检索结果-内蒙古大学图书馆

An Improved YOLO Detection Approach for Pinpointing Cucumber Diseases and Pests

computers, Materials & Continua 2024年第12期81卷 3989-4014页

作者： Ji-Yuan Ding Wang-Su Jeon Sang-Yong Rhee Chang-Man Zou Department of IT Convergence Engineering University of KyungnamChangwon51767Republic of Korea Department of Computer Engineering University of KyungnamChangwon51767Republic of Korea College of Computer Science and Technology Beihua UniversityJilin132013China

In complex agricultural environments,cucumber disease identification is confronted with challenges like symptom diversity,environmental interference,and poor detection *** paper presents the DM-YOLO model,which is an enhanced version of the YOLOv8 framework designed to enhance detection accuracy for cucumber *** detection models have a tough time identifying small-scale and overlapping symptoms,especially when critical features are obscured by lighting variations,occlusion,and background *** proposed DM-YOLO model combines three innovative modules to enhance detection performance in a collective ***,the MultiCat module employs a multi-scale feature processing strategy with adaptive pooling,which decomposes input features into large,medium,and small *** approach ensures that high-level features are extracted and fused effectively,effectively improving the detection of smaller and complex patterns that are often missed by traditional ***,the ADC2f module incorporates an attention mechanism and deep separable convolution,which allows the model to focus on the most relevant regions of the input features while reducing computational *** identification and localization of diseases like downy mildew and powdery mildew can be enhanced by this combination in conditions of lighting changes and ***,the C2fe module introduces a Global Context Block that uses attention mechanisms to emphasize essential regions while suppressing those that are not *** design enables the model to capture more contextual information,which improves detection performance in complex backgrounds and small-object scenarios.A custom cucumber disease dataset and the PlantDoc dataset were used for thorough *** results showed that DM-YOLO achieved a mean Average Precision(mAP50)improvement of 1.2%p on the custom dataset and 3.2%p on the PlantDoc dataset over the baseline *** results highlight

关键词： ADC2f C2fe cucumber diseases YOLOv8n MultiCat pest detection

来源：评论

学校读者我要写书评

暂无评论

BNSENTMIX: A Diverse Bengali-English Code-Mixed Dataset for Sentiment Analysis 1

BNSENTMIX: A Diverse Bengali-English Code-Mixed Dataset for ...

引用

1st Workshop on Language Models for Low-Resource Languages, LoResLM 2025 - co-located at the 31st International Conference on Computational Linguistics, COLING 2025

作者： Alam, Sadia Farhan Ishmam, Md. Alvee, Navid Hasin Shahnewaz Siddique, Md. Kamal, Abu Raihan Mostofa Azam Hossain, Md. Department of Computer Science and Engineering Islamic University of Technology Bangladesh

ISBN: (纸本)9798891762152

The widespread availability of code-mixed data in digital spaces can provide valuable insights into low-resource languages like Bengali, which have limited annotated corpora. Sentiment analysis, a pivotal text classification task, has been explored across multiple languages, yet code-mixed Bengali remains underrepresented with no large-scale, diverse benchmark. Code-mixed text is particularly challenging as it requires the understanding of multiple languages and their interaction in the same text. We address this limitation by introducing BNSENTMIX, a sentiment analysis dataset on code-mixed Bengali comprising 20,000 samples with 4 sentiment labels, sourced from Facebook, YouTube, and e-commerce sites. By aggregating multiple sources, we ensure linguistic diversity reflecting realistic code-mixed scenarios. We implement a novel automated text filtering pipeline using fine-tuned language models to detect code-mixed samples and expand code-mixed text corpora. We further propose baselines using machine learning, neural networks, and transformer-based language models. The availability of a diverse dataset is a critical step towards democratizing NLP and ultimately contributing to a better understanding of code-mixed languages. © 2025 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

A Comprehensive Analysis on Sentiment Using Deep Learning and Transformers 6

A Comprehensive Analysis on Sentiment Using Deep Learning an...

引用

6th International Conference on Cybernetics, Cognition and Machine Learning Applications, ICCCMLA 2024

作者： Mohsin, Md. Wankhede, Jaishri Fatima, Afshan Solanki, Vijender Kumar Department of Computer Science and Engineering Hyderabad India Stanley College of Engineering & Technology for women Department of Computer Science and Engineering Hyderabad India

ISBN: (纸本)9798331505790

Emotions describe the social attachment between the human that are ascendancy by cultural norms, social interactions, and Interpersonal bonds. So in this paper we are represent the application of deep learning models and transformers for emotion analysis on text based data. We are using different Word2Vec approaches for analyze the emotion from the textual data. As we know, Distributed Bag of Words (DBOW), Distributed Memory Concatenation (DMC), Distributed Memory Mean (DMM), and hybrid models DBOW+DMC and DBOW+DMM that's all are the part of Word2Vec model and all are based on deep learning techniques. In this paper we also explore the NLP transformer such as BERT, BART, DistilBERT, RoBERTa, and GPT to analyze the emotion of data and compare the performance of deep learning based technique with transformers based techniques. The aim of the exploration is compare the effectiveness of these models using standard sentiment analysis datasets, assessing performance through accuracy metrics. © 2024 IEEE.

关键词： Distribution transformers

来源：评论

学校读者我要写书评

暂无评论

A Novel Intelligent Task Offloading Scheme for Multicontroller Environment in Software Defined Internet of Vehicles

引用

IEEE Internet of Things Journal 2025年第12期12卷 20018-20029页

作者： Hussain, Mir Wajahat Sangaiah, Arun Kumar Reddy, K. Hemant Kumar Roy, Diptendu Sinha Alenazi, Mohammed J. F. Javvaji, Pavan Kumar Alliance University Department of Computer Science and Engineering Alliance College of Engineering and Design Anekal562106 India National Yunlin University of Science and Technology International Graduate Institute of Artificial Intelligence Yunlin Douliu64002 Taiwan Sunway University School of Engineering and Technology Subang Jaya47500 Malaysia University Centre for Research and Development Chandigarh University Mohali140413 India VIT-AP University School of Computer Science and Engineering Amaravati522237 India National Institute of Technology Department of Computer Science and Engineering Shillong793003 India King Saud University College of Computer and Information Sciences Department of Computer Engineering Riyadh11451 Saudi Arabia

The Internet of Vehicles (IoV), equipped with sensors, generates vast amounts of data, demanding rigorous computation and network. The cloud computing (CC) platform meets these stringent computation requirements, but it has a significant latency that fog computing (FC) effectively handles. Software defined network (SDN) has become the de facto standard for next-generation networking due to its unique and flexible features, which handle the network prerequisites in an agile manner. Task offloading (TO) is a crucial issue in software defined-IoV (SD-IoV), particularly when the vehicle’s resources are insufficient. Fog nodes schedule offloaded tasks;however, neglecting computational and network details during task scheduling can lead to longer completion times, thereby restricting the functionality of SD-IoV due to its time-sensitive nature. Therefore, this article introduces two proactive intelligent TO (ITO’ and ITO) schemes. These schemes schedule tasks onto fog nodes with enhanced computational capabilities, dynamically enable the network through the ovs-ofctl SDN utility, and consider vehicle mobility during task scheduling. Experiments with Mininet show that the proposed (ITO’ and ITO) schemes improve performance in terms of CPU availability, bandwidth, and throughput by 5.2 times, 50% and (30.7–13.9)% more than the existing scheme. They also reduce average packet loss, delay, round trip time (RTT), and offloaded node time selection by (45.35–21.3)%, (50–33.3)%, (53.9–9.9)% and (43.2–26.06)% which demonstrates the efficacy of the proposed scheme. © 2014 IEEE.

关键词： Fog computing

来源：评论

学校读者我要写书评

暂无评论

Realtime Object Detection Through M-ResNet in Video Surveillance System

引用

Intelligent Automation & Soft Computing 2023年第2期35卷 2257-2271页

作者： S.Prabu J.M.Gnanasekar Department of Computer Science&Engineering Karpaga Vinayaga College of Engineering and TechnologyChengalpattuTamilnadu603308India Department of Computer Science&Engineering Sri Venkateshwara College of Engineering&TechnologyChennai602001TamilnaduIndia

Object detection plays a vital role in the video surveillance *** enhance security,surveillance cameras are now installed in public areas such as trafﬁc signals,roadways,retail malls,train stations,and ***,monitor-ing the video continually at a quicker pace is a challenging *** a consequence,security cameras are useless and need human *** primary difﬁculty with video surveillance is identifying abnormalities such as thefts,accidents,crimes,or other unlawful *** anomalous action does not occur at a high-er rate than usual *** detect the object in a video,ﬁrst we analyze the images pixel by *** digital image processing,segmentation is the process of segregating the individual image parts into *** performance of segmenta-tion is affected by irregular illumination and/or low *** factors highly affect the real-time object detection process in the video surveillance *** this paper,a modiﬁed ResNet model(M-Resnet)is proposed to enhance the image which is affected by insufﬁcient *** results provide the comparison of existing method output and modiﬁcation architecture of the ResNet model shows the considerable amount improvement in detection objects in the video *** proposed model shows better results in the metrics like preci-sion,recall,pixel accuracy,etc.,andﬁnds a reasonable improvement in the object detection.

关键词： Object detection ResNet video survilence image processing object quality

来源：评论

学校读者我要写书评

暂无评论

A Review on Machine Learning and Deep Learning Based Systems for the Diagnosis of Brain Cancer

引用

SN computer science 2024年第1期5卷 28页

作者： Saha, Prottoy Das, Shanta Kumar Das, Rudra Department of Computer Science and Engineering Khulna University of Engineering and Technology Khulna 9203 Bangladesh

Brain cancer is a disease of the brain caused by a brain tumor. A brain tumor is the development of cells in the brain that grow in an unregulated and unnatural manner. Patients may suffer irreversible brain damage or even death if these tumors are not detected and treated properly. As with all types of treatment, Positional information and tumor size are critical for conventional systems. Thus, establishing a meticulous and automated approach to providing information to medical practitioners is required. With machine learning, deep learning, and several imaging modalities, physicians may now more reliably detect tumor types in a shorter period. The paper aims to provide an overview of newly developed systems that use machine learning and deep learning approaches to analyze various medical imaging modalities in the case of diagnosing brain tumors. Datasets used by the authors, dataset partitioning strategies, and different performance evaluation matrices are also described in this paper. To better understand the policy of categorization, we propose a taxonomy here where we have categorized deep learning and machine learning based systems with respect to single classifier, multiple classifiers, single dataset and multiple dataset. Finally, we focus on the challenges of deep learning algorithms for brain tumor classification and possible future trends in this field. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Brain cancer Convolutional neural network Deep learning Machine learning

来源：评论

学校读者我要写书评

暂无评论

A Framework for Industrial Inspection System using Deep Learning

引用

Annals of Data science 2024年第2期11卷 445-478页

作者： Hridoy, Monowar Wadud Rahman, Mohammad Mizanur Sakib, Saadman Department of Mechatronics & amp Industrial Engineering Chittagong University of Engineering & ampTechnology Chattogram Bangladesh Department of Mechanical Engineering Chittagong University of Engineering & ampTechnology Chattogram Bangladesh Department of Computer Science & amp Engineering Chittagong University of Engineering & ampTechnology Chattogram Bangladesh

Industrial Inspection systems are an essential part of Industry 4.0. An automated inspection system can significantly improve product quality and reduce human labor while making their life easier. However, a deep learning-based camera inspection system requires a large amount of data to classify the defective products accurately. In this paper, a framework is proposed for an industrial inspection system with the help of deep learning. Additionally, A new dataset of hex-nut products is proposed containing 4000 images, i.e., 2000 defective and 2000 non-defective. Moreover, different CNN architectures, i.e., Custom CNN, Inception ResNet v2, Xception, ResNet 101 v2, ResNet 152 v2, are experimented with the concept of transfer learning on the new hex-nut dataset. Fine-tuning the CNN architectures is performed by freezing the last 14 layers, which provided the optimal architecture, i.e., Xception (last 14 layers trainable, excluding the fully connected layer). The proposed framework can efficiently separate the defective products from the non-defective products with 100% accuracy on the hex nut dataset. Furthermore, the proposed optimal Xception architecture has experimented on a publicly available casting material dataset which produced 99.72% accuracy, outperforming existing methods. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2022. Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s);author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

关键词： Industry 4.0

来源：评论

学校读者我要写书评

暂无评论

Early detection of stroke disease using patients previous medical data instil with deep learning

引用

Multimedia Tools and Applications 2025年第16期84卷 16853-16881页

作者： Diwan, Tausif Gajbhiye, Saurav M. Goydani, Purva R. Gannarpwar, Vedant R. Khandait, Harshal R. Tembhurne, Jitendra V. Sahare, Parul Department of Computer Science and Engineering Indian Institute of Information Technology Maharashtra Nagpur441108 India Department of Electronics and Communication Engineering Indian Institute of Information Technology Maharashtra Nagpur441108 India

Early detection of any disease and starting its treatment in this early stage are the most important steps in case of any life-threatening disease. Stroke is not an exception in this regard which is one of the leading causes of death and disability worldwide. We develop a simple but efficient deep neural network for the stroke prediction that accurately evaluates the probability of occurrence of stroke disease by treating this as a binary classification problem on one of the standard datasets named as Stroke Prediction Dataset available on Kaggle. With the help of effective pre-processing techniques such as SMOTE and FastICA on the noisy and imbalanced dataset, we could achieve improved performance for the binary classification of stroke prediction by employing a deep dense neural network. The architectural hyper-parameters are critically designed with the help of Keras Tuner and consisting of just ten layers in the deep dense neural network. With the help of this light weight deep dense model, we report an accuracy of 95%, AUC Score of 100% and F1-Score of 95% on a testing set. Moreover, we also present comparative illustration of various machine learning models for the aforesaid task along with the comparative illustration of various state of the arts with our proposed model for the stroke prediction. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

WGSDMM+GA: A genetic algorithm-based service clustering methodology assimilating dirichlet multinomial mixture model with word embedding

引用

Future Generation computer Systems 2023年第1期145卷 254-266页

作者： Agarwal, Neha Sikka, Geeta Awasthi, Lalit Kumar Computer Science and Engineering Department Indian Institute of Information Technology Karnataka Raichur584135 India Department of Computer Science and Engineering National Institute of Technology Delhi Delhi110036 India 246174 India

Advancement of web 2.0 results in the expeditious growth of services in repositories and service portals, which raises the demand for service management. With the clusters of services, different processes like discovery, selection, ranking, etc., can be expedited efficiently. K-Means clustering is highly recommended for service clustering but it can be stuck in the local optimum values. Developers generally use short text to describe the functionality of Web services. For effective service clustering, it is required that relevant features of services are mapped in vector space based on semantic relations. Gibbs Sampling algorithm for Dirichlet Multinomial Mixture model (GSDMM) model is able to provide good performance on short text because it works on the assumption that there will be one topic related to one document. In this paper, a novel technique for web service clustering, i.e., WGSDMM+GA, is proposed in which relevant features, semantic relations, and their mappings in vector space are analyzed based on an assimilated approach of Word2Vec and GSDMM models. As K-means clustering cannot provide effective results due to local optima problem, GA-based clustering is applied in which fitness function is based on Manhattan distance. The proposed model is applied to five real-time datasets, and the performance is evaluated based on standard evaluation measures. Experimental results illustrate that WGSDMM+GA yields better results, and the accuracy score is increased by 23%, 17%,17%, 21%, and 25% on five real-time datasets. © 2023 Elsevier B.V.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Mango Pest Detection Using Entropy-ELM with Whale Optimization Algorithm

引用

Intelligent Automation & Soft Computing 2023年第3期35卷 3447-3458页

作者： U.Muthaiah S.Chitra Department of Computer Science and Engineering Sri Shanmugha College of Engineering and TechnologySankari637304India Department of Computer Science and Engineering Er.Perumal Manimekalai College of EngineeringHosur635117India

Image processing,agricultural production,andﬁeld monitoring are essential studies in the researchﬁ*** diseases have an impact on agricultural production and *** disease detection at a preliminary phase reduces economic losses and improves the quality of *** identifying the agricultural pests is usually evident in plants;also,it takes more time and is an expensive technique.A drone system has been developed to gather photographs over enormous regions such as farm areas and *** atmosphere generates vast amounts of data as it is monitored closely;the evaluation of this big data would increase the production of agricultural *** paper aims to identify pests in mango trees such as hoppers,mealybugs,inﬂorescence midges,fruitﬂies,and stem *** of the massive volumes of large-scale high-dimensional big data collected,it is necessary to reduce the dimensionality of the input for classify-ing *** community-based cumulative algorithm was used to classify the pests in the existing *** proposed method uses the Entropy-ELM method with Whale Optimization to improve the classiﬁcation in detecting pests in *** Entropy-ELM method with the Whale Optimization Algorithm(WOA)is used for feature selection,enhancing mango pests’classiﬁcation *** Vector Machines(SVMs)are especially effective for classifying while users get var-ious classes in which they are *** are created as suitable classiﬁers to categorize any dataset in Big Data *** proposed Entropy-ELM-WOA is more capable compared to the existing systems.

关键词： Whale optimization algorithm Entropy-ELM feature selection pests detection support vector machine mango trees classiﬁcation

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：