检索结果-内蒙古大学图书馆

IAENG International Journal of computer science 2024年第11期51卷 1720-1730页

作者： Li, Xudong Zhang, Yujun School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China

This paper introduces an advanced road damage detection algorithm that effectively addresses the shortcomings of existing models, including limited detection performance and large parameter sizes, by utilizing the YOLOv8n model. Key enhancements are integrated into the proposed algorithm to bolster its efficacy. First, the ConvNeXt V2 backbone network is integrated to improve the extraction of contextual features, thereby enhancing the effectiveness of road damage detection. Second, the algorithm employs a C2f_GhostNetV2 block structure to strengthen feature representation while simultaneously reducing computational costs. Additionally, PConv is utilized in the neck region to optimize spatial feature extraction, thereby minimizing redundant computations. The experimental results indicate that the proposed algorithm performs effectively on the Chinese subset of the RDD2022 dataset. Specifically, detection accuracy improved by 1.9%, recall by 1.8%, and mAP@0.5 by 2.1%, while the number of parameters decreased by 24% compared to the initial model. The optimized algorithm increases FPS by 4, meeting the dual requirements of mobile devices for accuracy and real-time object detection. © (2024), (International Association of Engineers). All rights reserved.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

An Improve Grey Wolf Optimizer Algorithm for Traveling Salesman Problems

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer science 2024年第6期51卷 602-612页

作者： Xu, Zhinan Zhang, Xiaoxia School of Computer Science and Software Engineering University of Science and Technology LiaoNing Anshan114051 China

The Traveling Salesman Problem (TSP) seeks the shortest closed tour that visits each city once and returns to the starting city. This problem is NP-hard, so it is not easy to solve using conventional methods. The grey wolf optimization (GWO) algorithm has shown outstanding per- formance in many practical applications. However, it is inclined towards premature convergence. This paper proposes an im- proved GWO (I-GWO) algorithm, which hybridizes GWO with genetic algorithms (GA) for the TSP. The main feature of the I-GWO algorithm is that it can make full use of the advantages of the GWO algorithm and the GA algorithm to make up for their respective shortcomings. Moreover, to make the GWO suitable for solving the TSP, both the 2-opt operator strategy and hamming distance h ave been designed to implement the discrete GWO directly. Additionally, to increase the diversity of solutions by expanding the search space, we present a new population update strategy with crossover and mutation operations in the next iteration. Meanwhile, the integration of the Simulated Annealing (SA) algorithm into the Improved Grey Wolf Optimizer (I-GWO) enhances its local search capabilities. Experimental results show that the I-GWO algorithm competes with established optimal methods for solving the TSP, suggesting its potential for different TSP variants and logistic transport domains. © (2024), (International Association of Engineers). All rights reserved.

关键词： Traveling salesman problem

来源：评论

学校读者我要写书评

暂无评论

Glaucoma Classification Using Enhanced Deep Transfer Learning Models with Hybrid ROI Cropped Optic Disc Technique

引用

SN computer science 2023年第6期4卷 1-15页

作者： Singh, Piyush Bhushan Singh, Pawan Dev, Harsh Chaurasia, Brijesh Kumar Amity School of Engineering and Technology Lucknow Amity University Uttar Pradesh Lucknow India Department of Computer Science and Engineering Pranveer Singh Institute of Technology Kanpur India

Glaucoma is a chronic, irrevocable eye ailment in which the optic nerve slowly deteriorates, impairing vision and quality of life. Specialists can diagnose clinical glaucoma;however, the procedures are either expensive or time-consuming. In this paper, we have proposed an enhanced version of various pre-trained CNN models (DenseNet, ResNet, VGG, Inception, and MobileNet) using batch normalization, inception module, and global pooling with the Swish activation function. We have used three datasets (ACREMA, LAG, and IEEE Fundus images) and pre-processed versions of these datasets in this work. Moreover, we have also proposed a hybrid region of interest (ROI) extraction approach in which we have extracted the region of interest using a combination of the Brightest Spot algorithm and Hough Circles implementation for pre-processing of the fundus images to improve the overall performance. We have also applied manual hyperparameter optimization for learning rate, batch size, SGD optimizer, and fine-tuning step. The proposed framework acquired an improved accuracy of 98–99% among all datasets used in our research. The extensive experimental simulation results indicate that the proposed framework is performing better than all the stand-alone models. The proposed hybrid model with DenseNet-169 acquired an accuracy of up to 99% using the ROI and the improved transfer learning model with customized layers on pre-processed IEEE dataset. The empirical results also validate the efficacy of our proposed model on pre-processed large dataset LAG, with the proposed hybrid approach, having 5824 fundus images and achieving up to 95% accuracy by customized DenseNet-169 model. The results show that the proposed framework using cutting-edge techniques can ensure its effectiveness. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： ConvBN DenseNet-169 Glaucoma Normal Optic disc Optic nerve head (ONH)

来源：评论

学校读者我要写书评

暂无评论

LDTR:Transformer-based lane detection with anchor-chain representation

引用

Computational Visual Media 2024年第4期10卷 753-769页

作者： Zhongyu Yang Chen Shen Wei Shao Tengfei Xing Runbo Hu Pengfei Xu Hua Chai Ruini Xue School of Computer Science and Engineering University of Electronic Science and Technology of ChinaChengdu 611731China Didi Chuxing Beijing 100081China

Despite recent advances in lane detection methods,scenarios with limited-or no-visual-clue of lanes due to factors such as lighting conditions and occlusion remain challenging and crucial for automated ***,current lane representations require complex post-processing and struggle with specific *** by the DETR architecture,we propose LDTR,a transformer-based model to address these *** are modeled with a novel anchorchain,regarding a lane as a whole from the beginning,which enables LDTR to handle special lanes *** enhance lane instance perception,LDTR incorporates a novel multi-referenced deformable attention module to distribute attention around the ***,LDTR incorporates two line IoU algorithms to improve convergence efficiency and employs a Gaussian heatmap auxiliary branch to enhance model representation capability during *** evaluate lane detection models,we rely on Fr´echet distance,parameterized F1-score,and additional synthetic *** results demonstrate that LDTR achieves state-of-the-art performance on well-known datasets.

关键词： transformer lane detection anchor-chain

来源：评论

学校读者我要写书评

暂无评论

Towards Accurate Heart Disease Diagnosis: An Overview of Machine Learning Approaches 6

Towards Accurate Heart Disease Diagnosis: An Overview of Mac...

引用

6th International Conference on Recent Trends in Advance Computing, ICRTAC 2023

作者： Deshmukh, Varun Pathak, Sunil Bhattacharya, Pronaya Verma, Ashwin Amity University Department of Computer Science & Engineering Jaipur India Amity University Amity School of Engineering and Technology Research and Innovation Cell Department of Computer Science and Engineering Kolkata India Nirma University Department of Computer Science & Engineering Ahmedabad India

ISBN: (纸本)9798350394702

cardiovascular diseases, especially heart ailments, pose substantial health challenges globally, often exacerbated by unhealthy lifestyle habits. Recent technological advances in machine learning (ML) have emerged as a beacon of hope, presenting avenues for swift and efficient heart disease detection. Despite the voluminous health data accumulated daily, its vast analytical potential is frequently overlooked, creating a pronounced knowledge disparity. Thus, it becomes imperative to study recent ML approaches for accurate heart diagnosis, which allows to fetch critical insights. In this paper, we discussed a comparative analysis of ML algorithms, datasets for heart diseases, and compared ML algorithms in terms of metrics like precision, recall, F1 and accuracy. We considered the benchmark Cleveland Heart Dataset for analysis. We observed an accuracy of 84% by Artificial Neural Network (ANN), while Support Vector Machine (SVM) has the highest recall of 95%. The presented findings and analysis indicate the efficacy of the analysis in heart prediction using ML in real-world setups. © 2023 IEEE.

关键词： Heart

来源：评论

学校读者我要写书评

暂无评论

Multi-lesion Segmentation of Fundus Images using Improved UNet++

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer science 2024年第10期51卷 1587-1595页

作者： Jiang, Haoyan Zhao, Ji School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China

Diabetes retinopathy (DR) is one of the complications of diabetes. Early diagnosis of retinopathy is helpful to avoid vision loss or blindness. The difficulty of this task lies in the significant differences in the size and shape of lesions between different DR samples, with a higher proportion of small lesions. We propose a new multi-disease segmentation method based on UNet++ to improve the segmentation accuracy of DR lesions. We chose Resnet50 as the backbone network and introduced a new hybrid residual module to replace the original residual module. At the same time, to compensate for the loss of information in DR small lesions during the feature extraction process, we introduce the Across Feature Map Attention (AFMA) is an auxiliary branch that enhances the segmentation accuracy of small-scale lesions. Finally, in response to the difficulty in extracting DR lesions in shallow models, the model abandoned the deep supervision structure in UNet++. In addition, we use a weighted mixed loss function to train the model. We conducted experiments on IDRID and DDR public datasets, simultaneously segmenting four typical DR lesions. The results on intersection over union (IOU) and dice similarity coefficient (Dice) showed that our method achieved competitive performance compared to other research methods. © (2024), (International Association of Engineers). All rights reserved.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Optimizing Stand-Alone PV Systems:A Metaheuristic-Enhanced Fuzzy Approach for Adaptive MPPT

引用

China Communications 2025年第1期22卷 61-74页

作者： Tina Samavat Mostafa Nazari Lin Fuhong Lei Yang Department of Computer Science and Engineering University of NevadaRenoNVUSA Faculty of Mechanical Engineering Shahrood University of TechnologyShahroodIran School of Computer and Communication Engineering University of Science and Technology BeijingBeijing 100083China

This paper introduces a simple yet effective approach for developing fuzzy logic controllers(FLCs)to identify the maximum power point(MPP)and optimize the photovoltaic(PV)system to extract the maximum power in different environmental *** propose a robust FLC with low computational complexity by reducing the number of membership functions and *** optimize the performance of the FLC,metaheuristic algorithms are employed to determine the parameters of the *** evaluate the proposed FLC in various panel configurations under different environmental *** results indicate that the proposed FLC can easily adapt to various panel configurations and perform better than other benchmarks in terms of enhanced stability,responsiveness,and power transfer under various scenarios.

关键词： genetic algorithm imperialist competitive algorithm invasive weed algorithm maximum power point tracking

来源：评论

学校读者我要写书评

暂无评论

IoT Based Smart Plant Monitoring System 13

IoT Based Smart Plant Monitoring System

引用

13th International Conference on Cloud Computing, Data science and engineering, Confluence 2023

作者： Nehra, Vibha Sharma, Megha Sharma, Vaibhav Amity University Department of Computer Science and Engineering Uttar Pradesh India Amity University Amity School of Engineering and Technology Uttar Pradesh India

ISBN: (纸本)9781665462631

Agriculture is a vast area that needs to be studied. technology development has brought us to the situation where there are sophisticated agricultural tools and systems. About 80% of India's surface water is used for agriculture. Applying IoT becomes necessary to ensure proper use of water resources and high-quality output. Because optimal irrigation in fields can provide better crops, this paper presents an open-source technology-based smart system to predict irrigation requirements of a field using sensing of ground parameters like soil moisture, environmental conditions like temperature and humidity, and external parameters like obstacle detection or intruder alert using an infrared sensor. The proposed system is based on an algorithm that takes into account sensed data in addition to the previously mentioned parameters. The entire system has been created and deployed on a pilot scale, and it wirelessly collects sensor node data over the cloud using web services, while a both web and mobile based information visualization and decision support system offers real-time insights based on the analysis of the sensor. The analysis displayed clear differences between the monitored and unmonitored environment and how it affects the plant health in both short and long term. The graphs help in visualizing the differences between the environment at fixed time intervals. In the future, this system can be combined with other technologies to predict rainfall, manage pH levels, and secure the vast amounts of data created and saved over the cloud by sensors. This would improve the system's long-term viability and make it more independent of human operation. © 2023 IEEE.

关键词： Soil moisture

来源：评论

学校读者我要写书评

暂无评论

Forecasting Smart Grid Stability Using a Hybrid CNN Bi-LSTM Approach

引用

SN computer science 2024年第5期5卷 1-18页

作者： Singhal, Divya Ahuja, Laxmi Seth, Ashish Amity Institute of Information Technology Amity University Uttar Pradesh Noida India School of Computer Science & amp Information Engineering INHA University Tashkent Uzbekistan

The prediction of smart grid stability is paramount for identifying contributors to network instabilities, yet it remains a challenging endeavor. To predict the stability of smart grid systems, researchers offer a novel method in this paper that uses a hybrid Convolutional Neural Network Bidirectional Long Short-Term Memory (CNN Bi-LSTM) model. By combining CNN's capacity to extract spatial characteristics with Bi-LSTM's capacity to record temporal dependencies, the suggested method overcomes the drawbacks of earlier techniques and provides a complete stability prediction solution. The research uses data transformation and normalization techniques to analyze data efficiently. To ensure fair handling of data, min–max normalization is utilized to reduce biases resulting from the Smart Grid database’s predilection for high-importance entries. This normalization process enhances algorithmic effectiveness by standardizing data sizes. The data analysis process begins with thorough exploration, allowing users to flexibly uncover preliminary patterns, primary points of interest, and distinctive attributes within the sizable dataset. Leveraging the Convolutional Neural Network framework, features are extracted, and the dimensionality problem stemming from the expanding smart grid systems is addressed using a principal component analysis-based feature selection technique. Finally, we introduce a novel Bidirectional Long Short-Term Memory technique tailored specifically for forecasting smart grid network stability. The proposed approach achieves superior accuracy, boasting a remarkable 96.9% accuracy rate compared to existing methods. This hybrid CNN Bi-LSTM methodology not only improves prediction accuracy but also enhances our understanding of smart grid dynamics, paving the way for more reliable and efficient energy management systems. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Bi-LSTM CNN Grid Stability Smart Grid Stability

来源：评论

学校读者我要写书评

暂无评论

Converse Attention Knowledge Transfer for Low-Resource Named Entity Recognition

引用

International Journal of Crowd science 2024年第3期8卷 140-148页

作者： Lyu, Shengfei Sun, Linghao Yi, Huixiong Liu, Yong Chen, Huanhuan Miao, Chunyan School of Computer Science and Technology University of Science and Technology of China Hefei230027 China School of Computer Science and Engineering Nanyang Technological University Singapore639798 Singapore

In recent years, great success has been achieved in many tasks of natural language processing (NLP), e.g., named entity recognition (NER), especially in the high-resource language, i.e., English, thanks in part to the considerable amount of labeled resources. More labeled resources, better word representations. However, most low-resource languages do not have such an abundance of labeled data as high-resource English, leading to poor performance of NER in these low-resource languages due to poor word representations. In the paper, we propose converse attention network (CAN) to augment word representations in low-resource languages from the high-resource language, improving the performance of NER in low-resource languages by transferring knowledge learned in the high-resource language. CAN first translates sentences in low-resource languages into high-resource English using an attention-based translation module. In the process of translation, CAN obtains the attention matrices that align word representations of high-resource language space and low-resource language space. Furthermore, CAN augments word representations learned in low-resource language space with word representations learned in high-resource language space using the attention matrices. Experiments on four low-resource NER datasets show that CAN achieves consistent and significant performance improvements, which indicates the effectiveness of CAN. © The author(s) 2024.

关键词： Labeled data

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：