检索结果-内蒙古大学图书馆

PLD-Det: plant leaf disease detection in real time using an end-to-end neural network approach based on improved YOLOv7

引用

Neural Computing and Applications 2024年第34期36卷 21885-21898页

作者： Mehedi, Md Humaion Kabir Nawer, Nafisa Ahmed, Shafi Khan, Md Shakiful Islam Hasib, Khan Md Mridha, M.F. Alam, Md. Golam Rabiul Nguyen, Thanh Thi Department of Computer Science and Engineering BRAC University Dhaka Bangladesh Department of Computer Science and Engineering Bangladesh University of Business and Technology Dhaka Bangladesh Department of Computer Science American International University Dhaka Bangladesh Department of Data Science and AI Monash University Wellington Rd Clayton MelbourneVIC3800 Australia

In order to maintain sustainable agriculture, it is vital to monitor plant health. Since all species of plants are prone to characteristic diseases, it necessitates regular surveillance to search for any symptoms, which is utterly challenging and time-consuming. Besides, farmers may struggle to identify the type of plant disease and its potential symptoms. Hence, the interest in research like image-based computer-aided automated plant leaf disease detection by analyzing the early symptoms has increased enormously. However, limitations in the plant leaf image database, for instance, unfitting backgrounds, blurry images, and so on, sometimes cause underprivileged feature extraction, misclassification, and overfitting issues in existing models. As a result, we have proposed a real-time plant leaf disease detection architecture incorporating proposed PLD-Det model, which is based on improved YOLOv7 with the intention of assisting farmers while reducing the issues in existing models. The architecture has been trained on the widely used PlantVillage dataset, which resulted in an accuracy of 98.53%. Furthermore, SHapley Additive exPlanations (SHAP) values have been analyzed as a unified measure of feature significance. According to the experimental findings, the proposed PLD-Det model, which is an improved YOLOv7 architecture, outperformed the original YOLOv7 model in test accuracy by approximately 4%. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Seed

来源：评论

学校读者我要写书评

暂无评论

Advanced skin lesion detection via efficientNetB0 and vision transformer model with spatial-aware attention

引用

Multimedia Tools and Applications 2025年 1-19页

作者： Shaheen, Hera Singh, Maheshwari Prasad Singh, Amit Kumar Department of Computer Science and Engineering National Institute of Technology Patna Ashok Rajpath Bihar Patna800005 India

Due to the high incidence and possibly fatal nature of skin cancer, early identification is crucial for enhancing patient results. This paper presents a unique deep learning network, EfficientNetB0 ViT, to accurately classify skin lesions. The proposed method encompasses the scalability and efficiency of EfficientNetB0 with the global pattern recognition capabilities of Vision Transformers (ViT), strengthened by the Spatial-Aware Squeeze and Excitation (SASE) attention mechanism. This innovative technique improves classification performance by enabling the model to dynamically highlight the image’s most pertinent elements. SASE attention successfully recalibrates feature importance, reducing overfitting and improving generalization. This research highlights the effectiveness of SASE attention mechanism enhances skin lesion classification, providing a valuable tool for dermatologists in early diagnosis. The model outperformed numerous methods, achieving a high accuracy of 94.64% as well as average precision of 95%, average recall of 95%, and F1-scores of 95% on HAM10000 dataset. It achieves an accuracy of 93% © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Diseases

来源：评论

学校读者我要写书评

暂无评论

An Efficient Encryption and Compression of Sensed IoT Medical Images Using Auto-Encoder

引用

computer Modeling in engineering & sciences 2023年第2期134卷 909-926页

作者： Passent El-kafrawy Maie Aboghazalah Abdelmoty M.Ahmed Hanaa Torkey Ayman El-Sayed School of Information Technology and Computer Science Nile UniversityGiza12677Egypt Math and Computer Science Department Faculty of ScienceMenoufia UniversityMenoufia32511Egypt Department of Computer Engineering College of Computer ScienceKing Khalid UniversityAbha61421Saudi Arabia Computer Science and Engineering Department Faculty of Electronic EngineeringMenoufia UniversityMenouf32952Egypt

Healthcare systems nowadays depend on IoT sensors for sending data over the internet as a common *** ofmedical images is very important to secure patient *** these images consumes a lot of time onedge computing;therefore,theuse of anauto-encoder for compressionbefore encodingwill solve such a *** this paper,we use an auto-encoder to compress amedical image before encryption,and an encryption output(vector)is sent out over the *** the other hand,a decoder was used to reproduce the original image back after the vector was received and *** convolutional neural networks were conducted to evaluate our proposed approach:The first one is the auto-encoder,which is utilized to compress and encrypt the images,and the other assesses the classification accuracy of the image after decryption and *** hyperparameters of the encoder were tested,followed by the classification of the image to verify that no critical information was lost,to test the encryption and encoding *** this approach,sixteen hyperparameter permutations are utilized,but this research discusses three main cases in *** first case shows that the combination of Mean Square Logarithmic Error(MSLE),ADAgrad,two layers for the auto-encoder,and ReLU had the best auto-encoder results with a Mean Absolute Error(MAE)=0.221 after 50 epochs and 75%classification with the best result for the classification *** second case shows the reflection of auto-encoder results on the classification results which is a combination ofMean Square Error(MSE),RMSprop,three layers for the auto-encoder,and ReLU,which had the best classification accuracy of 65%,the auto-encoder gives MAE=0.31 after 50 *** third case is the worst,which is the combination of the hinge,RMSprop,three layers for the auto-encoder,and ReLU,providing accuracy of 20%and MAE=0.485.

关键词： Auto-encoder cloud image encryption IoT healthcare

来源：评论

学校读者我要写书评

暂无评论

Analyzing Sentiment and Unveiling Geopolitical Perspectives: A Comprehensive Study of Reddit Comments on the Contemporary Israel-Palestine Conflict 6

Analyzing Sentiment and Unveiling Geopolitical Perspectives:...

引用

6th International Conference on Electrical engineering and Information and Communication technology, ICEEICT 2024

作者： Nushin, Kazi Nubila Uz Zaman, Md. Shahid Ahmed, Mohiuddin Rajshahi University of Engineering & Technology Department of Computer Science & Engineering Rajshahi Bangladesh Rajshahi University of Engineering and Technology Department of Computer Science and Engineering Rajshahi Bangladesh

ISBN: (纸本)9798350385779

The ongoing Israel-Palestine conflict has triggered intense discussions on various social media platforms, reflecting the diverse perspectives and sentiments of users worldwide. In this study, we present a comprehensive analysis of Reddit comments related to the conflict. The aim is to understand the prevailing sentiments and geopolitical stances expressed by users in this dynamic and sensitive context. Our research comprised three key phases. In the initial stage, we conducted data labeling on a dataset containing 436,725 Reddit comments using the VADER tool, labeling them as Positive, Negative, and Neutral sentiments. Subsequently, our work progressed to the second stage, wherein Latent Dirichlet Allocation (LDA) was employed for topic modeling, shedding light on the geopolitical perspectives expressed by users. Transitioning to the third stage, we utilized Logistic Regression (LR), Stochastic Gradient Descent (SGD), Support Vector Classifier (SVC), and Naive Bayes (NB) as learning algorithms to classify sentiments, with a focus on Accuracy, F1 Score, Precision, and Recall for performance evaluation. Notably, our experimental findings highlighted that the Logistic Regression model demonstrated the highest accuracy, achieving 96.96%. © 2024 IEEE.

关键词： Statistics

来源：评论

学校读者我要写书评

暂无评论

A combinatorial algorithm for finding straight skeleton of a digital object

引用

Multimedia Tools and Applications 2025年 1-21页

作者： Maity, Anukul Dutt, Mousumi Biswas, Arindam Department of Computer Science and Engineering Narula Institute of Technology 81 Nilgunj Road Agarpara West Bengal Kolkata700109 India Department of Computer Science and Engineering St. Thomas’ College of Engineering and Technology 4 Diamond Harbour Road Kidderpore West Bengal Kolkata700023 India Department of Information Technology Indian Institute of Engineering Science and Technology Botanical Garden Road Shibpur West Bengal Howrah711103 India

The paper presents a combinatorial algorithm to find the straight skeleton of the inner isothetic cover of a digital object imposed on a uniform background grid. The isothetic polygon (orthogonal polygon) tightly inscribes the given digital object. The algorithm stated here finds the straight skeleton of any isothetic polygon (simple or non-simple) in O(nglogng) time in a single traversal by applying combinatorial rules, where n is the number of pixels on the boundary of the digital object and g is the grid size on which the digital object is imposed. To find the straight skeleton of non-simple orthogonal polygon, the polygon is divided into sub-parts which are a set of simple orthogonal polygons. The straight skeleton is generated for each of the sub-parts and the corresponding results are merged. The orthogonal polygon containing holes is cut at each holes and the polygon becomes hole-free. The algorithm to obtain the straight skeleton is applied on it. The disconnected hole parts are rejoined and the result is merged. The straight skeleton is useful shape descriptor of digital object. This algorithm stated here has applications in shape analysis. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Combinatorial optimization

来源：评论

学校读者我要写书评

暂无评论

An efficient botnet detection approach based on feature learning and classification

引用

Journal of Control and Decision 2023年第1期10卷 40-53页

作者： B.Padmavathi B.Muthukumar Department of Computer Science and Engineering Sathyabama Institute of Science and TechnologyChennaiIndia Computer Science and Engineering Easwari Engineering CollegeChennaiIndia DMI College of Engineering ChennaiIndia

Bot detection is considered a crucial security issue that is extensively analysed in various existingapproaches. Machine Learning is an efficient way of botnet attack detection. Bot detectionis the major issue faced by the existing system. This research concentrates on adopting a graphbasedfeature learning process to reduce feature dimensionality. The incoming samples arecorrectly classified and optimised using an Adaboost classifier with an improved grey wolfoptimiser (g-AGWO). The proposed IGWO optimisation approach is adopted to fulfil the multiconstraintissues related to bot detection and provide better local and global solutions (to satisfyexploration and exploitation). The extensive results show that the proposed g-AGWO model outperformsexisting approaches to reduce feature dimensionality, under-fitting/over-fitting andexecution time. The error rate prediction shows the feasibility of the given model to work over thechallenging environment. This model also works efficiently towards the unseen data to achievebetter generalization.

关键词： Botnet machine learning AdaBoost grey wolf optimisation exploration exploitation

来源：评论

学校读者我要写书评

暂无评论

Learning continuous network emerging dynamics from scarce observations via data-adaptive stochastic processes

引用

science China(Information sciences) 2024年第12期67卷 240-255页

作者： Jiaxu CUI Qipeng WANG Bingyi SUN Jiming LIU Bo YANG College of Computer Science and Technology Jilin University Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Public Computer Education and Research Center Jilin University Department of Computer Science Hong Kong Baptist University

Learning network dynamics from the empirical structure and spatio-temporal observation data is crucial to revealing the interaction mechanisms of complex networks in a wide range of domains. However,most existing methods only aim at learning network dynamic behaviors generated by a specific ordinary differential equation instance, resulting in ineffectiveness for new ones, and generally require dense *** observed data, especially from network emerging dynamics, are usually difficult to obtain, which brings trouble to model learning. Therefore, learning accurate network dynamics with sparse, irregularly-sampled,partial, and noisy observations remains a fundamental challenge. We introduce a new concept of the stochastic skeleton and its neural implementation, i.e., neural ODE processes for network dynamics(NDP4ND), a new class of stochastic processes governed by stochastic data-adaptive network dynamics, to overcome the challenge and learn continuous network dynamics from scarce observations. Intensive experiments conducted on various network dynamics in ecological population evolution, phototaxis movement, brain activity, epidemic spreading, and real-world empirical systems, demonstrate that the proposed method has excellent data adaptability and computational efficiency, and can adapt to unseen network emerging dynamics, producing accurate interpolation and extrapolation with reducing the ratio of required observation data to only about 6% and improving the learning speed for new dynamics by three orders of magnitude.

关键词： complex networks network dynamics emerging spatio-temporal dynamics neural processes

来源：评论

学校读者我要写书评

暂无评论

Newborn Weight Prediction And Interpretation Utilizing Explainable Machine Learning 3

Newborn Weight Prediction And Interpretation Utilizing Expla...

引用

3rd International Conference on Advancement in Electrical and Electronic engineering, ICAEEE 2024

作者： Chowdhury, Safiul Haque Mamun, Mohammad Hossain, Muhammad Minoar Hossain, Mohammad Ibrahim Iqbal, Md. Sadiq Kashem, Mohammod Abul Bangladesh University Department of Computer Science and Engineering Dhaka Bangladesh Mawlana Bhashani Science and Technology University Department of Computer Science and Engineering Dhaka Bangladesh Dhaka University of Engineering and Technology Department of Computer Science and Engineering Gazipur Bangladesh

ISBN: (纸本)9798350388282

Newborn weight refers to the weight of a baby at the time of birth. It is typically measured in units such as pounds and ounces or kilograms. The weight of a newborn is an essential metric used by healthcare professionals to assess the health of an infant and its development. This research emphasizes the prediction of newborn weight using Machine Learning (ML) and Explainable Artificial Intelligence (XAI). The proposed approach involves preprocessing a dataset encompassing various parameters of newborns and their mothers during pregnancy. The primary objective is to develop an automated system capable of accurately predicting the weight of a newborn. Employing a range of regression ensemble ML models, namely Bootstrap Aggregating regression (BAGGINGR), Random Forest Regression (RFR), Ensemble Voting Regression (VOTIINGR), Extreme Gradient Boosting Regression (XGBOOSTR), Stacked Generalization Regression (STACKINGR), and Gradient Boosting Decision Trees Regression (GBDTR). This research develops a cross-validation approach, and notably, the GBDT model emerges as the top performer, yielding impressive results with average metrics Mean Squared Error (MSE) of 247.26, Mean Absolute Error (MAE) of 12.29, R-squared (R2) of 0.23, Peak Signal-to-Noise Ratio (PSNR) at 20.99 dB, and Signal-to-Noise Ratio (SNR) at 48.65 dB. The outcomes of the research are interpreted through Local Interpretable Model-agnostic Explanations (LIME) and Shapley Additive Explanations (SHAP) in XAI, emphasizing the significance of interpreting these parameters. The proposed research has valuable insights to enhance the long-term health conditions of newborns, reduce mortality rates, and provide crucial support to healthcare professionals. © 2024 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

A comprehensive review and open issues on energy aware resource allocation in cloud

引用

International Journal of Web engineering and technology 2023年第4期17卷 353-371页

作者： Singh, Shubham Singh, Pawan Tanwar, Sudeep Department of Computer Science and Engineering Amity University Uttar Pradesh Lucknow Campus India Department of Computer Science and Engineering Institute of Technology Nirma University India

Cloud computing is an on-demand service resource that includes applications to data centres on a pay-per-use basis. While allocating resources, the node failure causes the cloud service failures. This reduces the quality of service along with ‘extra’ energy consumption. This survey makes an evaluation on 65 papers regarding energy aware resource allocation (RA) techniques in cloud. The performances and associated highest attainments of each contribution are described in this review. In addition, the tools used for implementing the RA techniques in each paper along with type of workloads were considered. We examine a number of current or developed energy-aware RA approaches for cloud settings in this study. We also provide a thorough analysis of the various energy-conscious RA and selection strategies for cloud-based virtual machines. Lastly, the clear sketch of issues and weakness are determined that may be supportive to resolve the existing problems in this aspect. Copyright © 2022 Inderscience Enterprises Ltd.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Automated Irrigation System Using Improved Fuzzy Neural Network in Wireless Sensor Networks

引用

Intelligent Automation & Soft Computing 2023年第1期35卷 853-866页

作者： S.Sakthivel V.Vivekanandhan M.Manikandan Department of Computer Science and Engineering Sona College of TechnologySalem 636005TamilnaduIndia Department of Computer Science and Engineering Adhiyamaan College of EngineeringHosur 635109TamilnaduIndia

Irrigation plays a significant role in various agricultural cropping methods deployed in semiarid and arid regions where valuable water applications and managing are considered crucial *** factors such as weather,soil,water,and crop data need to be considered for irrigation maintenance in an efficient besides uniform manner from multifaceted and different information-based systems.A Multi-Agent System(MAS)has been proposed recently based on diverse agent subsystems with definite objectives for attaining global MAS objective and is deployed on Cloud Computing paradigm capable of gathering information from Wireless Sensor Networks(WSNs)positioned in rice,cotton,cassava crops for knowledge discovery and decision *** radial basis function network has been used for irrigation ***,in recent work,the security of data has not focused on where intruder involvement might corrupt the data at the time of data transferring to the cloud,which would affect the accuracy of decision *** handle the above mentioned issues,an efficient method for irrigation prediction is used in this *** factors considered for decision making are soil moisture,temperature,plant height,root *** above-mentioned data will be gathered from the sensors that are attached to the *** data will be forwarded to the local server,where data encryption will be performed using Adaptive Elliptic Curve Cryptography(AECC).After the encryption process,the data will be forwarded to the *** the data stored in the cloud will be decrypted key before being given to the deci-sion-making ***,the uniform distribution-based fuzzy neural network is formulated based on the received data information in the decisionmaking *** decision regarding the level of water required for cropfields would be *** on this outcome,the water volve opening duration and the level of fertilizers required will be *** results demons

关键词： Irrigation multi-agent system precision irrigation accuracy elliptic curve cryptography encryption wireless sensor networks fertilizers

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：