharing data using out-of-band (OOB) mobile authentication. Here we perform client-side encryption and decryption using simple hash functions and concatenation operator and achieve dynamic scalability. The encryption algorithms, sharing mechanism, and the use of OOB mobile authentication are extensively analysed to prove its efficiency and applicability.

An out-of-band mobile authenticating mechanism for controlling access to data outsourced in the mobile cloud environment

'Mobile Cloud Computing' (MCC), a cloud environment, formed by mobile users at the client-side and cloud servers at the back-end enables users to store and pervasively access a huge amount of data via different mobile devices in a distributive manner. This increases the risk of data abuse, since data is outsourced on the distrusted cloud servers. The drawbacks of the current frameworks, such as overloaded computations in key distribution, reduced flexibility, and scalability, are unable to achieve fine-graininess and confidentiality. Thus, we propose an 'access control mechanism' which is lightweight with minimal computational overhead and provides fine-grained access control for sharing data using out-of-band (OOB) mobile authentication. Here we perform client-side encryption and decryption using simple hash functions and concatenation operator and achieve dynamic scalability. The encryption algorithms, sharing mechanism, and the use of OOB mobile authentication are extensively analysed to prove its efficiency and applicability.

Automated Brittle Fracture Rate Estimator for Steel Property Evaluation Using Deep Learning After Drop-Weight Tear Test

This study proposes an automated brittle fracture rate (BFR) estimator using deep learning. As the demand for line-pipes increases in various industries, the need for BFR estimation through drop-weight tear test (DWTT) increases to evaluate steel's property. Conventional BFR or ductile fracture rate (DFR) estimation methods require an expensive 3D scanner. Alternatively, a rule-based approach is used with a single charge-coupled device (CCD) camera. However, it is sensitive to the hyper-parameter. To solve these problems, we propose an approach based on deep learning that has recently been successful in the fields of computer vision and image processing. The method proposed in this study is the first to use deep learning approach for BFR estimation. The proposed method consists of a VGG-based U-Net (VU-Net) which is inspired by U-Net and fully convolutional network (FCN). VU-Net includes a deep encoder and a decoder. The encoder is adopted from VGG19 and transferred with a pre-trained model with ImageNet. In addition, the structure of the decoder is the same as that of the encoder, and the decoder uses the feature maps of the encoder through concatenation operation to compensate for the reduced spatial information. To analyze the proposed VU-Net, we experimented with different depths of networks and various transfer learning approaches. In terms of accuracy used in real industrial application, we compared the proposed VU-Net with U-Net and FCN to evaluate the performance. The experiments showed that VU-Net was the accuracy of approximately 94.9 %, and was better than the other two, which had the accuracies of about 91.8 % and 93.7 %, respectively.

Mirror symmetry detection in curves represented by means of the Slope Chain Code

Symmetry is an important feature in natural and man-made objects; particularly, mirror symmetry is a relevant task in fields such as computer vision and pattern recognition. In the current work, we propose a new method to characterize mirror-symmetry in open and closed curves represented by means of the Slope Chain Code. This representation is invariant under scale, rotation, and translation, highly desirable properties for object recognition applications. The proposed method detects symmetries through simple inversion, concatenation and reflection operations on the chains, thus allowing the classification of symmetrical and asymmetrical contours. It also introduces a measure to quantify the degree of symmetry in quasi-mirror-symmetrical objects. Furthermore, it allows the identification of multiple symmetry axes and their location. Results show high performances in symmetrical/asymmetrical classification (0.9 recall, 0.9 accuracy, 0.97 precision) and axes’ detection (0.8 recall, 0.84 accuracy, 0.99 precision). Compared to other methods, the proposed algorithm provides properties such as: global, local, and multiple axes’ detection, as well as the capability to classify symmetrical objects, which makes it adequate for several practical applications, like the three exemplified in the paper.

An Improved Essential Secret Image Sharing Scheme with Smaller Shadow Size

This article describes how in conventional (k, n) secret image sharing (SIS), any k or more shadows can be collaborated together to reveal the secret image. Usually, all shadows have the same importance. Recently, an essential (t, s, k, n) secret image sharing (ESIS) with different importance of shadows were proposed where the n shadows are divided into s essential shadows and n − s non-essential shadows. In reconstruction, it needs at least k shadows which should include at least t essential shadows. In this article, the authors propose an improved (t, s, k, n)-ESIS scheme, where t<k. In their scheme, essential shadows and non-essential shadows have the equal sizes. No concatenation operation is needed in generating shadows. Besides, all shadow sizes are reduced to 1/k times of the secret image. Theoretical analyses and experimental results show that the proposed scheme is feasible with optimal sharing ratios among related works.

A Multiscale Fusion Convolutional Neural Network for Plant Leaf Recognition

Plant leaf recognition is a computer vision task used to automatically recognize plant species. It is very challenging since rich plant leaf morphological variations, such as sizes, textures, shapes, venation, and so on. Most existing plant leaf methods typically normalize all plant leaf images to the same size and recognize them at one scale, resulting in unsatisfactory performances. In this letter, a multiscale fusion convolutional neural network (MSF-CNN) is proposed for plant leaf recognition at multiple scales. First, an input image is down-sampled into multiples low resolution images with a list of bilinear interpolation operations. Then, these input images with different scales are step-by-step fed into the MSF-CNN architecture to learn discriminative features at different depths. At this stage, the feature fusion between two different scales is realized by a concatenation operation, which concatenates feature maps learned on different scale images from a channel view. Along with the depth of the MSF-CNN, multiscale images are progressively handled and the corresponding features are fused. Third, the last layer of the MSF-CNN aggregates all discriminative information to obtain the final feature for predicting the plant species of the input image. Experiments show the proposed MSF-CNN method is superior to multiple state-of-the art plant leaf recognition methods on the MalayaKew Leaf dataset and the LeafSnap Plant Leaf dataset.

Free monoids and generalized metric spaces

Let A be an ordered alphabet, A∗ be the free monoid over A ordered by the Higman ordering, and let F(A∗) be the set of final segments of A∗. With the operation of concatenation, this set is a monoid. We show that the submonoid F∘(A∗)≔F(A∗)∖{0̸} is free. The MacNeille completion N(A∗) of A∗ is a submonoid of F(A∗). As a corollary, we obtain that the monoid N∘(A∗)≔N(A∗)∖{0̸} is free. We give an interpretation of the freeness of F∘(A∗) in the category of metric spaces over the Heyting algebra V≔F(A∗), with the non-expansive mappings as morphisms. Each final segment F of A∗ yields the injective envelope SF of a two-element metric space over V. The uniqueness of the decomposition of F is due to the uniqueness of the block decomposition of the graph GF associated to this injective envelope.