检索结果-内蒙古大学图书馆

4th International Conference on Technological Advancements in Computational sciences, ICTACS 2024

作者： Nisha Chhabra, Megha Abidi, Ali Imam Setia, Sonia Maurya, Ajay Anjli, Km. Noida Institute of Engineering and Technology Department of Data Science U.P. Greater Noida India School of Engineering and Technology Sharda University Department of Computer Science & Engineering U.P. Greater Noida India Poornima Institute of Engineering and Technology Department of Artificial Intelligence and Data Science Rajasthan Jaipur India Delhi New Delhi India

ISBN: (纸本)9798350387490

In the dynamic realm of artificial intelligence, the evolution of Machine Translation Systems (MTS) for Indian languages faces unique challenges due to linguistic diversity. The diversity of morphology and the inherited dialects make it difficult to translate. The translation can be improved with pre-processing and optimization of MTS. Our MTS proposes the steps to improve the translation of Punjabi to English by employing two models, one with a Gated Recurrent Unit (GRU) layer and another with a Long Short-Term Memory (LSTM) layer. Another step involves Hyperparameter tuning to reduce complexity and increase model's efficiency. The experimentation shows accuracy rate on Samanantar dataset. The research aims to provide solutions to gaps identified in MTS for Punjabi, contributing to the advancement of machine translation technology in Punjabi's linguistic landscape. © 2024 IEEE.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

CNN Based Model for Underwater Image Classification and Enhancement

CNN Based Model for Underwater Image Classification and Enha...

引用

2024 International Conference on Advances in Modern Age Technologies for Health and engineering science, AMATHE 2024

作者： Jamandlamudi, Sahit Isravel, Deva Priya Dhas, Julia Punitha Malar Karunya Institute Of Technology And Sciences Division Of Data Science And Cyber Security Coimbatore India Karunya Institute Of Technology And Sciences Division Of Computer Science And Engineering Coimbatore India

ISBN: (纸本)9798350371567

The challenge of enhancing underwater images is difficult because of several issues, including color distortion, inadequate illumination, and dispersion from suspended particles in the water. Conventional picture enhancing methods frequently find it difficult to resolve these problems. Thanks to its ability to learn intricate representations and pick up on minute features, deep learning has become a potent tool for picture improvement tasks in recent years. In this work, we suggest a novel deep learning-based method for improving underwater images. To automatically learn and improve the visual quality of underwater scenes, we use a convolutional neural network architecture that is specifically made to handle underwater photos. A sizable dataset of underwater photos is used to train the network, with an emphasis on contrast improvement and color correction. We compare our method with many state-of-the-art techniques and undertake extensive tests on a benchmark dataset to evaluate our approach. The outcomes show that our strategy performs much better than current approaches both in terms of subjective visual quality and objective criteria. Our technique improves visibility, minimizes noise, and successfully restores colors in underwater photos. By presenting a deep learning-based method that enhances the visual quality of underwater photographs, this work advances the field of underwater image processing and aids in marine research, underwater exploration, and surveillance. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Quantum-enhanced Vehicle Detection and Speed Monitoring using OpenCV

Quantum-enhanced Vehicle Detection and Speed Monitoring usin...

引用

2024 International Conference on Integration of Emerging Technologies for the Digital World, ICIETDW 2024

作者： Simonthomas, S. Naresh, B. Rohith, K. Aarupadai Veedu Institute of Technology Department of Computer Science and Engineering Chennai India Aarupadai Veedu Institute of Technology Department of Artificial Intelligence and Data Science Chennai India

ISBN: (纸本)9798331541170

Quantum based vehicle detection is the innovative integration of Quantum Machine Learning (QML) techniques with classical computer vision methods to enhance vehicle detection and speed tracking systems using OpenCV. Traditional approaches in vehicle detection and speed tracking are often hindered by computational inefficiencies and accuracy limitations in complex traffic scenarios. Our proposed hybrid quantum-classical framework aims to overcome these complications by exploiting the computational advantages of quantum computing. In this study, we employ Quantum Principal Component Analysis (QPCA) for feature extraction and data dimensionality reduction, optimizing the input data for improved processing efficiency. The core detection mechanism integrates Quantum Neural Networks (QNNs) with classical Convolutional Neural Networks (CNNs), enhancing the model's accuracy in identifying vehicles within video frames. For speed tracking, we apply quantum optimization techniques to refine the tracking of vehicle positions across consecutive frames, ensuring precise speed calculation. Using OpenCV for real-time video processing and Qiskit for quantum computations, the findings from the experiments indicate significant improvements in both detection a precision, accuracy and computational efficiency. The hybrid quantum-classical model showcases the potential of quantum computing in advancing computer vision applications, particularly in intelligent transportation systems. This research highlights the transformative impact of QML techniques on practical applications and paves the way for further advancements in quantum-enhanced vehicle detection and speed tracking. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Diagnosis of Plant Leaf Disease using Vision Transformer 10

Diagnosis of Plant Leaf Disease using Vision Transformer

引用

10th International Conference on Communication and Signal Processing, ICCSP 2024

作者： Sundaraj, Ashik Isravel, Deva Priya Dhas, Julia Punitha Malar Karunya Institute of Technology and Sciences Division of Data Science and Cyber Security Coimbatore India Karunya Institute of Technology and Sciences Division of Computer Science and Engineering Coimbatore India

ISBN: (纸本)9798350353068

In precision agriculture, it is crucial to have a reliable system for identifying diseases and suggesting measures to maintain crop health and enhance yield performance. Addressing the persistent challenge of accurate and timely identification of plant illnesses, this study introduces a new understanding of the approach utilizing the Vision Transformer (ViT) architecture. Leveraging a diverse dataset encompassing both diseased and healthy plant images, the ViT model demonstrates efficacy by achieving an precision of 94% in recognizing and categorizing diseases based on plant leaves. The success of this transformer-based methodology underscores its potential to significantly enhance the field of agricultural diagnostics, offering a promising solution to improve crop management practices and mitigate potential losses in yield. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Dynamic Resource Allocation in Edge Computing via Deep Deterministic Policy Gradient Reinforcement Learning 3

Dynamic Resource Allocation in Edge Computing via Deep Deter...

引用

3rd International Conference on Integrated Circuits and Communication Systems, ICICACS 2025

作者： Gopal, Vidhya Surendar, U. Bajulunisha, A. Tamilselvi, M. Sathiyapriya, V. Saranya, N. Sona College of Technology Department of Computer Science and Engineering Salem India Srm institute of Science and Technology Department of Electronics and Communication Engineering Chennai India Chennai Institute of Technology Department of Computer Science and Engineering Kundrathur India Roever Engineering College Department of Computer Science and Engineering Perambalur India Knowledge Institute of Technology Department of Computer Science and Engineering Salem India Karpagam College of Engineering Department of Artificial Intelligence and Data Science Coimbatore India

ISBN: (纸本)9798331508456

The difficulty with dynamic and heterogeneous natured edge computing environments is resource provisioning. Reinforcement Learning (RL) can be promising to solve the problems of resource allocation under conditions of dynamics and complexity. This research work presents a novel DDPG-based (Deep Deterministic Policy Gradient) approach for optimizing the dynamics of resource allocation in edge computing environments by exploiting the CRAWDAD KAIST-WIBRO dataset for the proposed model. This approach implements a neural network that learns the optimal policies of allocations with dynamism towards new changes in workloads and resource availabilities. In order to test the performance of this method, conducted several simulations over real-world workload and resource data, mapping the parameters like rewards and variation in episodes. The simulation results indicate that proposed approach performs superior compared to traditional heuristic-based approaches as well as other RL algorithms for resource usage, response time, and energy efficiency. This proposed approach permits optimized resource utilization along with quality of service, of paramount importance in executing complex and demanding applications at the edge computing environment. © 2025 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Machine Learning for Detecting Subtle Signs of EyeDisease 4

Machine Learning for Detecting Subtle Signs of EyeDisease

引用

4th International Conference on Expert Clouds and Applications, ICOECA 2024

作者： Dafny Neola, J. Ebenezer, V. Edwin, E.Bijolin Kirubakaran, S.Stewart Thanka, M.Roshni Yona Davincy, R. Karunya Institute of Technology and Sciences Division of Computer Science and Engineering Coimbatore India Karunya Institute of Technology and Sciences Division of Data Science and Cyber Security Coimbatore India

ISBN: (纸本)9798350385793

Eye disorders can have serious repercussions, such as blindness or visual impairment, if they are not identified or addressed, Treatment that works and preventing permanent harm depend on early diagnosis and prediction. Using cutting-edge clinical observations, this method looks for diseases such diabetic retinopathy, cataracts, myopia, glaucoma, age-related eye illnesses, and hypertension-related eye disorders. This study presents an innovative method for early detection of eye health disorders: The Eye Health Station. It uses machine learning algorithm, by leveraging EfficientNet's superior performance in image classification tasks, our system analyzes ocular images, such as retinal scans, to identify potential indicators of eye disorders at an early stage., The proposed system seeks to discover possible markers of eye problems by evaluating several factors taken from ocular pictures and patient data, including retinal scans and medical history. The system is able to learn from various datasets and gradually increase its accuracy since machine learning techniques are included into the system. © 2024 IEEE.

关键词： Ophthalmology

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

Improved optimizer with deep learning model for emotion detection and classification

引用

Mathematical Biosciences and engineering 2024年第7期21卷 6631-6657页

作者： Joseph, Willson C. Kathrine, G. Jaspher Willsie Shanmuganathan, Vimal Sumathi, S. Pelusi, Danilo Valencia, Xiomara Patricia Blanco Verdú, Elena Department of Computer Science and Engineering Karunya Institute of Technology and Sciences Coimbatore India Department of Computer Science and Engineering Adi Shankara Institute of Engineering and Technology Kerala India Department of Artificial Intelligence and Data Science Sri Eshwar College of Engineering Coimbatore India Sri Eshwar College of Engineering Coimbatore India Communication Sciences University of Teramo Coste Sant’agostino Campus Teramo Italy Universidad Internacional de La Rioja La Rioja Logroño Spain

Facial emotion recognition (FER) is largely utilized to analyze human emotion in order to address the needs of many real-time applications such as computer-human interfaces, emotion detection, forensics, biometrics, and human-robot collaboration. Nonetheless, existing methods are mostly unable to offer correct predictions with a minimum error rate. In this paper, an innovative facial emotion recognition framework, termed extended walrus-based deep learning with Botox feature selection network (EWDL-BFSN), was designed to accurately detect facial emotions. The main goals of the EWDL-BFSN are to identify facial emotions automatically and effectively by choosing the optimal features and adjusting the hyperparameters of the classifier. The gradient wavelet anisotropic filter (GWAF) can be used for image pre-processing in the EWDL-BFSN model. Additionally, SqueezeNet is used to extract significant features. The improved Botox optimization algorithm (IBoA) is then used to choose the best features. Lastly, FER and classification are accomplished through the use of an enhanced optimization-based kernel residual 50 (EK-ResNet50) network. Meanwhile, a nature-inspired metaheuristic, walrus optimization algorithm (WOA) is utilized to pick the hyperparameters of EK-ResNet50 network model. The EWDL-BFSN model was trained and tested with publicly available CK+ and FER-2013 datasets. The Python platform was applied for implementation, and various performance metrics such as accuracy, sensitivity, specificity, and F1-score were analyzed with state-of-the-art methods. The proposed EWDL-BFSN model acquired an overall accuracy of 99.37 and 99.25% for both CK+ and FER-2013 datasets and proved its superiority in predicting facial emotions over state-of-the-art methods. ©2024 the Author(s), licensee AIMS Press.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

MTSC-Net:A Semi-Supervised Counting Network for Estimating the Number of Slash pine New Shoots

引用

植物表型组学（英文） 2024年第4期6卷 953-966页

作者： Zhaoxu Zhang Yanjie Li Yue Cao Yu Wang Xuchao Guo Xia Hao College of Information Science and Engineering Shandong Agricultural UniversityTaian271018Shandong ProvinceChina Research Institute of Subtropical Forestry Chinese Academy of ForestryHangzhou 311400Zhejiang ProvinceChina School of Big Data Taishan College of Science and TechnologyTaian271034Shandong ProvinceChina

The new shoot density of slash pine serves as a vital indicator for assessing its growth and photosynthetic capacity,while the number of new shoots offers an intuitive reflection of this *** deep learning methods becoming increasingly popular,automated counting of new shoots has greatly improved in recent years but is still limited by tedious and expensive data collection and *** resolve these issues,this paper proposes a semi-supervised counting network(MTSC-Net)for estimating the number of slash pine new ***,based on the mean-teacher framework,we introduce the improved VGG19 to extract multiscale new shoot ***,to connect local new shoot feature information with global channel features,attention feature fusion module is introduced to achieve effective feature ***,the new shoot density map and density probability distribution are processed in a fine-grained manner through multiscale dilated convolution of the regression head and classification *** addition,a masked image modeling strategy is introduced to encourage the contextual understanding of global new shoot features and improve the counting *** experimental results show that MTSC-Net outperforms other semi-supervised counting models with labeled percentages ranging from 5％to 50％.When the labeled percentage is 5％,the mean absolute error and root mean square error are 17.71 and 25.49,*** findings demonstrate that our work can be used as an efficient semi-supervised counting method to provide automated support for tree breeding and genetic utilization.

关键词： network counting estimating mtsc-net number pine semi-supervised shoots slash

来源：评论

学校读者我要写书评

暂无评论

Efficient allocation of independent gridlet on small, medium, and large grid

引用

Personal and Ubiquitous Computing 2023年第3期27卷 1029-1037页

作者： Rajeswari, D. Ramamoorthy, S. Srinivasan, R. Department of Data Science and Business Systems School of Computing College of Engineering and Technology SRM Institute of Science and Technology Tamil Nadu Kattankulathur603203 India Department of Computing Technologies School of Computing College of Engineering and Technology SRM Institute of Science and Technology Tamil Nadu Kattankulathur603203 India

Gridlet allocation in a computational grid environment is a major research issue to obtain not only the efficient gridlet allocation technique but also the time needed to obtain the efficient allocation technique. Grid computing networks have various nodes to process user jobs. To achieve the high performance of computational grid, task scheduling is an important issue. The users who are using services of grid systems are more cautious about time to complete their job. Hence, this work concentrates on gridlet allocation method used for time reduction by a genetic algorithm with the MapReduce programming model for independent tasks in computational grid. In computational grid environment, multi-objective problem formulation minimization of makespan and flowtime is considered. In this proposed technique, fitness function formulation for makespan and flowtime has been formulated mathematically. The genetic algorithm with the MapReduce programming model is implemented using MapReduce written in Java and then combined with GridSim. The experimental outcome with regard to time needed, flowtime, and makespan clearly reveals that the genetic algorithm with the MapReduce model effectively optimizes time, makespan, and flowtime in computational grid environment. A comparative study of performance efficiency among genetic algorithm with the MapReduce and sequential genetic algorithm (SGA) and parallel genetic algorithm (PGA) depicts the usefulness of the model. The execution time achieved by GA with the MapReduce model in small grid is 10.48 s, medium grid is 18.76 s, and large grid is 33.73 s. © 2023, The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature.

关键词： MapReduce

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：