检索结果-内蒙古大学图书馆

Atmospheric Noise-Resilient Image Classification in a Real-World Scenario: Using Hybrid CNN and Pin-GTSVM

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Mehendale, Shlok Sahoo, Jajati Keshari Roul, Rajendra Kumar Department of Computer Science & Engineering BITS Pilani K K Birla Goa Campus Goa India Department of Mathematics BITS Pilani K K Birla Goa Campus Goa India Department of Computer Science & Engineering Thapar Institute of Engineering & Technology Punjab Patiala India

Parking space occupation detection using deep learning frameworks has seen significant advancements over the past few years. While these approaches effectively detect partial obstructions and adapt to varying lighting conditions, their performance significantly diminishes when haze is present. This paper proposes a novel hybrid model with a pre-trained feature extractor and a Pinball Generalized Twin Support Vector Machine (Pin-GTSVM) classifier, which removes the need for a dehazing system from the current State-of-The-Art hazy parking slot classification systems and is also insensitive to any atmospheric noise. The proposed system can seamlessly integrate with conventional smart parking infrastructures, leveraging a minimal number of cameras to monitor and manage hundreds of parking spaces efficiently. Its effectiveness has been evaluated against established parking space detection methods using the CNRPark Patches, PKLot, and a custom dataset specific to hazy parking scenarios. Furthermore, empirical results indicate a significant improvement in accuracy on a hazy parking system, thus emphasizing efficient atmospheric noise handling. © 2025, CC BY.

关键词： Deep learning

A Systematic Review of the Future of Education in Perspective of Block Chain

学校读者我要写书评

暂无评论

Journal of Mobile Multimedia 2025年第5期19卷 1221-1254页

作者： Shams Tabrez Siddiqui Haneef Khan Md Imran Alam Kamal Upreti Sarika Panwar Sheela Hundekari Department of Computer Science Jazan University Saudi Arabia Department of Computer and Network Engineering Jazan University Saudi Arabia CHIRST (Deemed to be University) Ghaziabad India Department of Electronics and Telecommunication Engineering AISSMS's Institute of Information Technology Pune India MIT ADT University Pune India

Blockchain is one of the most revolutionary technologies in the past decade due to its decentralisation, data integrity, reliability, and security. Blockchain technology is the next popular topic, and it has the potential to significantly alter the educational environment in many ways. Blockchain technology must be used in the education sector despite its challenges. Education is one of the sectors where blockchain-based solutions are still in use. Many academics possess extensive knowledge of the societal benefits blockchain technology might bring. The vast potential of blockchain can only be realised if education expands its knowledge of the technology. The primary goal of this research is to identify current problems related to educational institutions and identify blockchain features that could assist in addressing them. This research article will provide an overview of existing activities and address several perspectives on how blockchain can revolutionize the education sector. In prolongation, this article investigates the categories of blockchain technology applications, especially in the field of education. An in-depth discussion on the benefits and impact that blockchain brings to education is explored. Further, deliberate the abundant challenges of adopting blockchain in education. This study will direct the organizations/institutions to decide which blockchain application will benefit the most based on their requisites. The analysis will also provide information about other educational fields that may benefit from blockchain technology.

关键词： Privacy Electronic learning Federated learning Scalability Education Blockchains Security Reliability Secure storage Systematic literature review

Two-Dimensional Unknown View Tomography from Unknown Angle Distributions

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Shah, Kaishva Chintan Gurumoorthy, Karthik S. Rajwade, Ajit Department of Electrical Engineering Indian Institute of Technology Bombay Mumbai India Walmart Global Tech Bengaluru India Department of Computer Science and Engineering Indian Insititute of Technology Bombay Mumbai India

This study presents a technique for 2D tomography under unknown viewing angles when the distribution of the viewing angles is also unknown. Unknown view tomography (UVT) is a problem encountered in cryo-electron microscopy and in the geometric calibration of CT systems. There exists a moderate-sized literature on the 2D UVT problem, but most existing 2D UVT algorithms assume knowledge of the angle distribution which is not available usually. Our proposed methodology formulates the problem as an optimization task based on cross-validation error, to estimate the angle distribution jointly with the underlying 2D structure in an alternating fashion. We explore the algorithm’s capabilities for the case of two probability distribution models: a semi-parametric mixture of von Mises densities and a probability mass function model. We evaluate our algorithm’s performance under noisy projections using a PCA-based denoising technique and Graph Laplacian Tomography (GLT) driven by order statistics of the estimated distribution, to ensure near-perfect ordering, and compare our algorithm to intuitive baselines. © 2025, CC BY-NC-ND.

关键词： computerized tomography

An Automated Deep Learning-Based VAR for Enhancing Referee Decision-Making in Football

学校读者我要写书评

暂无评论

An Automated Deep Learning-Based VAR for Enhancing Referee D...

Intelligent Data Communication Technologies and Internet of Things (IDCIoT), International Conference on

作者： Hamza Abu Zaken A. Prasanth Mohamad Ali Jahjah Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Chennai India

ISBN: (数字)9798331527549

ISBN: (纸本)9798331527556

While football is the most popular sport the world has ever seen, the question of referee decision-making has always been a shadow that hampers the spirit and justice of the game. Video Assistant Referee (VAR) was introduced to solve officiating errors through video reviews. Nonetheless, this technology has drawn heavy criticism for causing huge delays and not completely eradicating referee bias. In this research, a Tube Convolutional Neural Network (T-CNN) model is incorporated into a new VAR system as an advanced solution. By allowing for the input of spatial and temporal features for football footage, the T-CNN model processes features which allows for rapid and accurate analysis of complex match events such as offsides, fouls, and goal-line incidents. The proposed system makes gameplay more real-time while ensuring fairness by minimizing human involvement and the consequent delays in the decision-making process. Additionally, its unbiased AI-driven decisions also eliminate the challenge of home team bias. Comparative analyses with traditional VAR methods established that the T-CNN system achieves greater accuracy rate, real-time responsiveness, and fairness, ultimately transforming football officiating.

关键词： Reactive power Accuracy Video reviews Decision making Real-time systems Delays Convolutional neural networks Reliability Internet of Things Sports

Satellite-UAV Networks for 6G Control: A Sensing-Communication-Computing-Control Closed Loop Perspective

学校读者我要写书评

暂无评论

IEEE Network 2025年

作者： Lei, Chengleyang Fang, Xinran Feng, Wei Chen, Yunfei Xiao, Ming Ge, Ning Mao, Shiwen The Department of Electronic Engineering State Key Laboratory of Space Network and Communications Tsinghua University Beijing100084 China The Department of Engineering University of Durham DurhamDH1 3LE United Kingdom The Department of Information Science and Engineering School of Electrical Engineering and Computer Science Royal Institute of Technology Sweden The Department of Electrical and Computer Engineering Auburn University AuburnAL36849 United States

The future sixth-generation (6G) networks are envisioned to support machines and robots besides human beings. In order to efficiently support machines and robots working in remote and post-disaster areas, satellites and unmanned aerial vehicles (UAVs) can be utilized. In this article, we investigate how satellite-UAV networks can serve robots for their control tasks. We introduce the typical use cases of the satellite-UAV networks for control and outline the relevant challenges. The control process usually involves sensing, communication, computing, and control components, which forms a sensing-communication-computing-control (SC3) closed loop. In analogy with the reflex arc structure of humans, we reveal the indivisibility of the SC3 closed loops and propose a closed-loop-oriented system design framework. Within this framework, we take the closed-loop control performance as the design target, and highlight the holistic consideration of sensing, communication, computing, and control components in the SC3 closed loop. Furthermore, we discuss the extended SC3 closed loop with multiple sensors and robots to decompose complex control systems. A case study is provided to show the performance gain of the proposed closed-loop-oriented system design. Finally, we outline open issues in designing more efficient satellite-UAV networks for control. ©1986-2012 IEEE.

关键词： Unmanned aerial vehicles (UAV)

ARIES: AUTONOMOUS REASONING WITH LLMS ON INTERACTIVE THOUGHT GRAPH ENVIRONMENTS

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Gimenes, Pedro Cao, Zeyu Wong, Jeffrey Zhao, Yiren Department of Electrical & Electronic Engineering Imperial College London United Kingdom Department of Computer Science and Technology University of Cambridge United Kingdom

Recent research has shown that LLM performance on reasoning tasks can be enhanced by scaling test-time compute. One promising approach, particularly with decomposable problems, involves arranging intermediate solutions as a graph on which transformations are performed to explore the solution space. However, prior works rely on pre-determined, task-specific transformation schedules which are subject to a set of searched hyperparameters. In this work, we view thought graph transformations as actions in a Markov decision process, and implement policy agents to drive effective action policies for the underlying reasoning LLM agent. In particular, we investigate the ability for another LLM to act as a policy agent on thought graph environments and introduce ARIES, a multi-agent architecture for reasoning with LLMs. In ARIES, reasoning LLM agents solve decomposed subproblems, while policy LLM agents maintain visibility of the thought graph states, and dynamically adapt the problem-solving strategy. Through extensive experiments, we observe that using off-the-shelf LLMs as policy agents with no supervised fine-tuning (SFT) can yield up to 29% higher accuracy on HumanEval relative to static transformation schedules, as well as reducing inference costs by 35% and avoid any search requirements. We also conduct a thorough analysis of observed failure modes, highlighting that limitations on LLM sizes and the depth of problem decomposition can be seen as challenges to scaling LLM-guided reasoning. © 2025, CC BY.

关键词： Markov processes

Fairness Testing of Machine Learning Models using Combinatorial Testing in Latent Space

学校读者我要写书评

暂无评论

Fairness Testing of Machine Learning Models using Combinator...

IEEE International Conference on Software Testing Verification and Validation Workshop, ICSTW

作者： Arjun Dahal Sunny Shree Yu Lei Raghu N. Kacker D. Richard Kuhn Department of Computer Science and Engineering The University of Texas at Arlington Arlington USA Information Technology Laboratory National Institute of Standards and Technology Gaithersburg USA

ISBN: (数字)9798331534677

ISBN: (纸本)9798331534684

Decision-making by Machine Learning (ML) models can exhibit biased behavior, resulting in unfair outcomes. Testing ML models for such biases is essential to ensure unbiased decision-making. In this paper, we propose a combinatorial testing-based approach in the latent space of a generative model to generate instances that assess the fairness of black-box ML models. Our approach involves a two-step process: generating t-way test cases in the latent space of a Variational AutoEncoder and performing fairness testing using the instances reconstructed from these test cases. We experimentally evaluated our approach against an approach that generates t-way instances in the input space for fairness testing. The results indicate that the latent-space approach produces more natural test cases while detecting the first fairness violation faster and achieving a higher ratio of discriminatory instances to the total number of generated instances.

关键词： Combinatorial testing Conferences Decision making Autoencoders Closed box Machine learning

Coherence analysis of phase-controlled HOM effects

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Ham, Byoung S. Department of Electrical Engineering and Computer Science Gwangju Institute of Science and Technology 123 Chumdangwagi-ro Buk-gu Gwangju61005 Korea Republic of

The second-order intensity correlation of entangled photons has been intensively studied for decades, particularly for the Hong-Ou-Mandel (HOM) effect and nonlocal correlation - key quantum phenomena that have no classical counterparts. Recently, a path-entangled two-photon state has been experimentally demonstrated for both bosonic (symmetric) and fermionic (anti-symmetric) HOM effects by manipulating the photon’s phase at one input port. Entanglement represents a quantum superposition of path- or energy-correlated two-photon states with a relative phase. According to the conventional quantum mechanics, this phase is not an individual property but collective attribute of interacting photons. Here, the wave nature of photons is employed to coherently analyze the phase-controlled HOM effects recently observed in npj Quantum Info. 5, 43 (2019). A pure coherence approach is applied to derive a general solution for these phase-controlled HOM effects. Consequently, the quantum mystery of HOM effects, traditionally interpreted through the particle nature of quantum mechanics, is revealed as a coherent phenomenon between entangled photons via a selective choice of correlated photons. © 2025, CC BY.

关键词： Quantum entanglement

Auxiliary-Variable Adaptive Control Barrier Functions

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Liu, Shuo Xiao, Wei Belta, Calin Department of Mechanical Engineering Boston University BrooklineMA02215 United States Computer Science and Artificial Intelligence Lab Massachusetts Institute of Technology CambridgeMA United States Department of Electrical and Computer Engineering Department of Computer Science University of Maryland College ParkMD United States

This paper addresses the challenge of ensuring safety and feasibility in control systems using Control Barrier Functions (CBFs). Existing CBF-based Quadratic Programs (CBF-QPs) often encounter feasibility issues due to mixed relative degree constraints, input nullification problems, and the presence of tight or time-varying control bounds, which can lead to infeasible solutions and compromised safety. To address these challenges, we propose Auxiliary-Variable Adaptive Control Barrier Functions (AVCBFs), a novel framework that introduces auxiliary variables in auxiliary functions to dynamically adjust CBF constraints without the need of excessive additional constraints. The AVCBF method ensures that all components of the control input explicitly appear in the desired-order safety constraint, thereby improving feasibility while maintaining safety guarantees. Additionally, we introduce an automatic tuning method that iteratively adjusts AVCBF hyperparameters to ensure feasibility and safety with less conservatism. We demonstrate the effectiveness of the proposed approach in adaptive cruise control and obstacle avoidance scenarios, showing that AVCBFs outperform existing CBF methods by reducing infeasibility and enhancing adaptive safety control under tight or time-varying control bounds. Copyright © 2025, The Authors. All rights reserved.

关键词： Self adjusting control systems