2025-07-13
Shruti Jaiswal, Himani Bansal, Shiv Naresh Shivhare, Gulshan Shrivastava
The popularity of Internet of Things (IoT) devices has surged due to their applications in diverse areas such as e-Health, smart vehicles, and smart cities. However, the rapid deployment of these devices has led to an exponential increase in security attacks targeting IoT systems, making security a prime concern for the community. Securing IoT-based systems is challenging because the devices involved are often resource-constrained. Providing security to these systems requires a thorough understanding of their specific security needs, along with a systematic security engineering approach. Previous research lacks a systematic methodology for identifying and implementing security requirements. Therefore, there is a growing demand for a structured approach to identify security requirements, select appropriate algorithms, and ensure their effective implementation. While existing studies have extensively explored IoT security threats, they fall short of offering a structured method to comprehensively address these threats. This paper proposes a comprehensive security engineering framework that systematically identifies security threats by analyzing assets present over various layers of IoT system, considering their diverse roles. It includes creating repositories to identify potential vulnerabilities and applicable threats. Once threats are identified, they are evaluated for their severity level based on risk analysis. Following this, the framework focuses on designing the security solutions, where we proposed to add two new security services namely trust and data freshness besides the existing security services, algorithms are selected to mitigate threats by considering the domain and constraints of the devices involved. Ultimately, the security of the entire system is validated to ensure robustness. Throughout this process, we have developed comprehensive repositories for asset management, vulnerability-threat mapping, and algorithm-threat matching to help identify and analyze security needs and recommend algorithms for implementation.
2025-07-13
Lili Kong, Chunqiu Yi
High processing needs and latency make it difficult to simulate motor drive systems in distributed environments, which affect accuracy and real-time performance. Optimizing motor control and cutting energy use require effective modeling. Conventional simulation techniques have trouble scaling up and down, frequently demanding large amounts of resources and being unable to adjust to changing load circumstances, which leads to sluggish or imprecise simulations. This method improves scalability and lowers latency by dynamically adjusting computing loads in a distributed system through the use of adaptive algorithms. It improves the accuracy and efficiency of simulation by utilizing real-time adaption and parallel processing. The purpose of this work was to suggest distributed systems for motor drive system simulation analysis utilizing adaptive algorithms. Initially, the dataset was gathered from a test bench-mounted momentum permanent magnet synchronous motor (PMSM) in a three-phase system motor vehicle. The exponentially weighted moving standard deviation (EWMS) utilized in standardized data process representations for training. We proposed the Adaptive Controller with dynamic fuzzy system ensemble (AC-DMFSE) for distributed systems for simulation analysis of motor drive systems. To optimize motor performance in dynamic situations, adaptive techniques are used, such as fuzzy logic-based optimization and model predictive control. Our test findings show that the suggested distributed technique reduced simulation times and MSE while enhancing the accuracy of system performance evaluation. The foundation for scalable and effective motor drive system simulations is laid by this work, which also offers insightful information for improving the systems’ performance in practical applications.
2025-07-13
Zezhi Ye, Chenghai Yu, Zhilong Lu
Software reliability prediction (SDP) theory is crucial for balancing software value and assessing efficiency. Traditional defect prediction relies on static code metrics for machine learning, but these handcrafted features fail to capture the code’s syntactic structure and semantic information. In order to further predict the defects of the software, the Abstract Syntax Tree (AST) of the program was parsed on the basis of the metric data, and extracted as feature vectors, and the data was encoded by dictionary mapping and word embedding as the input of Convolutional Neural Network (CNN). On this basis, the Long Short-Term Memory network (LSTM) and Multi-head attention mechanism were used to further optimize the network, and the particle swarm optimization (PSO) was used to select the hyperparameters of the model, and the defects were predicted by using the model. The results show that the model can learn syntax and semantic features well, and the estimation accuracy is higher and the bias is smaller.
2025-07-13
Weijuan An, Li Shen, Yali Yuan
With the widespread application and development of internet technology in public education scenarios in China, the application of deep learning technology on online learning platforms is becoming increasingly widespread. This study aims to address the difficulty in determining students’ cognitive state under the current network learning mode, and proposes a face recognition algorithm for student cognitive state detection using multitask convolutional neural network image recognition technology. At the same time, research is conducted on extracting two-dimensional feature points of facial color images through cascading regression tree localization methods. In the practical application experiments of the method, the research method can effectively detect images with facial offset angles greater than 15° for students, and the cognitive state of online learning can be analyzed from the frequency detection of students’ blinking and yawning. From the results of image simulation experiments, it can be seen that this study proposes a cascaded regression tree localization optimization multitask convolutional neural network face recognition method, which has the highest image recognition accuracy of 86%, a recall rate of 0.85, and an f1 value of 0.855. The experimental results show that online learning state recognition based on image analysis can effectively monitor abnormal states during students’ learning process, improve students’ online learning efficiency, and provide necessary student state information support for teachers, promoting the improvement of online teaching quality.
2025-07-13
Yanxiang Zhang
Color functions as a distinct type of ideographic symbol in animation for film and television, playing a crucial role in enhancing visual narratives and conveying emotions. In different types of animation, such as fantasy, horror, or children’s genres, color language influences audience perception and can convey meanings beyond the capabilities of image language alone. For instance, bright colors may symbolize innocence in children’s animation, while darker shades may evoke tension or fear in horror. However, current approaches to representing color in animation often fail to capture its full semantic richness and ideographic potential. Existing methods primarily focus on image-based analysis, overlooking the deeper layers of meaning encoded in color language. In this paper, we address these gaps by utilizing semantic segmentation techniques to combine the three modalities of color, content, and text to establish a consistent representation of color language in animation. We propose a method for semantic segmentation of color-depth (RGB-D) images using two-stream weighted Gabor convolutional network fusion. A weighted Gabor orientation filter builds a deep convolutional network (DCN) capable of extracting feature information adaptive to changes in orientation and scale, allowing for orientation- and scale-invariant features. Dual-stream picture features - color and depth - are extracted using a broad residual-weighted Gabor convolutional network and then combined into a lightweight feature extraction network. To evaluate the ideographic functions of color language quantitatively, we conducted extensive experiments using open databases. Our proposed method outperforms existing RGB-D picture semantic segmentation algorithms, demonstrating its effectiveness in representing color language in animation.
2025-07-13
Shilpa, Tarandeep Kaur, Rachit Garg, Deepak Prashar, Sudan Jha, Sultan Ahmad, Jabeen Nazeer
With the advancements of the technologies, healthcare industry has become more digitized, data driven, patient-centric, innovative as well as collaborative. It has become possible to access and share patient information globally irrespective of time and locations. The traditional healthcare facilities have proved inadequate in facilitating better patient treatment and services as per the patient requirements. The modern healthcare provisioning involves the use of digital technologies and operations and manifest digitized healthcare facilities. The digital technologies such as cloud computing, Internet of Things and many other supports the facilitation of smart healthcare utilities and services. This paper describes how cloud computing and IoT deliver healthcare-as-a-service. Further, the various proposed healthcare frameworks integrated with healthcare field have been discussed such as based on technologies implemented, offered services, and focused disease(s). Also, this survey enlightens the comparisons of existing frameworks based on technologies used and based on the focused methodologies.
2025-07-13
Natesh Mahadev, Sowmya V L, Shankar R, Anitha Premkumar, Syed Salim, Rajesh Natarajan
The popularity of smart home systems has greatly increased over the past several years, which provide convenience, automation, and control over various areas of daily life at home. However, these technologies still have issues with scalable user experience, mainly because there aren’t any engaging and intuitive user interfaces. Incorporating pervasive virtual reality (VR) interfaces into smart home systems is the unique strategy proposed in this study to improve scalable user experience. The study’s goal is to use VR technology to develop engaging and simple user interfaces that seamlessly integrate with the actual surroundings of a smart home. Users can engage with their smart home appliances and services through a mixed-reality experience in which virtual items and information are seamlessly incorporated into their environment by donning portable VR headsets. This application considers various difficulties related to developing Ambient Assisted Living (AAL) solutions, including unique characteristics of each end user, appliance, technology, deployment, and data-sharing problems. The Smart Home Systems take advantage of Semantic Web technologies integration abilities and their capacity facing represent significant information into legal models. A virtual reality application called Smart Home Systems allows for residential settings’ setup and customization in AAL systems. Additionally, it effectively uses VR technology to streamline the creation of specialized AAL settings. The application and underlying framework were evaluated for each through two scenarios: designing a home setting specifically for specific scalable user categories.
2025-07-13
Xiong Jian, Rajamohan Parthasarathy, Yinqing Tang, Binwen Huang
This paper proposes the design of electronic medical record system supported by multimedia communication based on blockchain technology. Blockchain technology ensures the secure storage and sharing of patient information through distributed ledger and smart contract algorithm. In this system, smart contracts are used to automatically execute cross-institutional data access control and audit functions to ensure the transparency and compliance of data access. At the same time, this paper introduces multimedia communication technology to support the efficient transmission and sharing of medical data, especially in diagnostic images, videos and voice. Simulation results show that the electronic medical record system based on blockchain has significantly improved data processing efficiency, security and reliability compared with traditional systems.
2025-07-13
Soumya K N, Praveen N
In the recent past, so many chronic diseases have been emerging and spreading in the world and even in the developing and advanced countries as well. One of such serious chronic diseases is Diabetes Mellitus that covers and impacts the health of people from early age. Nevertheless, the available Machine Learning (ML) and Deep Learning (DL) approaches are unable to provide good predictions in patients relating to diabetes. In addition, this study evaluated the proposed pre-processing procedure on large datasets for diabetes prediction that contained outliner detection and removal, missing values imputation, and standardization, to improve diabetes ascertainment. This research evaluated the proposed pre-processing procedure on a large set of data by outlier identification and removal, missing values imputation and data standardization were done to improve diabetes forecast. To ensure rapid and accurate classification of diabetes, the researchers employed and initialized an Artificial Neural Network (ANN). Data was gathered from the PIMA Dataset and North California State University (NCSU). Following this, Bivariate filter was applied to sort out features which were relevant. The selected features were subsequently subjected to Pearson correlation towards feature set refinement considering a threshold below which features were eliminated and only the most effective features selected. From the results it was evident that the proposed approach was significantly better than the existing methods in terms of accuracy as it achieved a classification accuracy of around 93% as opposed to the other methods.
2025-07-13
Ying Huang
Food consumed by humans is becoming more and more customized to fit each person’s unique demands, with a vast array of product labels readily available. As a result, many companies are beginning to concentrate on enhancing the practicality of contemporary packaging. Throughout the lifetime of a user-product engagement, sensory paradigms and emotional responses may shift. Traditional product packaging layout is largely based on the designer’s emotional imagination and prior events; however, it is limited by uncontrollable content and a lack of expert advice; most earlier studies involving mental analysis of images focused on predicting the most prevalent viewers feelings. There are situations when an image’s general impression is insufficient for practical purposes since the emotions it arouses are very subjective and differ from viewer to viewer. The proposed methodology uses Genetic Algorithm based Multi-Layer Ant Colony Optimization for analysing visual perception and emotion perception to identify the senses of human being. A significant set of images called Image-Emotion-Social-Net is utilized to assess categorized and multivariate attitude representations. The collection, which comes from Flickr, has more than a million photos uploaded by more than 9,000 members. The results of this dataset’s research indicate that the suggested approach performs better in personalized emotional identification than several contemporary methods. In comparison to other current methods, the experimental findings demonstrate that the suggested approach obtains a high packaging layout excellence rate of 95.1%, a performance success rate of 98.5%, and a mean square error rate of 1.5%.
2025-07-13
Hanyang Cui, Xinyu Yang
In order to solve the problems of long training content mastery time, poor training effect, and high cost in traditional training systems, the author proposes the use of computer big data technology in the research of intelligent algorithms for athlete training in university physical education. The author begins by gathering sports data from athletes, then utilizes a virtual reality perception interaction model to feed the data into the virtual environment generation unit. Data fusion is performed under the supervision of the simulation management module. Finally, combined with the motion behavior interpretation in the database and image card of the sports simulation training data unit, simulation 3D modeling is carried out in the 3D model processor according to user settings. The experimental results demonstrated that athletes trained with this system (the third group) had a markedly better understanding of the training content compared to those trained with the other two systems (the first and second groups) after the first week, achieving a mastery rate of 73.875%. As the duration of the training increased, all groups showed improved mastery of the content. This system enhances athletes’ training effectiveness while also reducing training costs, offering significant practical application value.