检索结果-内蒙古大学图书馆

Scene text visual question answering by using YOLO and STN

International Journal of Speech Technology 2024年第1期27卷 69-76页

作者： Nourali, Kimiya Dolkhani, Elham Computer Science Department Semnan University Semnan Iran Computer Engineering Department Ferdowsi University of Mashhad Mashhad Iran

Extracting text from an image using a Visual Question Answering (VQA) system is an application at the intersection of computer vision and Natural Language Processing (NLP) to help blind people better view and comprehend textual information within the image. A VQA system takes a natural language question and an image as its input and then targets different areas of the image to extract an answer from the text within the image. To detect and recognize text in the image, the Look-Read-Reason-Answer (LoRRA), DCD_ZJU and Multimodal Multi-Copy Mesh methods, classify image features by using an Optical-Character-Recognition (OCR) module. Due to the curvature, distortion, background noise, and different fonts, understanding the text in natural scenes is a challenge as OCR techniques are mainly designed to detect textual features from within clean image files. In this paper, we introduce a novel architecture that rectifies perspective and curved text in an image using Thin-Plate-Spline (TPS) transformations and detects objects through the utilization of You-Only-Look-Once networks to extract answers. The results of our experiments show that our proposed VQA method reaches higher accuracy levels on the OpenImage dataset as compared with the existing methods. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2023.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Deepfake Detection: Emerging Techniques and Evolving Challenges 12

Deepfake Detection: Emerging Techniques and Evolving Challen...

引用

12th International Conference on Internet of Everything, Microwave, Embedded, Communication and Networks, IEMECON 2024

作者： Natarajan, Krishnaraj Mathur, Abeer Bhargava, Kshitiz Singh, Manvendra Tejpal, Moulik Department of Database Systems School of Computer Science & Engineering Vellore Institute of Technology Vellore India Department of Software Systems School of Computer Science & Engineering Vellore Institute of Technology Vellore India

ISBN: (纸本)9798350387315

Several newly developed techniques and tools for manipulating images, audio, and videos have been introduced as an outcome of the recent and rapid breakthroughs in AI, machine learning, and deep learning. While most applications for these techniques or tools are in the fields of entertainment and education, some individuals with unlawful intent have also been benefited from them. These individuals use such techniques for various purposes, including the spread of misleading information and unnecessary propaganda, the incitement of political instability, hate and unrest, as well as for purposes of torture and blackmail. These high-quality and convincing manipulated images, audio, or videos are commonly referred to as ‘Deepfakes’. Since then, various solutions to the problems raised by Deepfakes have been proposed in academic studies. This literature review contains relevant publications that offered a variety of approaches to give an updated summary of the research activities in different types of Deepfake attacks, their detection, and countermeasures. It also assesses the effectiveness of the detection capabilities of different techniques with various datasets and algorithms applied in Deepfake detection, while also outlining the various benefits and drawbacks of various methodologies. © 2024 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Leveraging Satellite Imagery to Analyze Large-Scale Economic Trends

Leveraging Satellite Imagery to Analyze Large-Scale Economic...

引用

2024 International Conference on science, Technology, engineering and Management, ICSTEM 2024

作者： Lavanya, Raja Indira, K. Santhiya, C. Chakkaravarthy, G. Vinoth Abhinav, S. Department of Computer Science and Engineering Thiagarajar College of Engineering Madurai India Department of Computer Science and Engineering Velammal College of Engineering Madurai India

ISBN: (纸本)9798350376913

Poverty is still one of the factors that have been affecting the lives of humans over hundreds of decades in the world but still, it is not entirely eradicated. One of the main reasons is the lack of identification of the economical pattern in a particular area that has been affected by poverty over a long period which has not been noted by any NGOs or government. So, that is the main reason why poverty eradication is still the first goal of UN sustainable development goals.[1] Understanding economical patterns over a region can lead us in providing informed policy-making, targeted NGO, and government-aided efforts. If we could track them down in an easy method at the earlier stage or in the advanced method and predict this fatal enigma on the region, the poverty can surely prevent or at least save lives. The above dilemma leads us to the essential method of tracking the reliable and timely measurements of economic activities, which are key for understanding economic development and designing government policies. But still many countries, especially developing countries, lack reliable data. Though data is available, the lack of quality in the data remains a huge challenge. In this paper, we propose a low-cost yet efficient approach to predict the economical pattern in a rural region from satellite imagery, which is globally available, in the meantime, satellite images are continuously updating to the changeable environmental conditions. Since the satellite image is way more advantageous than traditional methods, they play a huge role in the model. Traditional methods like surveys are expensive to conduct and include a lot of manpower. [2]This contributes to infer socioeconomic indicators from large-scale, remotely-sensed data. The available dataset is used to extract the luminous intensity from the night-time satellite images and added along with the other features of the particular region that determines the socio- economic conditions. Then with machine learning al

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

SeaCap: Multi-Sight Embedding and Alignment for One-Stage Image Captioner

引用

IEEE Transactions on Multimedia 2025年 27卷 3411-3425页

作者： Wang, Bo Zhang, Zhao Zhao, Mingbo Jin, Xiaojie Xu, Mingliang Wang, Meng Hefei University of Technology School of Computer Science and Information Engineering Hefei230601 China Yunnan Key Laboratory of Software Engineering Yunan China Donghua University School of Computer Science and Engineering Shanghai201620 China ByteDance Culver CityCA90230 United States Zhengzhou University School of Computer and Artificial Intelligence Zhengzhou450001 China

Recent mainstream image captioning methods usually adopt two-stage captioners, i.e., calculating the object features of the given image by a pre-trained detector and then feeding them into a language model to generate the descriptive sentences. However, such a two-stage procedure will lead to a task-based information gap that decreases the performance of the captioners, because the object features learned from the detection task are suboptimal representations and cannot provide all the necessary information for subsequent sentence generation. Besides, the object features are usually represented by the last pooling features of the detector that lose the local details of images. In this paper, we propose a novel One-Stage Image Captioner using dynamic multi-sight embedding and alignment, called SeaCap, which directly transforms input images into descriptive sentences in one stage to eliminate the information gap. Specifically, to obtain rich features, we use the Swin Transformer to capture multi-level features, followed by a sights alignment module to alleviate the vision confusion, and then feed them into a novel dynamic multi-sight embedding module to exploit both the global structure and local texture of input images. To enhance the global modeling capacity of the visual encoder, we propose a new dual-dimensional refining module to non-locally model the interaction of the embedded features. As a result, SeaCap can obtain rich and useful information to improve the performance of the captioner. Extensive comparisons on the benchmark MS-COCO, Flickr8K and Flickr30 K datasets verified the superior performance of our method. © 1999-2012 IEEE.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Facial Feature Extraction Review for Contemporary applications 15

Facial Feature Extraction Review for Contemporary applicatio...

引用

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Singh, Gaganjot Kaur, Shinepreet Devi, Shivali Chandigarh University Department of Computer Science and Engineering Mohali India Department of Computer Science and Engineering India

ISBN: (纸本)9798350370249

Recognizing a face is an intricate cognitive process that showcases the remarkable capabilities of the human brain in visual perception, a phenomenon deeply rooted in evolutionary biology. In an attempt to emulate this complex phenomenon, computer vision technology has been developed to serve as a sophisticated tool capable of recognizing and distinguishing faces, akin to the human visual system, particularly in the domain of biometric identification methods employed for verifying individuals' identities. This is done by identifying the unique features of a human's face through various algorithms. This paper reviews the efficacy of popular approaches to portraying facial features in face pictures. Geometrical-based methods, template-based methods, and appearance-based methods were traditional ways for extracting facial features before the onset of Machine learning for this application. Accuracy through confusion matrix have been comprehended and traditional methods have been found more fit for static and atomic systems although, machine learning algorithms give better and consistent results at the cost of developing and maintaining the computational system underneath. © 2024 IEEE.

关键词： 468 landmark detection discrete wavelet transform eigen faces Facial features detection geodesic distance local binary pattern morphometry

来源：评论

学校读者我要写书评

暂无评论

Boosting Write Performance of KV Stores: An NVM -Enabled Storage Collaboration Approach 40

Boosting Write Performance of KV Stores: An NVM -Enabled Sto...

引用

40th IEEE International Conference on Data engineering, ICDE 2024

作者： Wang, Yi He, Jiajian Sun, Kaoyi Dong, Yunhao Chen, Jiaxian Ma, Chenlin Zhou, Amelie Chi Mao, Rui College of Computer Science and Software Engineering Shenzhen University China Hong Kong Baptist University Department of Computer Science Hong Kong

ISBN: (纸本)9798350317152

As the most common data structure for key-value stores, LogStructured Merge Tree (LSM-tree) can eliminate random write operations and keep acceptable read performance. However, write stall and write amplification introduced by the leveled compaction of LSM-tree significantly degrade the system performance. The emerging non-volatile memory (NVM) provides byte-addressable access and low-latency data persistence. Integrating DIMM-interface NVM in the design of the LSM-tree can potentially alleviate the write stall and write amplification issue, as the access speed of NVM is several orders of magnitude faster than hard disk drives or flash memory-based solid-state drives. This hybrid storage should be carefully designed, requiring new architectural and key-value structural support. This paper presents ZigZagDB, an NVM-enabled data man-agement scheme for LSM-tree-based key-value stores. ZigZagDB adds additional layers of key-value stores and uses non-volatile memory as the storage media to hold these additional layers of data. The newly designed key-value stores alternately access the data from either SSD or NVM. This 'ZigZag' shape of storage collaboration and synchronization can benefit write efficiency and space utilization. By utilizing the NVM with very limited capacity, the redesigned organization of LSM-tree can effectively solve the write stall and write amplification issue. We demonstrate the viability of the proposed ZigZagDB using a set of extensive experiments. Experimental results show that ZigZagDB can significantly reduce the write amplification and boost the throughput in comparison with representative schemes. © 2024 IEEE.

关键词： Flash memory

来源：评论

学校读者我要写书评

暂无评论

Hybrid feature selection method for predicting software defect

引用

Journal of engineering and Applied science 2024年第1期71卷 124页

作者： Anju, A.J. Judith, J.E. Department of Computer Science Engineering Eranad Knowledge City Technical Campus Kerala Manjeri India Department of Computer Science Engineering Noorul Islam Center for Higher Education Tamil Nadu Kanyakumari India

To address the challenges associated with the abundance of features in software datasets, this study proposes a novel hybrid feature selection method that combines quantum particle swarm optimization (QPSO) and principal component analysis (PCA). The objective is to identify a subset of relevant features that can effectively contribute to the accuracy of a predictive model based on an artificial neural network (ANN). The quantum particle swarm optimization algorithm is employed to optimize the selection of features by simulating the behavior of quantum particles in a search space. This approach enhances the exploration and exploitation capabilities, allowing for a more effective identification of relevant features. Furthermore, principal component analysis is integrated into the hybrid method to reduce dimensionality and remove multicollinearity among features, thereby improving the efficiency of the feature selection process. The proposed hybrid method is applied to software defect datasets, where the selected subset of features is fed into an artificial neural network for defect prediction. The performance of the hybrid model is compared with traditional feature selection methods, standalone QPSO, and PCA. Experimental results demonstrate the effectiveness of the hybrid approach in achieving superior predictive accuracy while reducing the dimensionality of the dataset. The proposed approach not only enhances prediction accuracy but also provides a more interpretable and efficient subset of features for building robust defect prediction models. © The Author(s) 2024.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

A Mathematical and Dynamical Model for Analyzing H-Shaped PM Vernier Motor for Electric Motorcycle Mid-Drive Applications

引用

IEEE Transactions on Energy Conversion 2025年第1期40卷 529-543页

作者： Ghods, Mehrage Faiz, Jawad Bazrafshan, Mohammad Amin Gorginpour, Hamed Toulabi, Mohammad Sedigh University of Tehran School of Electrical and Computer Engineering College of Engineering Tehran*** Iran Persian Gulf University Department of Intelligent Systems Engineering and Data Science Bushehr*** Iran Oakland University Department of Electrical and Computer Engineering RochesterMI48309 United States

The permanent magnet (PM) Vernier machines enhance torque density and decrease cogging torque compared to conventional permanent magnet synchronous motor. This paper presents a novel fractional-slot H-shaped PM Vernier machines, which incorporates unique design for the PM housing and rotor core construction. These designs are specifically optimized for the small-scale electric vehicles application. The choice of the PM arrangement is based on the combination of V-shaped and spoke-array PM topologies. The aim is to achieve a higher torque density than V-shaped PM machines, while reducing cogging torque compared to spoke-array PM machines. By using this methodology, leakage fluxes in the terminal region of the PMs reduce, resulting in an augmentation of both flux-linkage and power factor. A prototype 680 W motor was fabricated with 12 slots and 8 poles to verify the accuracy and effectiveness of the proposed model. Mathematical model and dynamical analysis are proposed for the analytical prediction of machine performance. Due to the unique shape feature of the machine, the conformal mapping approach is applied to design the rotor core. To validate the model, the predicted performance of the motor was compared with those obtained by finite element analysis and experiments. © 1986-2012 IEEE.

关键词： Permanent magnets

来源：评论

学校读者我要写书评

暂无评论

Privacy-Enhanced Federated Expanded Graph Learning for Secure QoS Prediction

引用

IEEE Transactions on Services Computing 2025年第03期18卷 1641-1654页

作者： Zou, Guobing Yan, Zhi Hu, Shengxiang Gan, Yanglan Zhang, Bofeng Chen, Yixin Shanghai University School of Computer Engineering and Science Shanghai200444 China Donghua University School of Computer Science and Technology Shanghai201620 China Shanghai Polytechnic University School of Computer and Information Engineering Shanghai201209 China Washington University Department of Computer Science and Engineering St. LouisMO63130 United States

Current state-of-the-art QoS prediction methods face two main limitations. Firstly, most existing QoS prediction approaches are centralized, gathering all user-service invocation QoS records for training and optimization, which causes privacy breaches. While some federated learning-based methods consider user privacy in a distributed way, they either directly upload local trained parameters or use simple encryption for global aggregation at the central server, thus failing to truly protect user privacy. Secondly, existing federated learning-based methods neglect distributed user-service topology and latent behavior-attribute correlations, compromising QoS prediction accuracy. To address these limitations, we propose a novel framework named Privacy-Enhanced Federated Expanded Graph Learning (PE-FGL) for secure QoS prediction. It first conducts user-service expansion on the invocation graph with advanced privacy-preserving techniques, upgrading first-order local QoS invocations to high-order interaction relationships. Then, it extracts hybrid features from the expanded invocation graph via deep learning and graph residual learning. Finally, a two-layer secure mechanism of federated parameters aggregation is designed to enable collaborative learning among users through local parameter segmentation and global aggregation, achieving effective and secure QoS prediction. Extensive experiments on WS-DREAM demonstrate effective QoS prediction across multiple metrics while preserving privacy in user-service invocations. © 2025 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Recommendation systems-based software requirements elicitation process—a systematic literature review

引用

Journal of engineering and Applied science 2024年第1期71卷 29页

作者： Akram, Faiz Ahmad, Tanvir Sadiq, Mohd. Department of Computer Engineering Faculty of Engineering and Technology Jamia Millia Islamia A Central University New Delhi110025 India Software Engineering Lab. Computer Engineering Section UPFET Jamia Millia Islamia A Central University New Delhi110025 India

Requirements elicitation is one of the fundamental sub-processes of requirements engineering which is used to find the needs of stakeholders. There are several activities in this sub-process, i.e., identification of stakeholders and their requirements, software requirements prioritization, and analysis. Recommendation systems have been intertwined with the requirements elicitation process to predict the stakeholders’ requirements based on their preferences for functional and non-functional requirements. A number of systematic literature reviews (SLRs) have been carried out in the area of requirements elicitation. These SLRs do not support the applications of the recommendation systems during the requirements elicitation process. To deal with this issue, we present an SLR on recommendation systems-based software requirements elicitation processes, from 2009 to 2022, undertaking four research questions: (a) What are the different activities of the software requirements elicitation methods? (b) What are the applications of recommendation systems in the identification of the software requirements? (c) How the recommendation systems can facilitate the identification of stakeholders in the requirements elicitation process? (d) What are the ways to automate the selection of requirements elicitation techniques? The aim of this study is to identify the research gaps in the area of recommendation systems-based requirements elicitation processes and suggest future research directions. © 2024, The Author(s).

关键词： Recommender systems

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：