Search

The dissemination of emerging electronic devices increases considerably alongside with the Internet of Things (IoT) paradigm. Some key elements considered in these devices (e.g. rare-earth) are becoming critical materials and are associated with serious environmental concerns. Therefore, there is a major requirement for designing differently our electronic technologies at the early stages of the research process. Many reports present the promising properties of paper-based electronics as meant to address this challenge. Through technical and life cycle assessments, this article discusses the main design levers in terms of material and fabrication process choices for the eco-design of paper-electronics water quality sensors. Our results suggest conscientious choices of conductive nanomaterials and additive fabrication techniques.

Integration of hazardous compounds emerging from modern human activities into the environment is a major concern. Given its versatility, paper is being exploited as a microfluidic network, preconcentration platform, and/or microextraction membrane accompanied by specialized reaction zones to perform analytical tasks aimed to solve real-world demands related to environmental analysis. Furthermore, advantageous paper-based analytical devices intended for environmental monitoring are being supported by new technologies such as smartphones and sample collection through drones. Herein, we offer a critical overview on this relevant field highlighting challenges and opportunities.

One of the benefits of Internet of Vehicles (IoV) is improved traffic safety and efficiency, for example due to the capability to share vehicular messages in real-time. While most of the vehicular messages only need to be shared by nearby vehicles, some messages (e.g., announcement messages) may need to be more broadly distributed, for example to vehicles in a wider region. Finding a single trusted entity to store and distribute such messages can be challenging, and vehicles may not be inclined to participate (e.g., generation and distribution of announcement messages) unless they can benefit from such participation. In addition, achieving both security and privacy can be challenging. In this paper, we propose a blockchain based secure data sharing system to address the above challenges in an IoV setting. Specifically, in our system, announcement messages are stored using blockchain. To encourage/incentivize participation, vehicles that faithfully broadcast the announcement messages and/or contribute to the block generation will be rewarded by some cryptocurrency. Our system is also designed to be privacy-preserving and realizes both priori and posteriori countermeasures.

With the surging energy price and environmental concerns, measures to improve energy efficiency have attracted increasing concerns of the manufacture sector, especially energy-intensive manufacturing industries such as tissue paper mills. Energy-efficiency scheduling, as a novel energy-efficient method, has attracted the attention of an increasing number of researchers in recent years. Drying process is the most energy-intensive production process in tissue paper mills, which has a great energy-saving potential. This paper aims to reduce the energy costs for the tissue paper mill, consisting of processing energy cost and set-up energy cost, through integrating drying process optimization with energy-efficient scheduling. First, the energy cost model and the scheduling model were built. Then, the energy cost of the drying process of every job in a given scheduling problem was optimized using particle swarm optimization (PSO). Afterwards, the energy cost was further optimized using energy-efficiency scheduling. In addition, a hybrid non-dominated sorting genetic algorithm II (NSGA-II) was utilized to solve the energy-efficiency scheduling problem. Finally, several real scheduling problems from a real tissue paper mill were addressed using the proposed approach to demonstrate its effectiveness in energy saving. The experiment result showed that there is a great energy-saving potential in the drying process, accounting for up to 12.53% of the total energy consumption. Moreover, the maximum energy saving ratio of the proposed approach could reach 9.03%. On the whole, the proposed approach can provide a new energy-saving method for tissue paper mills or other manufacturing industries.

Membrane bioreactor (MBR) models are useful tools for both design and management. The system complexity is high due to the involved number of processes which can be clustered in biological and physical ones. Literature studies are present and need to be harmonized in order to gain insights from the different studies and allow system optimization by applying a control. This position paper aims at defining the current state of the art of the main integrated MBR models reported in the literature. On the basis of a modelling review, a standardized terminology is proposed to facilitate the further development and comparison of integrated membrane fouling models for aerobic MBRs. ispartof: WATER SCIENCE AND TECHNOLOGY vol:81 issue:1 pages:1-9 ispartof: location:England status: published

Economically paper and pulp mills are important but at the same time it is facing problems due to pollutants that are produced during the process. In this paper we will discuss methods of waste water treatment in the industry. This industry use excessive amount of water because it is used in nearly each and every step of the process as a result large quantity of waste water and residual sludge waste is produced leading to the problems of wastewater treatment, discharge and sludge disposal but advanced technologies with new equipment with the help of some innovative strategies are making it possible for treating wastewater for its reuse by presenting solutions to the industries and helping them in choosing suitable and affordable methods for management of waste

Technology is shaping our lives in a multitude of ways. This is fuelled by a technology infrastructure, both legacy and state of the art, composed of a heterogeneous group of hardware, software, services, and organisations. Such infrastructure faces a diverse range of challenges to its operations that include security, privacy, resilience, and quality of services. Among these, cybersecurity and privacy are taking the centre-stage, especially since the General Data Protection Regulation (GDPR) came into effect. Traditional security and privacy techniques are overstretched and adversarial actors have evolved to design exploitation techniques that circumvent protection. With the ever-increasing complexity of technology infrastructure, security and privacy-preservation specialists have started to look for adaptable and flexible protection methods that can evolve (potentially autonomously) as the adversarial actor changes its techniques. For this, Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL) were put forward as saviours. In this paper, we look at the promises of AI, ML, and DL stated in academic and industrial literature and evaluate how realistic they are. We also put forward potential challenges a DL based security and privacy protection system has to overcome. Finally, we conclude the paper with a discussion on what steps the DL and the security and privacy-preservation community have to take to ensure that DL is not just going to be hype, but an opportunity to build a secure, reliable, and trusted technology infrastructure on which we can rely on for so much in our lives.

For years, achievements and discoveries made by researcher are made aware through research papers published in appropriate journals or conferences. Many a time, established s researcher and mainly new user are caught up in the predicament of choosing an appropriate conference to get their work all the time. Every scienti?c conference and journal is inclined towards a particular ?eld of research and there is a extensive group of them for any particular ?eld. Choosing an appropriate venue is needed as it helps in reaching out to the right listener and also to further one’s chance of getting their paper published. In this work, we address the problem of recommending appropriate conferences to the authors to increase their chances of receipt. We present three di?erent approaches for the same involving the use of social network of the authors and the content of the paper in the settings of dimensionality reduction and topic modelling. In all these approaches, we apply Correspondence Analysis (CA) to obtain appropriate relationships between the entities in question, such as conferences and papers. Our models show hopeful results when compared with existing methods such as content-based ?ltering, collaborative ?ltering and hybrid ?ltering.

Purpose The purpose of this paper is to propose a cradle-to-cradle (C2C) zero discharge production planning system with a fuzzy hybrid optimization model that uses fuzzy comprehensive evaluation to establish fuzzy indicators, and then defuzzify the fuzzy indicators to construct a fuzzy multiobjective programming (FMOP) model. Design/methodology/approach The FMOP model pursues overall satisfaction using a particle swarm optimization algorithm to produce the best output values for the maximum waste paper recovery rate, the condensate reuse quality and minimum total cost of the zero discharge production planning system. Findings Recovered waste heat is seldom recycled and consumed in Taiwan. There is a need to capture and utilize heat recovery and use it in the production process supply. In contrast, waste materials are used as resources to perform waste paper recovery and recycle the waste heat of evaporation collection in the production process. Originality/value This paper develops a system to establish the best output value for an overall high satisfaction level. According to the results, the waste paper recovery rate is 99.8 percent, condensate reuse quality water is 102.6 tons and the total cost of the zero discharge production planning system is NT$1,312,012.

In this paper a compact, flexible, Multiband antenna is designed. It is having flexible Substrate of photographic paper which makes the antenna conformal. The antenna with two U slots cut in radiating patch is offset fed by microstrip lines gives the multiple bands of WLAN /WIMAX Application and two parasitic Elements improves the bandwidth of a typical microstrip antenna. The Measured results shows that the antenna resonates between Frequency band 2.10-2.95 GHz, 3.35-3.54 GHz and 5.04-6.0Hz.which are used for application such as Bluetooth, Wi-Fi, Zigbee, ISM 2.4GHz,WLAN WiMAX, and WLAN 5.2GHz.

Due to increased awareness and strictly imposed legislations, Supply Chain (SC) managers are often forced to include recovery system in their SC. Recovery systems make use of 3 Rs, namely, recyclin...

Digitalisation provides many opportunities for process improvements by using smart devices such as Augmented-reality (AR) displays in production. Nevertheless, in-depth studies about the effects of these applications on key performance process indicators are sparse. The goal of this study is to evaluate AR-based remote maintenance processes supported by cutting edge optical head-mounted display technology. A total of 12 test runs conducted with real-world maintenance engineers in a real-world industry setting were analysed based on video recordings and semi-structured interviews. The study identified both case-specific and general issues, and the inherent potential of the technology.

Sensitivity analysis (SA) has been used to evaluate the behavior and quality of environmental models by estimating the contributions of potential uncertainty sources to quantities of interest (QoI) in the model output. Although there is an increasing literature on applying SA in environmental modeling, a pragmatic and specific framework for spatially distributed environmental models (SD-EMs) is lacking and remains a challenge. This article reviews the SA literature for the purposes of providing a step-by-step pragmatic framework to guide SA, with an emphasis on addressing potential uncertainty sources related to spatial datasets and the consequent impact on model predictive uncertainty in SD-EMs. The framework includes: identifying potential uncertainty sources; selecting appropriate SA methods and QoI in prediction according to SA purposes and SD-EM properties; propagating perturbations of the selected potential uncertainty sources by considering the spatial structure; and verifying the SA measures based on post-processing. The proposed framework was applied to a SWAT (Soil and Water Assessment Tool) application to demonstrate the sensitivities of the selected QoI to spatial inputs, including both raster and vector datasets - for example, DEM and meteorological information - and SWAT (sub)model parameters. The framework should benefit SA users not only in environmental modeling areas but in other modeling domains such as those embraced by geographical information system communities.

Even though AI/ML is of strategic importance for many companies, it is not easy to come up with an AI life cycle in a multi-cloud world that is being increasingly embraced. In this paper, we present our position on an AI strategy that is future proof and does not force vendor lock in.

In Paper and Pulp (P&P) industry resource management scenarios, such as processing cell(s) efficiency reduction, equipment break-down or failures, alternative energy source (biomass and solar energy) utilization/ synchronization with regular energy sources and management decided change in production rate or product quality often occur in industrial processes causing material flow disruption, energy and production losses. In such scenarios, immediate and appropriate actions are required to be taken to minimize energy and production losses. Emerging resource management options tackle these issues through four different ways i.e., by mixing of readymade pulp at various processing stages, by varying the percentage of input raw material, by adopting a parallel utilization of alternative energy sources and by switching on/off some processing units or parallel working segments. These ways are implementable with immediate effect, accurately monitorable and their degree of application can be regulated within a wide range, thus denoted as adaptive solutions. This research work proposes a discrete-event simulation based real-time resource management technique for paper and pulp industry, wherein some resource management challenges are addressed and solved strategically with adaptive solutions. The presented examples validate that using the developed approach, the effective adaptive strategy can be timely planned to resolve occasionally or permanently happening resource management challenges ensuring an un-interrupted material flow throughout the plant along with the minimization of productivity and energy losses.

Most existing studies and public datasets for handwritten Chinese text recognition are based on the regular documents with clean and blank background, lacking research reports for handwritten text recognition on challenging areas such as educational documents and financial bills. In this paper, we focus on examination paper text recognition and construct a challenging dataset named examination paper text (SCUT-EPT) dataset, which contains 50 000 text line images (40 000 for training and 10 000 for testing) selected from the examination papers of 2 986 volunteers. The proposed SCUT-EPT dataset presents numerous novel challenges, including character erasure, text line supplement, character/phrase switching, noised background, nonuniform word size, and unbalanced text length. In our experiments, the current advanced text recognition methods, such as convolutional recurrent neural network (CRNN) exhibits poor performance on the proposed SCUT-EPT dataset, proving the challenge and significance of the dataset. Nevertheless, through visualizing and error analysis, we observe that humans can avoid vast majority of the error predictions, which reveal the limitations and drawbacks of the current methods for handwritten Chinese text recognition (HCTR). Finally, three popular sequence transcription methods, connectionist temporal classification (CTC), attention mechanism, and cascaded attention-CTC are investigated for HCTR problem. It is interesting to observe that although the attention mechanism has been proved to be very effective in English scene text recognition, its performance is far inferior to the CTC method in the case of HCTR with large-scale character set.

Fault diagnosis of sequential batch reactor (SBR), a widely applied wastewater treatment technology in papermaking industry with huge discharge, has been a significant challenge due to the inherent multi-period characteristics of the process. In this paper, based on the conventional multi-way principal component analysis (MPCA) method (Scenario 0), two sub-period division strategies based on the processing phases of SBR process (Scenario 1) and the similarities of the loading matrices between the adjacent time slices (Scenario 2) are proposed for the detection of faults. Combined with Scenario 0, using the field data of blower current, level of SBR reactor, dissolved oxygen of wastewater and the blower valve opening in the SBR process of paper mill, two different fault diagnosis models with Scenarios 1 and 2 are developed and evaluated, respectively. The study results revealed that, both the calculated statistics of T2 and sum of prediction errors (SPE) of the fault diagnosis models with Scenarios 1 and 2 could detect the faults and identify the fault locations and sources. Compared to Scenario 0 which neglects the correlations between the different stages of SBR process, the fault diagnosis model by Scenario 2 demonstrated a superiority ability in fault identification in terms of the fault’s time onset and fault’s sources with adequate accuracy. The results enable the feasible and reliable implementation of the developed sub-MPCA diagnosis model with Scenario 2 in the actual SBR plants.

The paper describes the stages of paper production, considers the structure of a paper-making machine. Questions related to the proof and use of the Pontryagin’s maximum principle in the theory of optimal control are considered. Optimal paper web weight control system based on the Pontryagin’s maximum principle is presented. Adaptive learning methods for modeling nonlinear systems represent some of the latest advances in adaptive algorithms and machine learning techniques designed to model and identify nonlinear systems. Real-world problems always involve a certain degree of non-linearity, which makes linear models a suboptimal choice. This article may be of interest to research engineers and practitioners in the study and application of control systems using adaptive regulators. This book serves as an essential resource for researchers, graduate students and doctoral students working in the field of machine learning, signal processing, adaptive filtering, nonlinear control, system identification, cooperative systems, and computational intelligence. This book may also be of interest to the industry market and practitioners working with a wide range of nonlinear systems.

Nowadays, there is a rapidly increasing number of conferences and journals in computer science that make a lot of challenges for researchers to find an appropriate venue to submit their scientific work. There is a need for a recommendation system that can support researchers for a better process of paper submission. In this paper, we present an efficient approach for building such a recommendation model by using embedding methods, Global Vector (GloVe) 1 created by Pennington et al. [1] and FastText 2 proposed by Facebook [2], Convolutional Neural Network (CNN) [3], and LSTM. The experimental results show that the combination of CNNs and FastText, CNN + FastText, can achieve the best performance in terms of the Top 1 Accuracy compared with other techniques, including the S2RSCS model, as presented in [4]. Moreover, the performance by using GloVe or FastText is much better, faster, and more stable than S2RSCS in most cases.

Prior work on texture analysis of historic, photographic papers has focused primarily on measures of texture similarity. However, automated grouping or clustering of photographic paper textures in a way that is meaningful to art conservators remains an open problem. In this work a deep learning approach to automated classification is presented, for clusters derived from a human sorting experiment conducted by 19 art conservators and paper experts and subsequently extended through crowd-sourcing. The proposed approach uses a deep convolutional neural network, and results are presented on the performance in automatically classifying images when compared to human experts.

American democracy is critically threatened by the use of insecure voting systems. Many existing electronic voting machines have malfunctioned during recent elections, and many are also vulnerable to hacking (Appel et al. 2019, Blaze et al. 2019, Blaze 2020). Most states have switched to secure, hand-marked paper ballots, but roughly 30% of Americans will continue to vote using vulnerable voting machines in 2020 (Cordova et al. 2019, Bajak 2020). While Congress allocated $380 million in 2018 and $425 million in 2020 to improve election security, these funds were neither targeted at nor sufficient for replacing all electronic voting machines. We propose that Congress (1) allocate $110 million exclusively for transitioning away from electronic voting machines and (2) prohibit the use of federal funds for purchasing voting systems that do not primarily use hand-marked paper ballots. This one-time transition cost is much smaller than even annual expenditures on other critical infrastructure (Copeland 2010; Halderman 2019). Replacing electronic voting machines with hand-marked paper ballots is the most affordable and secure option.

In Nigeria, poor waste management practices are evident in many urban centers. This study considered a University community as the model to explore the use of digital solutions in proper waste management. This study developed a web-based waste management system for the collection of plastics and paper in a University community. The framework developed utilized the use case and system block diagram. In this study, HTML, CSS, MySQL and Java programming language was used in the design of the web-based system. The result of this study was presented using screenshots. The web-based system had an interface to show profiles of the users, waste pickup information, incentive platform, and the admin dashboard to monitor and approve pickups and incentives. Past studies have indicated the success of using incentives to build sustainable behavioural change towards efficient waste disposal. Therefore, this study was able to introduce the incentive platform which converts weight of paper and plastics waste to points then to rewards for the users. The web-based waste management system for the collection of plastics and paper can be deployed for University communities to help reduce the number of paper and plastics that end up in landfills. Also, by picking the paper and plastics waste from source helps to prevent bin contamination.

Keyphrases in a research paper succinctly capture the primary content of the paper and also assist in indexing the paper at a concept level. Given the huge rate at which scientific papers are published today, it is important to have effective ways of automatically extracting keyphrases from a research paper. In this paper, we present a novel method, Syntax and Semantics Aware Keyphrase Extraction (SaSAKE), to extract keyphrases from research papers. It uses a transformer architecture, stacking up sentence encoders to incorporate sequential information, and graph encoders to incorporate syntactic and semantic dependency graph information. Incorporation of these dependency graphs helps to alleviate long-range dependency problems and identify the boundaries of multi-word keyphrases effectively. Experimental results on three benchmark datasets show that our proposed method SaSAKE achieves state-of-the-art performance in keyphrase extraction from scientific papers.

Physical synthesis has emerged as a core component in a modern circuit design flow. Large-scale optimization problem is often involved in the process, which requires substantial efforts to solve and no optimality is guaranteed. Reinforcement learning provides one direction to deal with the above issue by automatically acquiring knowledge through experience, which has shown great success in various applications. In this paper, we introduce the foundation of and the progress in reinforcement learning, and review some recent approaches in applying reinforcement learning to physical synthesis. We hope to inspire more work and to see more talented ideas in this field.

This paper considers a cell-free massive MIMO system with multiple-antenna access points and single-antenna users. The APs use conjugate beamforming to beamform the data to all users in the network. Total energy efficiency maximization is investigated. This optimization problem is nonconvex and thus difficult to solve. Existing solutions are based on second-order optimization methods in connection with convex approximations. These methods have been shown to perform very well but their complexity does not scale favorably with the network size. To tackle this issue, in this paper we propose to use a first-order method for nonconvex programming to our energy efficiency problem. Compared to the second-order methods, the proposed method achieves the same performance, while its run time is much faster. Thus, it is considered as a feasible solution for resource allocation in cell-free massive MIMO systems.

Facial Expressions play a vital role in the process of recognizing emotions and also to have non-verbal communication as well as in identifying people. Emotional Recognition turns out to be very important in everyday life just next to the tone of voice. Novel applications in Human-Computer Interaction (HMI) have been enabled by this and in many other areas. Inevitably most of the recent research on this area focuses on Convolutional Neural Networks (CNN) for extraction of features and inference from those features. In this paper we used CNN algorithm along with Histogram of gradients (HOG) features for higher accuracy. We perform emotional analysis of lecturers from Impartus videos. By detecting their emotions from a sequence of videos during their coursework, we evaluate the feedback they are expected to receive at the end of the duration of their course.

Diabetic Retinopathy (DR) is a polygenic disorder issue that affects human eyes. Bruise to the blood vessels of the photosensitive tissue of the retina causes this complication. It’s most frequent in patients who had diabetes for more than ten years. This downside is going on in several individuals worldwide. However, the number of medical practitioners and also the tools needed for the detection of DR are very less for serving the mass population. In this paper, we have proposed DRDNet (Diabetic Retinopathy Diagnosis Network), a neural network framework based on capsule networks (CapsNets) for DR diagnosis. Experiments on a dataset with 1,265 images demonstrate that CapsNet shows better accuracy and convergence behavior for the complex data than the state-ofthe-art techniques. The proposed DRDNet performs with an overall accuracy of 80.59% for five class, as compared to the closest competitor with an accuracy of 75.83%. We performed a study on a mixed dataset for two class and found that testing accuracy was 80.59%. We have also done training on a two class model and testing on other unseen datasets. Moreover, we observed that DRDNet has much higher confidence for the predicted probabilities as compared to other state-of-the-art techniques.

We present an open-source online dictionary editing system, Ve′rdd, that offers a chance to re-evaluate and edit grassroots dictionaries that have been exposed to multiple amateur editors. The idea is to incorporate community activities into a state-of-the-art finite-state language description of a seriously endangered minority language, Skolt Sami. Problems involve getting the community to take part in things above the pencil-and-paper level. At times, it seems that the native speakers and the dictionary oriented are lacking technical understanding to utilize the infrastructures which might make their work more meaningful in the future, i.e. multiple reuse of all of their input. Therefore, our system integrates with the existing tools and infrastructures for Uralic language masking the technical complexities behind a user-friendly UI.

To identify quality issues within the production and prevent defect products to be delivered to customers is critical for most manufacturing companies, and usually performed both within and at the end of each production section. In this paper we investigate the use of deep neural networks for performing automatic quality inspections based on image processing, with the aim of eliminating today’s manual inspection processes. A deep neural network is implemented on a real-world industrial case study and its performance is evaluated and analyzed when it comes to detecting quality problems in produced products. The results show that the network has an accuracy of 94.5% which is considered good in comparison to the 70-80% accuracy that a trained human inspector can achieve. CC BY-NC-ND 4.0Edited by Robert X. Gao, Kornel EhmannThe authors would like to thank Jonsac AB for their support in the study and for allowing us to work in their facility. The authors also want to thank Vinnova for financing the VISION project through the strategic innovation program Produktion2030, within which this work has been undertaken.

Video detection and classification constantly involve high dimensional data that requires a deep neural network (DNN) with huge number of parameters. It is thereby quite challenging to develop a DNN video comprehension at terminal devices. In this paper, we introduce a deeply tensor compressed video comprehension neural network called DEEPEYE for inference at terminal devices. Instead of building a Long Short-Term Memory (LSTM) network directly from raw video data, we build a LSTM-based spatio-temporal model from tensorized time-series features for object detection and action recognition. Moreover, a deep compression is achieved by tensor decomposition and trained quantization of the time-series feature-based spatio-temporal model. We have implemented DEEPEYE on an ARM-core based IOT board with only 2.4W power consumption. Using the video datasets MOMENTS and UCF11 as benchmarks, DEEPEYE achieves a 228.1× model compression with only 0.47% mAP deduction; as well as 15k× parameter reduction yet 16.27% accuracy improvement.

Neural networks are widely used in many applications ranging from classification to control. While these networks are composed of simple arithmetic operations, they are challenging to formally verify for properties such as reachability due to the presence of nonlinear activation functions. In this paper, we make the observation that Lipschitz continuity of a neural network not only can play a major role in the construction of reachable sets for neural-network controlled systems but also can be systematically controlled during training of the neural network. We build on this observation to develop a novel verification-aware knowledge distillation framework that transfers the knowledge of a trained network to a new and easier-to-verify network. Experimental results show that our method can substantially improve reachability analysis of neural-network controlled systems for several state-of-the-art tools.

Robustness is a key requirement for deploying a machine learning (ML) based solution. When a solution involves a ML model whose robustness is not guaranteed, ensuring robustness of the solution might rely on continuous checking of the ML model for its validity after the solution is deployed in production. Using wafer image classification as an example, this paper introduces tensor-based methods that help improve robustness of a neural-network-based classification approach and facilitate its deployment. Experiment results based on data from a commercial product line are presented to explain the key ideas behind the tensor-based methods.

Deep Neural Networks (DNNs) are the algorithm of choice for image processing applications. DNNs present highly parallel workloads that lead to the emergence of custom hardware accelerators. Deep Learning (DL) models specialized in different tasks require a programmable custom hardware and a compiler/mapper to efficiently translate different DNNs into an efficient dataflow in the accelerator. The goal of this paper is to present a compiler for running DNNs on Snowflake, which is a programmable hardware accelerator that targets DNNs. The compiler correctly generates instructions for various DL models: AlexNet, VGG, ResNet and LightCNN9. Snowflake, with a varying number of processing units, was implemented on FPGA to measure the compiler and Snowflake performance properties upon scaling up. The system achieves 70 frames/s and 4.5 GB/s of off-chip memory bandwidth for AlexNet without linear layers on Xilinx’s Zynq-SoC XC7Z045 FPGA.

Steganography is the art of hiding a secret message inside a publicly visible carrier message. Ideally, it is done without modifying the carrier, and with minimal loss of information in the secret message. Recently, various deep learning based approaches to steganography have been applied to different message types. We propose a deep learning based technique to hide a source RGB image message inside finite length speech segments without perceptual loss. To achieve this, we train three neural networks; an encoding network to hide the message in the carrier, a decoding network to reconstruct the message from the carrier and an additional image enhancer network to further improve the reconstructed message. We also discuss future improvements to the algorithm proposed.

In recent times, Generative Adversarial Networks (GANs) have created a lot of buzz in the research community. GANs are formulated on the zero-sum game theory, where two neural nets compete against each other. The resultant deep model is capable of generating data similar to any data distribution provided. It utilizes the adversarial learning approach and is far more capable in learning features than the traditional machine learning models. This review focusses on the origin and evolution of GANs. Firstly, the traditional GAN is explored in terms of its structure and loss functions. Then come the common challenges of training GANs. Thirdly, the review dives into numerous GAN variants and explains their improvements. The review then lists the wide variety of applications and ends with the conclusion.

We propose a method to learn 3D deformable object categories from raw single-view images, without external supervision. The method is based on an autoencoder that factors each input image into depth, albedo, viewpoint and illumination. In order to disentangle these components without supervision, we use the fact that many object categories have, at least in principle, a symmetric structure. We show that reasoning about illumination allows us to exploit the underlying object symmetry even if the appearance is not symmetric due to shading. Furthermore, we model objects that are probably, but not certainly, symmetric by predicting a symmetry probability map, learned end-to-end with the other components of the model. Our experiments show that this method can recover very accurately the 3D shape of human faces, cat faces and cars from single-view images, without any supervision or a prior shape model. On benchmarks, we demonstrate superior accuracy compared to another method that uses supervision at the level of 2D image correspondences., CVPR 2020 Oral. Project page: https://elliottwu.com/projects/unsup3d/

The presence of microplastics (MPs) in the environment is becoming a problem for soils and seas, as well as for the food chain of animals and humans. The scientific community has been called upon to contribute to solving the problem and several papers have been published, especially in the last decade. The aim of this work is to carry out a bibliometric analysis of the scientific literature dedicated to the problem of MPs, highlighting its course over the years, and to identify the sectors to which the research could be profitably addressed. The VOSviewer software has been used to perform the analysis of the data in which specific maps were used to represent the network of the relationships among countries, journals, organizations, authors, and keywords related to the investigated topic and subtopics. The results of the survey demonstrated that during the investigated range of time, most attention has been paid to the individuation of the MPs, and to marine pollution, while a gap seems to exist in the possible advanced oxidation processes specifically addressing the degradation of MPs and their derivates.

Gesture recognition is one of the most popular tasks in computer vision, where convolutional neural networks (CNNs) based method has obtained the state-of-the-art performance. However, it is generally acknowledged that CNNs need a large amount of data to achieve such performance. Event Camera is a kind of biologically inspired event-based camera, which can keep the information of moving objects and remove the redundant background data. In this paper, we set up a rock-paper-scissor hand gesture recognition system based on FlowNet and Event Camera. Event camera is used to acquire event data. Then we propose an algorithm for the proposed gesture recognition. Specifically, FlowNet2.0 is employed to extract the motion representation of the pre-processed visual data, and a CNN classification network is applied to recognize the symbols extracted according to the motion representation. As a comparison, we also apply the rock-paper-scissor gesture recognition algorithm on traditional camera. The experimental results show that the proposed system based on Event Camera gets better performance, and to some extent, weaken the dependence on the training data. The whole system achieves 94.0\(\%\) out-of-sample accuracy and allows computation at up to 30 fps.

In this paper, we provide a simple and effective baseline for classifying both patents and papers to the well-established Cooperative Patent Classification (CPC). We propose a label-informative classifier based on the Wide & Deep structure, where the Wide part encodes string-level similarities between texts and labels, and the Deep part captures semantic-level similarities via non-linear transformations. Our model trains on millions of patents, and transfers to papers by developing distant-supervised training set and domain-specific features. Extensive experiments show that our model achieves comparable performance to the state-of-the-art model used in industry on both patents and papers. The output of this work should facilitate the searching, granting and filing of innovative ideas for patent examiners, attorneys and researchers.

Near infrared (NIR) spectroscopy is used in many applications to gather information about chemical composition of materials. For paper waste sorting, a small number of scores computed from NIR-spectra and assuming more or less unimodal clustered data, a pixel classifier can still be crafted by hand using knowledge about chemical properties and a reasonable amount of intuition. Additional information can be gained by visual data (VIS). However, it is not obvious what features, e.g. based on color, saturation, textured areas, are finally important for successfully separating the paper classes in feature space. Hence, a rigorous feature analysis becomes inevitable. We have chosen a generic machine-learning approach to successfully fuse NIR and VIS information. By exploiting a classification tree and a variety of additional visual features, we could increase the recognition rate to 78% for 11 classes, compared to 63% only using NIR scores. A modified feature ranking measure, which takes redundancies of features into account, allows us to analyze the importance of features and reduce them effectively. While some visual features like color saturation and hue showed to be important, some NIR scores could even be dropped. Finally, we generalize this approach to analyze raw NIR-spectra instead of score values and apply it to plastic waste sorting.

Large-scale machine learning problems are often solved using distributed optimization techniques by dividing tasks across multiple compute nodes. Due to variability in compute time amongst nodes, distributed systems suffer from slow nodes known as stragglers that can have a big impact on convergence speed. Recently, Anytime Minibatch (AMB) technique has been proposed to speed up convergence by exploiting work done by stragglers rather than avoiding them. In AMB, workers are given a fixed time to calculate gradients, followed by a fixed communication time to average gradients. We observe that nodes stay idle during communication time. In our work, we allow workers to compute gradients during communication time resulting in what is known as stale gradients due to gradients calculated on out-of-date parameters. Our simulation results show that such approach achieves up to 60% faster convergence in wall time.

River is a major source of water in Malaysia and one of the major threats to its sustainability is pollution. The existing methods for monitoring of water quality in rivers are manual monitoring and continuous monitoring. These methods are costly and less efficient. Hence, we propose a smart river monitoring system (SRMS) that uses unmanned aerial vehicles (UAVs) or drones and low power wide area (LPWA) communication technology. The Internet of Things (IoT) and data analytic are promising techniques which provide real-time monitoring and enhances efficiency. However, due to the span of river that needs to be monitored, conventional communication technology such as Wi-Fi, Zigbee, Bluetooth are not suitable. Hence, there is the need for LPWA communication technology. We discuss the application of LPWA and UAV for sustainability of rivers in Malaysia as a case study. Preliminary results show that the use of UAV will increase the efficiency of measuring the water quality parameters compared to manual monitoring method. Also, real-time monitoring enables us to study the changes in water quality. Finally, we provide future direction in the application of UAV and LPWA for sustainability in river.

This paper constructs a time-varying state space model to empirically explore the dynamic relationship between the environmental regulation and innovation by using Chinese paper-making industry data during the period of 1996–2013. The empirical results show that the environmental regulation has a significant positive effect on the technological innovation, and the change of this effect level follows a “W” curve over time, indicating that environmental regulation could promote innovation and the impact of regulation shows the periodicity and volatility. Lastly, this paper provided suggestions for government policy makers and business managers.

In the area of cartography and geographic information science, the center points of area features are related to many fields. The centroid is a conventional choice of center point of area feature. However, it is not suitable for features with a complex shape for the center point may be outside the area or not fit the visual center so well. This paper proposes a novel method to calculate the center point of area feature based on triangulation skeleton graph. This paper defines two kinds of centrality of vertices in skeleton graph according to the centrality theory in graph and network analysis. Through the measurement of vertices centrality, the center points of polygon area features are defined as the vertices with maximum centrality.

We present a paper abstract writing system based on an attentive neural sequence-to-sequence model that can take a title as input and automatically generate an abstract. We design a novel Writing-editing Network that can attend to both the title and the previously generated abstract drafts and then iteratively revise and polish the abstract. With two series of Turing tests, where the human judges are asked to distinguish the system-generated abstracts from human-written ones, our system passes Turing tests by junior domain experts at a rate up to 30% and by non-expert at a rate up to 80%., * Equal contribution. 6 pages. Accepted by ACL 2018; The code and dataset are available at https://github.com/EagleW/Writing-editing-Network

In this attractive paper (Battiato et al. in Transp Porous Media, 2019), the authors review different methods of upscaling heterogeneous media descriptions to continuous macroscopic equivalent descriptions. I would like to introduce some comments to complete the presentation of the homogenisation theory in Section 7.

[Departement_IRSTEA]Ecotechnologies [TR1_IRSTEA]TED [Axe_IRSTEA]Co TED-SOWASTE et TED-EPURE; International audience; Increasingly stringent effluent limits and an expanding scope of model system boundaries beyond activated sludge has led to new modelling objectives and consequently to new and often more detailed modelling concepts. Nearly three decades after the publication of Activated Sludge Model No1 (ASM1), the authors believe it is time to re-evaluate wastewater characterisation procedures and targets. The present position paper gives a brief overview of state-of-the-art methods and discusses newly developed measurement techniques on a conceptual level. Potential future paths are presented including on-line instrumentation, promising measuring techniques, and mathematical solutions to fractionation problems. This is accompanied by a discussion on standardisation needs to increase modelling efficiency in our industry.

The large number of user-generated videos uploaded on to the Internet everyday has led to many commercial video search engines, which mainly rely on text metadata for search. However, metadata is often lacking for user-generated videos, thus these videos are unsearchable by current search engines. Therefore, content-based video retrieval (CBVR) tackles this metadata-scarcity problem by directly analyzing the visual and audio streams of each video. CBVR encompasses multiple research topics, including low-level feature design, feature fusion, semantic detector training and video search/reranking. We present novel strategies in these topics to enhance CBVR in both accuracy and speed under different query inputs, including pure textual queries and query by video examples. Our proposed strategies have been incorporated into our submission for the TRECVID 2014 Multimedia Event Detection evaluation, where our system outperformed other submissions in both text queries and video example queries, thus demonstrating the effectiveness of our proposed approaches.