检索结果-内蒙古大学图书馆

Informatica (Slovenia) 2025年第14期49卷 63-78页

作者： Li, Yiming Xie, Tao Mei, Dongdong Department of Computer Science and Technology School of Computer Science and Engineering Ningxia Institute of Science and Technology Shizuishan753000 China

This study proposes a malicious code detection model DTL-MD based on deep transfer learning, which aims to improve the detection accuracy of existing methods in complex malicious code and data scarcity. In the feature extraction process, the weighted sum method of GIST and LBP features is used to combine the advantages of the two features. Online transfer learning is used to reduce the data distribution difference between the target domain and the source domain. The model uses ResNet50V2 as the backbone network and combines SimAM to enhance the feature extraction and representation capabilities. In addition, in order to further improve the robustness of detection, GAN is used to generate malicious code variants and expand the training data set. In the experiment, the public CICIDS 2017 data set is used for model training and testing. The performance test results show that when the threshold is 0.7, the accuracy of DTL-MD is 95.8% and the F1 score is 0.93. In a performance test involving 30,000 samples, the throughput of the DTL-MD model under Trojans, viruses, worms, and adware is 11, 12, 11, and 12 tasks/s, respectively, and the inference time is 211, 225, 239, and 234 samples/s, respectively. Compared with GAN, DTL-MD increases the throughput by about 10% and the inference speed by about 15%. The research aims to provide new ideas for improving the intelligence and automation level of malicious code detection technology, which has certain application value and practical significance. © 2025 Slovene Society Informatika. All rights reserved.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Dual strategy for single image denoising and generation using deep neural network

引用

Multimedia Tools and Applications 2025年第18期84卷 19255-19271页

作者： J, Ranjith Sarobin, Vergin Raja Bhargavi, Renta Chintala School of Computer Science and Engineering Vellore Institute of Technology Chennai India

Removing noise in the real-world scenario has been a daunting task in the field of natural language processing. Research has shown that Deep Neural Networks (DNN) have proven to be very useful in terms of noise generation in order to synthesize noise and clean image couples in order to train the deep denoiser architectures. In our work, we develop a method that denoises an image while dealing hand in hand with image generation tasks so as to facilitate the former and reduce the load on the denoiser. We use a probabilistic model by learning the joint distribution of the noisy and clean image couples. To be more specific, we use an approximation of the above-mentioned probabilistic distribution by using two different forms, one that can map clean images to noisy, hereafter referred to as the generator, and the vice versa, i.e., mapping a noisy image to a clean image hereafter referred to as the image denoiser. The advantage of this method is that as compared to traditional models where manual design of the images beforehand and deducing information on the noise assumption was critical, the learned probabilistic joint distribution now implicitly holds the required details related to the clean and noisy images. The main aim of the symmetric generator is to enhance the performance of the denoiser by generating synthetic noisy images from the same dataset itself. The evaluation of the denoiser was done using two evaluation metrics, peak signal to noise ratio (PSNR) and the structural similarity index (SSIM). The entire experiment was conducted using the publicly available Smartphone Image Denoising Dataset (SIDD). © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

OPTIMIZING CONSTRUCTION AND DEMOLITION WASTE COLLECTION SYSTEMS THROUGH ADVANCED SCHEDULING

Journal of Solid Waste Technology and Management

引用

Journal of Solid Waste technology and Management 2024年第2期50卷 555-566页

作者： Ramalakshmi, Shree Smiti HaitherAli, Hammadhu Sivagami, M. Anjali, G. School of Computer Science and Engineering Vellore Institute of Technology Chennai India School of Civil Engineering Vellore Institute of Technology Chennai India

Construction and demolition (C&D) waste management is challenging in urban areas due to the high volume of waste generated and widespread illegal dumping. City authorities are struggling with environmental, economic, and social issues due to the lack of a systematic waste management process. An inefficient collection system leads to the loss of valuable urban spaces and clogging of drains with debris. This study aims to develop a simplified algorithm for effective C&D waste collection and transportation systems that maximize efficiency and minimize cost. It analyses various scheduling concepts such as Dijkstra's, Bellman-Ford's, and Floyd-Warshall's algorithm to identify the most efficient route for the collection fleet. The uniqueness of the study lies in providing an integrated solution to the collection and transportation issues of C&D waste management by estimating the volume of debris from photographs and scheduling the collection fleet based on real-world constraints. Waste type, volume, transportation fleet, route and vehicle capacities are considered to aid in efficient resource allocation and utilization. Optimization process involves determining the optimal allocation of waste to the collection fleet, scheduling transportation routes, and assigning human resources. On verification with a real-world case, the model was found to reduce the number of trips and enhance resource utilization, resulting in an efficiency improvement of over 30%. Since waste collection constitutes a significant cost in any solid waste management process, proposed methodology would be beneficial for waste recycling facilities and local waste management authorities to optimize the process, minimize costs, maximize efficiency and reduce environmental footprint. © 2024 Widener University school of Civil engineering. All rights reserved.

关键词： Debris

来源：评论

学校读者我要写书评

暂无评论

A comprehensive survey of image compression methods: from prediction models to advanced techniques

引用

Multimedia Tools and Applications 2025年 1-39页

作者： Devadason, Joshua Rajah D, Aju School of Computer Science and Engineering Vellore Institute of Technology Vellore India

Images are used widely nowadays. Images are used in many fields such as medicine to terrain mapping. There is a need to compress the images and represent them in shorter form for effective transmission. Several techniques are reviewed in this paper. The first group of techniques is the data compression technique. The data compression methods are Huffman encoding, Lempel–Ziv Welch (LZW), arithmetic encoding, run length encoding, and Shannon-Fano encoding. The second group of methods is prediction, here the neighboring pixels are utilized in the prediction of the present pixel and the error is encoded using an entropy encoder. In wavelet-based techniques, the source image is transformed from the spatial domain to the wavelet domain, and the coefficients are encoded. In learning-based techniques, a neural network is used to model the image data for compression. Fractal encoding is a compression method that searches for similar blocks in an image and compresses the difference between the blocks. With an average bitrate of 4.87 bpp (bits per pixel) on the first 10 photos of the Digital Imaging and Communications in Medicine (DICOM) image collection, the results indicate that LZW is the top performing data compression technique among the data compression techniques. In prediction-based techniques, Context Adaptive Lossless Image Codec (CALIC) is the best performing with 3.5 bpp. The modern neural method, compressive autoencoder outperforms established methods such as Joint Photographic Experts Group (JPEG), by achieving a PSNR of 35 on the Kodak dataset compared with 32 PSNR on the same dataset at 1.0 bpp. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

Research on ride control strategy of electromagnetic suspension based on sliding mode algorithm

引用

Noise and Vibration Worldwide 2025年第1-2期56卷 25-37页

作者： Xiong, Xin Chen, JiaLing Wei, JinCheng Xu, Fei School of Automotive Engineering Yancheng Institute of Technology Yancheng China School of Automotive Engineering Jilin University Changchun China

This study proposes a main inner loop control technique to address the dynamic response of the electromagnetic suspension during vehicle driving. The genetic algorithm is employed to optimize the control parameters of the linear quadratic adjustment controller in the control main loop, with the aim of achieving the desired control force. In the control inner loop, a closed-loop sliding mode controller is designed based on the exponential approximation law to regulate the current. The Lyapunov function was utilized to assess the stability of the controller in terms of its theoretical feasibility. The simulation results of Matlab/Simulink show that under the given working conditions, compared with the passive suspension, sliding mode control and fuzzy sliding mode control for references, the root mean square value and peak value of the vertical acceleration of the body and the dynamic load of the tire with the improved sliding mode control are significantly reduced, the root mean square value of the dynamic deflection of the suspension was not improved. However, on the whole, the ride comfort of the vehicle has been improved. © The Author(s) 2024.

关键词： Sliding mode control

来源：评论

学校读者我要写书评

暂无评论

An enhanced framework for smart automated evaluations of answer scripts using NLP and deep learning methods

引用

Multimedia Tools and Applications 2025年第11期84卷 8491-8513页

作者： G, Mohanraj R.K, Nadesh M, Marimuthu V, Sathiyapriya School of Computer Science Engineering and Information Systems Vellore Institute of Technology Vellore632014 India School of Computer Science and Engineering Vellore Institute of Technology Chennai600127 India Knowledge Institute of Technology Salem637504 India

The manual process of evaluating answer scripts is strenuous. Evaluators use the answer key to assess the answers in the answer scripts. Advancements in technology and the introduction of new learning paradigms need automation of the evaluation process. This work aims to develop an enhanced novel hybrid framework that can evaluate answer scripts and automatically assign marks for different type of questions based on keywords, grammar, symbols, special keywords, and the given factors. First, the proposed system uses Optical Character Recognition (OCR) to convert image answer scripts into an editable text format. Second, the sentence transformers, the Natural Language Processing (NLP) technique flips the answer script and answers key texts into word embedding vectors. To find similarity measures, these vectors are matched using BERT encoding, spearmanś rank-order correlation, and fuzzy search. At last, the proposed model is trained using Deep Columnar Convolutional Neural Network (DCCNN) in the third step with MINST and Kaggle handwritten mathematical symbols and tested with the segmented mathematical equations to find the similarity. The performance of proposed model is measured using precision, recall, accuracy, and F1-score, and its gives highest accuracy of 93% and 95% when compared to the existing methodologies. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Optical character recognition

来源：评论

学校读者我要写书评

暂无评论

Agricultural Machinery Inter-regional Cooperative Job Scheduling Method based on An Improved SPBO Algorithm

引用

Journal of engineering Science and technology Review 2025年第2期18卷 186-194页

作者： Lu, Jianwei Fan, Xiaoling Li, Hui Song, Dianhui School of Information Engineering Changzhou Vocational Institute of Industry Technology Changzhou213164 China School of Management College of Humanities & Information Changchun University of Technology Changchun130122 China

Given the lack of scientific and reasonable optimization scheduling strategies, multimachine cross-regional operations often encounter high operation costs, low operation efficiency, and the inability to complete farmland tasks within the suitable operation time during the harvest season. To address this issue, the authors of this study constructed a multimachine cross-regional collaborative operation scheduling model under the constraint of operation time windows to minimize the transfer distance cost of agricultural machinery operations and the total scheduling cost. An improved standard particle swarm optimization (SPBO) algorithm was designed, a double-layer coding method was adopted to encode genes, and an individual repair strategy was designed during decoding to reduce the number of iterations. Three evolutionary operators, namely, strong disturbance, weak disturbance, and learning strategies, were introduced to enhance the algorithm’s local and global search capabilities. According to the priority rule of time windows, combine harvesters were allocated for operations in sequence, and a reoptimization strategy was adopted. An improved SPBO algorithm with the addition of worst removal and greedy insertion operations for orders was proposed as the scheduling optimization algorithm. Simulation experiments and comparative analysis verify that, the improved SPBO algorithm proposed in this study can effectively avoid local optimal problems, accelerate the global convergence speed, and avoid the waste or redundant use of resources, thereby reducing costs. The conclusion provides a new solution for the path planning problem of cross-regional collaborative operations of agricultural machinery. © 2025 school of Science, DUTH. All rights reserved.

关键词： Harvesters

来源：评论

学校读者我要写书评

暂无评论

Realizing the potential of Internet of Things (IoT) in Industrial applications

引用

Discover Internet of Things 2025年第1期5卷 1-16页

作者： Rathi, Bharat Thapaswi, S. Kambhampati, Meghana Jain, Vineet Akshay, P. Pandey, Trilok Nath Pradhan, Sunil Kumar School of Electronics Engineering Vellore Institute of Technology Chennai600127 India School of Computer Science and Engineering Vellore Institute of Technology Chennai600127 India

The Internet of Things (IoT), which enables seamless connectivity and effective data exchange between physical items and digital systems, has completely changed the way we interact with our surroundings. This study evaluates IoT applications in healthcare, transportation, agriculture, and smart cities, highlighting advancements, opportunities, and challenges. Using a systematic literature review and case study analysis, it identifies essential IoT components, including sensors, communication protocols, data analytics, and security methods. Key findings demonstrate IoT's potential to enhance healthcare through real-time monitoring and cost reduction, optimize transportation systems for resource efficiency, improve agricultural productivity via precision farming, and enable sustainable urban development in smart cities. However, challenges like data privacy, interoperability, and scalability remain critical barriers to widespread adoption. This study underscores the need for interdisciplinary collaboration among researchers, policymakers, and industry stakeholders to address these challenges and unlock IoT's full potential. It provides actionable insights and future directions for realizing a connected, efficient, and sustainable IoT-driven future. © The Author(s) 2025.

关键词： Precision agriculture

来源：评论

学校读者我要写书评

暂无评论

Analytical Formulation and Numerical Investigation of Electromechanical Dynamics in Circular Biconvex Diaphragm Capacitive Pressure Sensor

引用

IEEE Sensors Journal 2025年第11期25卷 18867-18875页

作者： Rukshana Bi, Gajula Motwani, Mayank Jindal, Sumit Kumar Vellore Institute of Technology School of Electronics Engineering Vellore632014 India Vellore Institute of Technology School of Computer Science and Engineering Vellore632014 India

Capacitive pressure sensors (CPSs) have attracted considerable interest due to their high sensitivity, low energy consumption, and potential for miniaturization, making them suitable for applications in automotive systems, consumer electronics, and industrial automation. The conventional designs of CPSs often face challenges such as limited sensitivity, mechanical instability, and difficulties in achieving increased sensitivity. This study presents a biconvex diaphragm using silicon, featuring a gradual thickness variation, with a maximum thickness at the center. This article offers a mathematical analysis of the sensor, featuring step-by-step derivations for parameters such as the maximum thickness of the diaphragm, working range, capacitance, and capacitive sensitivity. Finally, curves of capacitance and capacitive sensitivity with pressure are obtained in MATLAB, and the results are compared with those of prior works. The proposed sensor shows a more than threefold increase in capacitance and a nearly double improvement in capacitive sensitivity compared to the highest previously reported double touch mode capacitive pressure sensor (DTMCPS). The deflection of the diaphragm is obtained using COMSOL. © 2001-2012 IEEE.

关键词： MATLAB

来源：评论

学校读者我要写书评

暂无评论

Automatically Generating Front-End Source Code Method for Complex Table Structures Based on Vue 7

Automatically Generating Front-End Source Code Method for Co...

引用

7th International Conference on Electrical, Mechanical and computer engineering, ICEMCE 2023

作者： She, Xiangfei Bian, Beibei School of Computer Technology and Engineering Changchun Institute of Technology Changchun130012 China

ISBN: (纸本)9798350382877

In enterprise software development, component-based front-end page technology is becoming increasingly important. However, traditional manual component-based programming is difficult to adapt to increasingly complex enterprise data table structures. This can lead to a large workload and slow down the development process. To address this issue, this paper proposes a method for automatically generating front-end source code for complex table structures based on Vue (AGT-Complex). This method uses hierarchical data structure recognition and expression to identify complex Excel table structures. It then forms a semantic mapping between business data tables and front-end program tables. Finally, it can generate code based on Vue components. Experimental results show that this method is effective and can greatly improve development efficiency. It can be used to develop complex data software with a high degree of automation. © 2023 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：