检索结果-内蒙古大学图书馆

3rd International Conference on Artificial Intelligence and Signal Processing, AISP 2023

作者： Nandhini, T.J. Thinakaran, K. Saveetha School of Engineering Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350320749

Object detection algorithms must first identify all the objects inside an image before machine vision can properly categorize and localize them. Many methods have been proposed to handle this problem, with most of the motivation coming from computer vision and deep learning methods. However, prevailing technologies have never effectively recognized tiny, dense things and often failed to detect objects that have undergone random geometric alterations. We analyze the current state of the art in object identification and propose a deformable convolutional network with adjustable depths to address these concerns. The results of our research suggest that they are better than the current best practices, blend deep convolutional networks with flexible convolutional structures to account for geometric variations, and get multi-scaled features. Next, we perform the remaining phases of object identification and region regress by up-sampling the fusion of multi-scaled elements. Experimental validation of our proposed framework demonstrates a considerable improvement in accuracy relative to time spent recognizing small target objects with geometric distortion. © 2023 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Automatic text inpainting and quality elevation in video sequences

引用

Multimedia Tools and Applications 2024年 1-34页

作者： Palivela, Lakshmi Harika Dharmalingam, Vivekanandan Gayathri, D. Bala School of Computer Science and Engineering Vellore Institute of Technology Chennai Vandalur-Kelambakkam Road Chennai600127 India Department of Information Technology Madras Institute of Technology Anna University Chennai India Department of Computer Science and Engineering B.S.Abdur Rahman Crescent Institute Of Science and Technology Chennai India

Scene text removal is a recent development in computer vision that replaces text patches in natural images with the appropriate background. Text removal is a difficult process leading to faulty areas of text containing text strokes with their hazy backgrounds. Text in the real world uses a variety of font kinds, some of which are difficult to localize due to their chaotic shapes, varied shading degrees, and orientation *** text erasing may include the subtasks of text detection as well as text inpainting. Both subtasks require a large amount of data to be successful;but, existing approaches were limited by insufficient real-world data for scene-text elimination. Eventhough the existing works produced considerable performance improvement in scene text removal, they often leave many text remains like text strokes, thus producinglow-quality visual outcomes. Therefore, this paper proposes an automatic text inpainting and video quality elevation model by using the Improved Convolutional Network-based ***, the video samples are collected from the diverse datasets and then converted into frames. Next, the frames are deblurred using an enhanced Convolutional Neural Network (CNN) model that has three convolutional layers for accurately localizing the texts in frames. Subsequently, the texts are detected by utilizing the CLARA-based VGG-16 network. Afterward, the text strokes are removed using a convolutional Encoder and decoder network to eliminate the presence of text on complex backgrounds and textures. Here, the coordinates of text in the deblurred frames are used to crop out the text stroke regions. So, the texts are in-painted, and then, the text in-painted regions are pasted back to their original positions in the frames. Furthermore, the video quality is elevated with the help of the DenseNet-centric Enhancement network. The experimental outcomes demonstrate that the proposed model effectively removed scene texts and enhanced the video qu

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Data-driven ML Approaches for the concept of Self-healing in CWN, Including its Challenges and Possible Solutions 8

Data-driven ML Approaches for the concept of Self-healing in...

引用

8th IEEE International Conference on science, Technology, engineering and Mathematics, ICONSTEM 2023

作者： Caleb, S. John Justin Thangaraj, S. Saveetha School of Engineering Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350347791

The idea of a self-organizing network, or SON for short, was developed to facilitate the autonomous deployment and administration of cellular networks. SON capabilities have the potential to improve service quality, increase network performance, and lower operating and capital expenditures (OPEX and CAPEX). Self-healing refers to a network paradigm in which faults in intended networks are repaired through the execution of a predetermined set of steps, and recompense. SON self-heals as a network paradigm. SON requires self-healing. It is now widely acknowledged that data-driven machine learning is an effective method for infusing networks with intelligence and enabling them to repair themselves. However, there are significant obstacles to overcome before practical implementations of machine-learning approaches for self-healing may be developed. In the first part of this essay, we will begin by dividing these difficulties into Data imbalance, data insufficiency, data cost insensitivity, data reaction time, multi-source data fusion, and data latency are the five main areas of concern. Then, after identifying these difficulties, we provide viable technological solutions to overcome them. In addition, A cost-sensitive defect detection case study employing unbalanced data shows the approaches' feasibility and efficiency © 2023 IEEE.

关键词： Data fusion

来源：评论

学校读者我要写书评

暂无评论

ResNet50-based Classification of Coffee Cherry Maturity using Deep-CNN 5

ResNet50-based Classification of Coffee Cherry Maturity usin...

引用

5th International Conference on Smart systems and Inventive Technology, ICSSIT 2023

作者： Raveena, S. Surendran, R. Saveetha School of Engineering Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9781665474672

up of coffee is a part of the daily human routine life. Any small changes in the coffee cherries yield that leads a huge change in coffee quality, smell, and taste. Types and stages of coffee cherries Identification and categorization of varieties and stages necessitate the presence of a farmer or crop pathologist. Manual illness diagnosis might result in mistaken identity and require a lengthy process. This work presents a comparison between the Deep Convolutional Neural Network-based approach and ResNet50 for classifying coffee type and maturity using transfer learning. Coffee cherries of different stages and varieties may be automatically sorted and recognized. Many ways to solve this problem have been proposed with transfer learning emerging as the dominant option due to its outstanding performance. This research work has implemented the Deep CNN Transfer Learning to detect distinct phases in the coffee cherries datasets. The suggested model employs VGG 16 for classifying coffee type and maturity got 96.38% accuracy and using ResNet50 for classifying coffee type and maturity with 99.01% accuracy. Experiments demonstrate that the suggested strategy is practical and that it would be used to detect efficiently and outperform other methodologies. © 2023 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Deep Neural Network-based Crime Scene Detection with Frames 8

Deep Neural Network-based Crime Scene Detection with Frames

引用

8th IEEE International Conference on science, Technology, engineering and Mathematics, ICONSTEM 2023

作者： Nandhini, J.T. Thinakaran, K. Department of Computer Science and Engineering Saveetha School of Engineering Chennai India

ISBN: (纸本)9798350347791

The field of computer vision stands to benefit significantly from automated crime scene detection. In this work, we demonstrate the application of DNN (Deep Neural Network) to identify a knife, blood, and gun in a picture and forecast whether a crime has occurred. Improving the accuracy of the system's detection capabilities was a top priority for us since we wanted to make the system as helpful as possible. To get a detection result, DNN's Non-linearity ReLu (Rectified Linear Unit), a Convolutional Neural Network Layer, a Fully Connected Layer, and a Dropout Function are employed. Additionally, this research makes use of a Dropout Function. To get the outcomes we want from Neural Networks, we use the open-source platform TensorFlow. Our system has a test accuracy of about 92.1% for the datasets that are currently accessible, which puts it in a very competitive position compared to other systems specifically created for this task. © 2023 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Few-Shot Object Detection Based on the Transformer and High-Resolution Network

引用

computers, Materials & Continua 2023年第2期74卷 3439-3454页

作者： Dengyong Zhang Huaijian Pu Feng Li Xiangling Ding Victor S.Sheng Hunan Provincial Key Laboratory of Intelligent Processing of Big Data on Transportation Changsha University of Science and TechnologyChangsha410114China School of Computer and Communication Engineering Changsha University of Science and TechnologyChangsha410114China School of Computer Science and Engineering Hunan University of Science and TechnologyXiangtan411004China Department of Computer Science Texas Tech UniversityLubbock79409TXUSA

Now object detection based on deep learning tries different *** uses fewer data training networks to achieve the effect of large dataset ***,the existing methods usually do not achieve the balance between network parameters and training *** makes the information provided by a small amount of picture data insufficient to optimize model parameters,resulting in unsatisfactory detection *** improve the accuracy of few shot object detection,this paper proposes a network based on the transformer and high-resolution feature extraction(THR).High-resolution feature extractionmaintains the resolution representation of the *** and spatial attention are used to make the network focus on features that are more useful to the *** addition,the recently popular transformer is used to fuse the features of the existing *** compensates for the previous network failure by making full use of existing object *** on the Pascal VOC and MS-COCO datasets prove that the THR network has achieved better results than previous mainstream few shot object detection.

关键词： Object detection few shot object detection transformer high-resolution

来源：评论

学校读者我要写书评

暂无评论

Salp Swarm Algorithm Based Optimal Network Reconfiguration with Distributed Generation for Power Loss Minimization and Voltage Profile Improvement

Salp Swarm Algorithm Based Optimal Network Reconfiguration w...

引用

2023 International Conference on Sustainable Technology and engineering, i-COSTE 2023

作者： Alende, Azmerawu Argawu School of Electrical and Computer Engineering Hawassa University Department of Power System and Energy Engineering Hawassa Ethiopia

ISBN: (纸本)9798350329711

In a distribution system, power loss and voltage deviation are the main concern for the customers and utility. The distribution system has a lower voltage level and higher amount of flowing current than the transmission system, so that, the percentage real power loss in the distribution system is higher. This study addresses power loss and voltage deviation concerns in distribution systems. It optimizes the Yirgalem-Ethiopia radial distribution network using the Salp Swarm Algorithm (SSA) and solar Distributed Generation (DG). SSA outperforms other algorithms in reducing real and reactive power losses and improving the voltage profile. The approach minimizes network costs and maintains voltage within acceptable limits. It proves cost-effective, demonstrating significant power loss reduction and improved voltage profiles. The resource feasibility of solar and wind power in Yirgalem city was analyzed and the outcomes showed that solar power generation is more desirable. For this reason, the solar-type Distributed Generation (DG) is used. The proposed SSA algorithm was compared with Particle Swarm Optimization (PSO) and Whale Optimization Algorithm (WOA) in three various scenarios (only optimal network reconfiguration, only optimal DG size and site, and simultaneous optimal network reconfiguration and DG allocation) for the Aposto feeder. As stated, the SSA method performs better in terms of reducing both real and reactive power losses and improvement of voltage profile. The model has been formulated to minimize the total cost of the network by determining the optima of the substation locations and power, the load transfers between the demand centers, the feeder routes and the load flow in the network subject to a set of constraints. From the point of view of economic evaluations, the proposed approach is cost-effective. Generally, the simulation results show that the proposed technique is effective to maintain all buses voltage magnitude within the IEEE acceptable lim

关键词： Distributed power generation

来源：评论

学校读者我要写书评

暂无评论

ConU: Conformal Uncertainty in Large Language Models with Correctness Coverage Guarantees

ConU: Conformal Uncertainty in Large Language Models with Co...

引用

2024 Findings of the Association for Computational Linguistics, EMNLP 2024

作者： Wang, Zhiyuan Duan, Jinhao Cheng, Lu Zhang, Yue Wang, Qingni Shi, Xiaoshuang Xu, Kaidi Shen, Hengtao Zhu, Xiaofeng School of Computer Science and Engineering University of Electronic Science and Technology of China China Department of Computer Science Drexel University United States Department of Computer Science University of Illinois Chicago United States

ISBN: (纸本)9798891761681

Uncertainty quantification (UQ) in natural language generation (NLG) tasks remains an open challenge, exacerbated by the closed-source nature of the latest large language models (LLMs). This study investigates applying conformal prediction (CP), which can transform any heuristic uncertainty notion into rigorous prediction sets, to black-box LLMs in open-ended NLG tasks. We introduce a novel uncertainty measure based on self-consistency theory, and then develop a conformal uncertainty criterion by integrating the uncertainty condition aligned with correctness into the CP algorithm. Empirical evaluations indicate that our uncertainty measure outperforms prior state-of-the-art methods. Furthermore, we achieve strict control over the correctness coverage rate utilizing 7 popular LLMs on 4 free-form NLG datasets, spanning general-purpose and medical scenarios. Additionally, the calibrated prediction sets with small size further highlights the efficiency of our method in providing trustworthy guarantees for practical open-ended NLG applications. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Recognition of American sign language using modified deep residual CNN with modified canny edge segmentation

引用

Multimedia Tools and Applications 2025年 1-28页

作者： Hariharan, U. Devarajan, Devikanniga Kumar, P. Santhosh Rajkumar, K. Meena, M. Akilan, T. Department of Computer Science and Engineering Apex Institute of Technology Chandigarh University Punjab Mohali140413 India Department of Computer Science and Engineering GITAM University Karnataka Bengaluru561203 India Department of Information Technology Faculty of Engineering and technology SRM Institute of Science and Technology Ramapuram Tamil Nadu Chennai600089 India Department of Data Science and Business Systems School of Computing Faculty of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Tamil Nadu Chennai603203 India Department of Computer Science and Engineering Sri Sairam Engineering College Tamil Nadu Tambaram602109 India Research Coordinator Dept. of IQAC Amity University Kolkata West Bengal 700135 India

American Sign Language (ASL) recognition aims to recognize hand gestures, and it is a crucial solution to communicating between the deaf community and hearing people. However, existing sign language recognition algorithms still have some drawbacks, such as difficulty recognizing hand movements and low recognition accuracy for most sign language recognition. A Modified Convolutional Neural Network (MCNN) deep residual 101 classifier-based American Sign Language identification system has been developed to address this problem. There are three main parts presents in the method. The first part is pre-processing the images to remove the noise, enhance the contrast of the picture, adjust the contrast level and smoothen the picture using various filters. The second part is the segmentation, and it’s used to partition the image using a modified canny edge detection method by removing all weak edges present in the image. Finally, classification will be done using the Modified CNN deep residual 101 classifiers. By classifying the image, the American Sign Language is accurately identified. This process is conducted through images. The outcome shows that the suggested approach has a 0.97% accuracy and a 0.05% False Positive Rate. Other CNN architectures such as resNet 50 and resNet 18 reached 0.90% and 0.80% accuracy, respectively, which is lower than our proposed method. In addition, the resNet 101 classifier effectively recognizes the difficult hand gestures through the image data. It obtains high recognition accuracy for 36 signs is 0 to 9 numbers and a to z alphabets from American Sign Language. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Exploring the possibilities of integration of cyber-psychology for human behaviour in a smart city

引用

International Journal of Information and Communication Technology 2024年第2期25卷 150-167页

作者： Wen, Liping Ting, Zhou Zheng, Huang Alfred Daniel, J. Antonidoss, A. Department of Psychology University of Chinese Academy of Sciences Beijing100101 China Institute of Psychology Chinese Academy of Sciences Beijing100101 China Department of Medical Psychology School of Health Humanities Peking University Beijing100191 China Department of Computer Science & Engineering SNS College of Technology India Department of Computer Science and Engineering Hindustan Institute of Technology and Science India

The smart city idea differs between cities and nations. In all meanings and characteristics of a smart city, public involvement is the only thing that remains common. Therefore, it is a very significant field to study human behaviour and development in smart cities. This paper presents a framework for identifying qualities necessary for people to be classified as intelligent persons and to integrate these human behavioural characteristics in cyber technology. Human behaviour in a smart city has been analysed using the machine learning algorithm and big data analytics. The integrated machine learning and big data analytics framework (iML-BD) classifies the cyber behaviour of intelligent persons in a smart city by observing the cyber activities performed by the individuals. Furthermore, this paper handles the risk factors for cyber-acquired and cyber-dependant crime violence and abuse that vulnerable internet and public access devices using blockchain technology. Blockchain is a method of storing data that takes too long to alter, modify, or manipulate. A blockchain is an electronic accounting system that is reproduced and spread through the Bitcoin protocol’s entire communication network. The case study performed on iML-BD has resulted in the highest performance in terms of prediction accuracy of 94.98%. Copyright © 2024 Inderscience Enterprises Ltd.

关键词： Smart city

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：