Traditional Computer Research Articles

Intellectualized Collection and Analysis of Multimedia Computer Teaching Based on the Fusion of Audio-visual Information

In the post-epidemic era, online classrooms and multimedia computers are increasingly connected. In order to efficiently conduct online teaching and answer questions after class, this paper intelligently collects and analyzes current multimedia computer teaching materials based on the latest audio-visual information fusion algorithm. First, the spatial frequency SF of the online classroom is selected as the input excitation of audio-visual information, and the direction information at the pixel is fixed as the link strength slice relationship of PCNN. Compared with traditional computer teaching, this method has many advantages. Multimedia electronic classrooms are very suitable for computer teaching, and the use of multimedia electronic classrooms can diversify the forms of computer teaching. Then, the source image is decomposed in a two-dimensional experience mode, and the obtained residual images are fused to complete the intelligent collection and analysis of the characteristics of teaching information; the function design is carried out under the framework of learning activities, and the functional upgrade of technical products requires the teaching and learning.

Generative Adversarial Networks (GAN) and HDFS-Based Realtime Traffic Forecasting System Using CCTV Surveillance

The most crucial component of any smart city traffic management system is traffic flow prediction. It can assist a driver in selecting the most efficient route to their destination. The digitalization of closed-circuit television (CCTV) systems has resulted in more effective and capable surveillance imaging systems for security applications. The number of automobiles on the world’s highways has steadily increased in recent decades. However, road capacity has not developed at the same rate, resulting in significantly increasing congestion. The model learning mechanism cannot be guided or improved by prior domain knowledge of real-world problems. In reality, symmetrical features are common in many real-world research objects. To mitigate this severe situation, the researchers chose adaptive traffic management to make intelligent and efficient use of the current infrastructure. Data grow exponentially and become a complex item that must be managed. Unstructured data are a subset of big data that are difficult to process and have volatile properties. CCTV cameras are used in traffic management to monitor a specific point on the roadway. CCTV generates unstructured data in the form of images and videos. Because of the data’s intricacy, these data are challenging to process. This study proposes using big data analytics to transform real-time unstructured data from CCTV into information that can be shown on a web dashboard. As a Hadoop-based architectural stack that can serve as the ICT backbone for managing unstructured data efficiently, the Hadoop Distributed File System (HDFS) stores several sorts of data using the Hadoop file storage system, a high-performance integrated virtual environment (HIVE) tables, and non-relational storage. Traditional computer vision algorithms are incapable of processing such massive amounts of visual data collected in real-time. However, the inferiority of traffic data and the quality of unit information are always symmetrical phenomena. As a result, there is a need for big data analytics with machine learning, which entails processing and analyzing vast amounts of visual data, such as photographs or videos, to uncover semantic patterns that may be interpreted. As a result, smart cities require a more accurate traffic flow prediction system. In comparison to other recent methods applied to the dataset, the proposed method achieved the highest accuracy of 98.21%. In this study, we look at the construction of a secure CCTV strategy that predicts traffic from CCTV surveillance using real-time traffic prediction analysis with generative adversarial networks (GAN) and HDFS.

Fog computing security: A review

AbstractFog computing, also known as edge computing, is a decentralized computing architecture that brings computing and data storage closer to the users and devices that need it. It offers several advantages over traditional cloud computing, such as lower latency, improved reliability, and enhanced security. As the Internet of Things continues to grow, the demand for fog computing is also increasing, making it an important topic for research and development. However, the deployment of fog computing also brings new technical challenges and security risks. For example, fog nodes are often deployed in resource‐constrained environments and are exposed to potential security threats, such as malware and attacks on devices connected to the network. In addition, the decentralized nature of fog computing creates new challenges in terms of privacy, security, and data management. This survey aims to address these technical challenges and research gaps in the field of fog computing security. It provides an overview of the current state of fog computing and its security challenges, and identifies key areas for future research. The survey also highlights the importance of fog computing security and the need for continued investment in this area in order to fully realize the potential of this promising technology.

A Novel Data Partitioning Method for Active Privacy Protection Applied to Medical Records

In recent years, cloud computing has attracted extensive attention from industry and academia due to its convenience and ubiquity. As a new Internet-based IT service model, cloud computing has brought revolutionary changes to traditional computing and storage services. More and more individual users and enterprises are willing to deploy their own data and applications on the cloud platform, but the accompanying security issues have also become an obstacle to the development of cloud computing. Multi-tenancy and virtualization technologies are the main reasons why cloud computing faces many security problems. Through the virtualization of storage resources, multi-tenant data are generally stored as shared physical storage resources. To distinguish the data of different tenants, labels are generally used to distinguish them. However, this simple label cannot resist the attack of a potential malicious tenant, and data still has the risk of leakage. Based on this, this paper proposed a data partitioning method in a multi-tenant scenario to prevent privacy leakage of user data. We demonstrate the use of the proposed approach in protecting patient data in medical records in health informatics. Experiments show that the proposed algorithm can partition the attributes more fine-grained and effectively protect the sensitive information in the data.

Deep Reinforcement Learning Based Computation Offloading in UAV-Assisted Edge Computing

Traditional multi-access edge computing (MEC) often has difficulty processing large amounts of data in the face of high computationally intensive tasks, so it needs to offload policies to offload computation tasks to adjacent edge servers. The computation offloading problem is a mixed integer programming non-convex problem, and it is difficult to have a good solution. Meanwihle, the cost of deploying servers is often high when providing edge computing services in remote areas or some complex terrains. In this paper, the unmanned aerial vehicle (UAV) is introduced into the multi-access edge computing network, and a computation offloading method based on deep reinforcement learning in UAV-assisted multi-access edge computing network (DRCOM) is proposed. We use the UAV as the space base station of MEC, and it transforms computation task offloading problems of MEC into two sub-problems: find the optimal solution of whether each user’s device is offloaded through deep reinforcement learning; allocate resources. We compared our algorithm with other three offloading methods, i.e., LC, CO, and LRA. The maximum computation rate of our algorithm DRCOM is 142.38% higher than LC, 50.37% higher than CO, and 12.44% higher than LRA. The experimental results demonstrate that DRCOM greatly improves the computation rate.

Device‐mediated customer behaviour on the internet: A systematic literature review

AbstractThe proliferation of smartphones, tablets and other digital devices in addition to traditional computers has transformed the Internet into a device‐mediated environment. While these devices provide immediate access to similar Internet sources, they differ significantly in their characteristics, such as screen size, operation mode and context of use. As a consequence, behaviours on the Internet along the customer journey vary substantially depending on the device used. To summarize the fast‐growing body of research on device‐mediated customer behaviour, a systematic, framework‐based literature review of 59 articles from the last decade was conducted. Through an examination of the antecedents, decisions and outcomes investigated in the publications, the review presents a conceptual framework that highlights the relation between device characteristics, decision processes and behavioural outcomes. The review further summarizes the theories, contexts and methods employed in the studies and sets an extensive future research agenda. We found that the extant literature lacks comprehensive theories and clear definitions of digital devices in the omnichannel environment. Furthermore, existing findings should be generalized for other contexts (e.g. industries and countries) and validated via the introduction of other research designs and methods. The understanding of device‐mediated behaviour and the consequently arising marketing measures remains scarce. Thus, this review advances the comprehension of customer behaviour on the Internet and provides researchers and practitioners with information on the implications for customer experience and omnichannel management.

Classification of cassava leaf diseases using deep Gaussian transfer learning model

AbstractIn Sub‐Saharan Africa, experts visually examine the plants and look for disease symptoms on the leaves to diagnose cassava diseases, a subjective method. Machine learning algorithms have been employed to quickly identify and classify crop diseases. In this study, we propose a model that integrates a transfer learning approach with a deep Gaussian convolutional neural network model. In this study, two pre‐trained transfer learning models were used, that is, MobileNet V2 and VGG16, together with three different kernels: a hybrid kernel (a product of a squared exponential kernel and a rational quadratic kernel), a squared exponential kernel, and a rational quadratic kernel. In experiments using MobileNet V2 and the three kernels, the hybrid kernel performed better, with an accuracy of 90.11%, compared to 86.03% and 85.14% for the squared exponential kernel and a rational quadratic kernel, respectively. Additionally, experiments using VGG16 and the three kernels showed that the hybrid kernel performed better, with an accuracy of 88.63%, compared to the squared exponential kernel's accuracy of 84.62% and the rational quadratic kernel's accuracy of 83.95%, respectively. All the experiments were done using a traditional computer with no access to GPU and this was the major limitation of the study.

Data Extraction and Intelligent Analysis of Aesthetic Laws Accepted by VR Digital Media Art on Film Art Based on NVIDIA GPU Hardware

On the premise of analyzing traditional computer VR technology, this paper explores the application value of VR technology based on NVIDIA GPU hardware in the design of digital media systems, and proposes corresponding application measures to further improve the practical application capability of VR technology. These technologies integrate the achievements of modern high-performance computer systems, artificial intelligence, computer graphics, human-machine interfaces, stereoscopic images, stereophonic sound, measurement control, simulation simulation and other technologies. Analytical models and multi-grained formal conceptual analysis.

Comparative study on different energy-saving plans using water-side economizer to retrofit the computer room air conditioning system

An energy efficient, environmentally friendly and economical retrofit plan of traditional computer room air conditioning system is necessary to reduce energy consumption of data centers. In this study, two energy-saving retrofit plans, a computer room air handler system with water-side economizer (Plan 1) and a loop thermosyphon system with water-side economizer (Plan 2), are compared with an air-cooled computer room air conditioning system (existing cooling system) in a data center. To analyze and compare the energy efficiency, environmental and economic potential, energy models of the existing cooling system and retrofit plans are established. Under different set points (including supply chilled water temperature and supply air temperature) and 2.08–6 kW average electric power of racks, the free cooling periods of Plan 1 and Plan 2 reach 3878–5867 h and 4541–7661 h, respectively, which lead to the annual energy consumption of Plan 1 and Plan 2 decreases by 27.6–47.3% and 53.4–63.6%, respectively. Due to the lower annual energy consumption of main equipment (including chillers, chilled water pumps, cooling water pumps and terminal equipment), Power Usage Effectiveness of Plan 2 is 0.049–0.090 lower than that of Plan 1, and Global Coefficient of Performance of Plan 2 is 23.3–67.3% higher than that of Plan 1. Furthermore, the annual CO2 emissions of Plan 1 and Plan 2 are 337.3–557.1 tons and 599.8–1106.9 tons less than those of the existing cooling system at different operating stages respectively. Finally, the dynamic payback periods of Plan 1 and Plan 2 are 4.11–5.42 years and 1.64–3.57 years.

TSDNet: A New Multiscale Texture Surface Defect Detection Model

Industrial defect detection methods based on deep learning can reduce the cost of traditional manual quality inspection, improve the accuracy and efficiency of detection, and are widely used in industrial fields. Traditional computer defect detection methods focus on manual features and require a large amount of defect data, which has some limitations. This paper proposes a texture surface defect detection method based on convolutional neural network and wavelet analysis: TSDNet. The approach combines wavelet analysis with patch extraction, which can detect and locate many defects in a complex texture background; a patch extraction method based on random windows is proposed, which can quickly and effectively extract defective patches; and a judgment strategy based on a sliding window is proposed to improve the robustness of CNN. Our method can achieve excellent detection accuracy on DAGM 2007, a micro-surface defect database and KolektorSDD dataset, and can find the defect location accurately. The results show that in the complex texture background, the method can obtain high defect detection accuracy with only a small amount of training data and can accurately locate the defect position.

Let’s Unleash the Network Judgment: A Self-Supervised Approach for Cloud Image Analysis

Abstract Accurate cloud type identification and coverage analysis are crucial in understanding the Earth’s radiative budget. Traditional computer vision methods rely on low-level visual features of clouds for estimating cloud coverage or sky conditions. Several handcrafted approaches have been proposed; however, scope for improvement still exists. Newer deep neural networks (DNNs) have demonstrated superior performance for cloud segmentation and categorization. These methods, however, need expert engineering intervention in the preprocessing steps—in the traditional methods—or human assistance in assigning cloud or clear sky labels to a pixel for training DNNs. Such human mediation imposes considerable time and labor costs. We present the application of a new self-supervised learning approach to autonomously extract relevant features from sky images captured by ground-based cameras, for the classification and segmentation of clouds. We evaluate a joint embedding architecture that uses self-knowledge distillation plus regularization. We use two datasets to demonstrate the network’s ability to classify and segment sky images—one with ∼ 85,000 images collected from our ground-based camera and another with 400 labeled images from the WSISEG database. We find that this approach can discriminate full-sky images based on cloud coverage, diurnal variation, and cloud base height. Furthermore, it semantically segments the cloud areas without labels. The approach shows competitive performance in all tested tasks, suggesting a new alternative for cloud characterization.

Compute in-Memory with Non-Volatile Elements for Neural Networks: A Review from a Co-Design Perspective.

Deep learning hasbecome ubiquitous, touching daily lives across the globe. Today, traditional computer architectures are stressed to their limits in efficiently executing the growing complexity of data and models. Compute-in-memory (CIM) can potentially play an important role in developing efficient hardware solutions that reduce data movement from compute-unit to memory, known as the von Neumann bottleneck. At its heart is a cross-bar architecture with nodal non-volatile-memory elements that performs an analog multiply-and-accumulate operation, enabling the matrix-vector-multiplications repeatedly used in all neural network workloads. The memory materials can significantly influence final system-level characteristics and chip performance, including speed, power, and classification accuracy. With an over-arching co-design viewpoint, this review assesses the use of cross-bar based CIM for neural networks, connecting the material properties and the associated design constraints and demands to application, architecture, and performance. Both digital and analog memory are considered, assessing the status for training and inference, and providing metrics for the collective set of properties non-volatile memory materials will need to demonstrate for a successful CIM technology.

Acceleration of Wheel Factoring Techniques

The efficiency with which an integer may be factored into its prime factors determines several public key cryptosystems’ security in use today. Although there is a quantum-based technique with a polynomial time for integer factoring, on a traditional computer, there is no polynomial time algorithm. We investigate how to enhance the wheel factoring technique in this paper. Current wheel factorization algorithms rely on a very restricted set of prime integers as a base. In this study, we intend to adapt this notion to rely on a greater number of prime integers, resulting in a considerable improvement in the execution time. The experiments on composite numbers n reveal that the proposed algorithm improves on the existing wheel factoring algorithm by about 75%.

Art Installation Design and Algorithm Research Oriented to Heterogeneous Computing Architecture and Particle Swarm Algorithm

Traditional high-performance computers generally use commercial general-purpose processors. When constructing large-scale parallel computing systems, they will face many challenges in system efficiency, power consumption, system maintenance, and cost. Heterogeneous architectures have begun to become the key to constructing supercomputer systems. In order to improve the optimization efficiency of the particle swarm algorithm, based on the simplified particle swarm algorithm, an improved strategy for fusion of population information is proposed. Only using the optimal position of the individual particle and the optimal position of the population to update the particle position makes the algorithm description simpler, construct a random term that depends on the population information to increase the diversity of the population, and design an adaptive control function equation to balance the algorithm's global detection and local detection. This article aims to study the design and algorithm of art installations for heterogeneous computing architecture and particle swarm algorithm. Aiming at the two heterogeneous system components of GPU and FPGA in heterogeneous computing systems to solve the dual heterogeneous problem between various computing components and applications, a task flow model for heterogeneous computing systems is proposed, which adopts heterogeneous systems. The structure focuses on the deterministic and nondeterministic calculation methods of particle swarm algorithm and carries out parallel research algorithms of particle transport data to analyze the design of art installations. The experimental results show that by making full use of the GPU hardware storage structure to improve memory access speed and reduce cache failure, the combination of particle swarm algorithm and heterogeneous computing system enables the calculation of the carrier components to obtain 3 times the acceleration effect; through the energy analysis model, Heterogeneous computing system components can obtain at least 1.5 times performance improvement.

Are virtual reality serious video games more effective than web video games?

Virtual Reality is widely used for many educational purposes. However, despite the increasing trend in the use of this technology, it is not yet clear whether it is more effective for learning than other technologies such as traditional computer video games. This paper presents a serious video game to learn Scrum, a methodology widely used in the software industry. The game is available in mobile Virtual Reality and Web (using WebGL) formats. Both versions of the game are compared in terms of knowledge acquisition and motivation enhancement through a robust empirical study involving 289 students and instruments such as pre-post tests and a questionnaire. The obtained results suggest that both formats of the game are useful for acquiring knowledge and promoting aspects such as fun, motivation, and engagement. The results also suggest, strikingly, that there is no difference in terms of learning effectiveness between the two versions of the game.

Output-Feedback Flexible Performance-Based Safe Formation for Underactuated Unmanned Surface Vehicles

This article studies a cooperative time-varying formation of underactuated unmanned surface vehicles (USVs) with position-heading measurements, following the philosophies of (i) balance operation between performance constraints and safety requirements including collision avoidance and connectivity maintenance, and (ii) concise finite-time control design based on the integrated finite-time extended state observer (IFTESO) to facilitate the practical implementation. At first, to accomplish the guidance design of networked vehicles in the complex ocean environment and guarantee the performance of cooperative tracking errors, a flexible performance-based control scheme is investigated by feeding the potential effects of safety requirements into the modification signals of the user-specified performance functions. Then, without the application of cumbersome nonlinear tracking differentiators and backstepping techniques, a output-feedback IFTESO-based method provides a concise solution with a convenient design by taking advantage of homogeneous system theories, which further reduces the traditional computation consumption. Finally, the performance of the proposed scheme is validated by numerical simulations.

Automatic measuring of finger joint space width on hand radiograph using deep learning and conventional computer vision methods

Hand osteoarthritis (OA) severity can be assessed visually through radiographs using semi-quantitative grading systems. However, these grading systems are subjective and cannot distinguish minor differences. Joint space width (JSW) compensates for these disadvantages, as it quantifies the severity of OA by accurately measuring the distances between joint bones. Current methods used to assess JSW require users’ interaction to identify the joints and delineate the initial joint boundary, which is time-consuming. To automate this process and offer a robust measurement for JSW, we proposed two novel methods to measure JSW: (1) The segmentation-based (SEG) method, which uses traditional computer vision techniques to calculate JSW; (2) The regression-based (REG) method, which is a deep learning approach employing a modified VGG-19 network to predict JSW. On a dataset with 3591 hand radiographs, 10,845 DIP joints were cut as ROI and served as input to the SEG and REG methods. The bone masks of the ROI images generated by a U-Net model were sent as input in addition to the ROIs. The ground truth of JSW was labeled by a trained research assistant using a semi-automatic tool. Compared with the ground truth, the REG method achieved a correlation coefficient (r) of 0.88 and a mean square error (MSE) of 0.02 mm on the testing set; the SEG method achieved a correlation coefficient of 0.42 and an MSE of 0.15 mm. Results show the REG method has promising performance in JSW measurement and, in general, Deep Learning approaches can facilitate the automatic quantification of distance features in medical images.

An improved method for computing acoustic ray incident angle based on secant method

Constant gradient acoustic ray tracking is usually used in high-precision underwater acoustic positioning. However, it needs to calculate the incident angle of the acoustic ray, and the traditional incident angle computation method is very inefficient. To resolve this problem, we propose an improved acoustic ray incident angle iterative computation method. It firstly constructs the mathematical model for incident angle by using the function between horizontal propagation distance and incident angle based on secant method. The initial value of the incident angle is then determined by the incident angle calculated from the previous epoch and the last iteration. Meanwhile, the root-containing interval of the incident angle is reduced according to experimental parameters and coordinate correction values. Finally, the incident angle calculated in the last iteration is directly used for acoustic ray tracking and positioning solution until the end, when the coordinate correction values reach a threshold. Experiment conducted on the simulation cases demonstrates that the positioning calculation time of the proposed method is 77.7%, 77.1% and 88.1% lower than traditional methods in order of the dichotomy, secant method, and Newton method, respectively. As for the real experiment of the South China Sea, the computation efficiency improvement of the proposed method is about 62.9%, 67.6% and 66.1% compared with the dichotomy, secant method, and Newton method respectively. These results suggest that the proposed method can significantly improve the positioning calculation efficiency without loss of positioning accuracy compared with conventional methods.

Sign language recognition from digital videos using feature pyramid network with detection transformer

Sign language recognition is one of the fundamental ways to assist deaf people to communicate with others. An accurate vision-based sign language recognition system using deep learning is a fundamental goal for many researchers. Deep convolutional neural networks have been extensively considered in the last few years, and a slew of architectures have been proposed. Recently, Vision Transformer and other Transformers have shown apparent advantages in object recognition compared to traditional computer vision models such as Faster R-CNN, YOLO, SSD, and other deep learning models. In this paper, we propose a Vision Transformer-based sign language recognition method called DETR (Detection Transformer), aiming to improve the current state-of-the-art sign language recognition accuracy. The DETR method proposed in this paper is able to recognize sign language from digital videos with a high accuracy using a new deep learning model ResNet152 + FPN (i.e., Feature Pyramid Network), which is based on Detection Transformer. Our experiments show that the method has excellent potential for improving sign language recognition accuracy. For instance, our newly proposed net ResNet152 + FPN is able to enhance the detection accuracy up to 1.70% on the test dataset of sign language compared to the standard Detection Transformer models. Besides, an overall accuracy 96.45% was attained by using the proposed method.

Assessment of plant disease detection by deep learning

Plant disease and pest detection machines were originally used in agriculture and have, to some extent, replaced traditional visual identification. Plant diseases and pests are important determinants of plant productivity and quality. Plant diseases and pests can be identified using digital image processing. According to the difference in the structure of the network, this study presents research on the detection of plant diseases and pests based on three aspects of the classification network, detection network, and segmentation network in recent years, and summarizes the advantages and disadvantages of each method. A common data set is introduced and the results of existing studies are compared. This study discusses possible problems in the practical application of plant disease and pest detection based on deep learning. Conventional image processing algorithms or manual descriptive design and classifiers are often used for traditional computer vision-based plant disease and pest detection. This method usually uses various characteristics of plant diseases and pests to create an image layout and selects a useful light source and shooting angle to produce evenly lit images. The purpose of this work is to identify a group of pests and diseases of domestic and garden plants using a mobile application and display the final result on the screen of a mobile device. In this work, data from 38 different classes were used, including diseased and healthy leaf images of 13 plants from plantVillage. In the experiment, Inception v3 tends to consistently improve accuracy with an increasing number of epochs with no sign of overfitting and performance degradation. Keras with Theano backend used to teach architectures