检索结果-内蒙古大学图书馆

Farm-Level Smart Crop Recommendation Framework Using Machine Learning

Annals of Data science 2025年第1期12卷 117-140页

作者： Bhola, Amit Kumar, Prabhat Department of Computer Science and Engineering National Institute of Technology Patna Bihar Patna India

Agriculture is the primary source of food, fuel, and raw materials and is vital to any country’s economy. Farmers, the backbone of agriculture, primarily rely on instinct to determine what crops to plant in any given season. They are comfortable following customary farming practices and standards and are oblivious to the fact that crop yield is highly dependent on current environmental and soil conditions. Crop recommendations involve multifaceted factors such as weather, soil quality, crop production, market demand, and prices, making it crucial for farmers to make well-informed decisions. An improper or imprudent crop recommendation can affect them, their families, and the entire agricultural sector. Modern technologies like artificial intelligence, machine learning, and data science have emerged as efficient solutions to combat issues like declining crop production and lower profits. This research proposes a Smart Crop Recommendation framework that leverages machine learning to empower farmers to make informed decisions about optimal crop selection. The framework consists of two phases: crop filtration and yield prediction. Crops are filtered in the first phase using an artificial neural network based on local input parameters. The second phase estimates yield for filtered crops, considering the season, farm area, and location data. The final recommendation provides farmers with crops aimed at maximizing profit. The remarkable 99.10% accuracy of the framework is demonstrated through experimentation using artificial neural networks and the 0.99 R2 error metric for the random forest. The uniqueness of this framework lies in its distinctive focus on the farm level and its consideration of the challenges and various agricultural features that change over time. The experimental results affirm the effectiveness of the framework, and its lightweight nature enhances its practicality, making it an efficient real-time recommendation solution. © The Author(s), under exclusi

关键词： Crops

来源：评论

学校读者我要写书评

暂无评论

Neural network-guided motion blur estimation and patch representation method for face hallucination

引用

Neural Computing and Applications 2025年 1-28页

作者： Kumar, Banti Rajput, Shyam Singh Department of Computer Science and Engineering National Institute of Technology Patna Bihar Patna India

Face hallucination (or super-resolution) is a technique of generating high-resolution face images from low-resolution (LR) inputs. The current face hallucination algorithms struggle with motion blur, a common issue in captured images due to various factors, including camera defocusing, objects in motion, etc. To address this problem, a new motion blur robust face hallucination algorithm via neural network-guided motion blur estimation and patch representation (NNMEPR) is proposed in this paper. The NNMEPR algorithm first estimates the motion blur kernels i.e., length (λ) and angle (α) from a motion-blurred LR test faces using a neural network fitting approach with scale conjugate gradient (SCG) learning algorithm. Then, the estimated factor is embedded in the LR dictionary face images to make them compatible with the test images and mitigate the impact of motion blur from the reconstruction process. Moreover, the proposed algorithm employs a neighboring position patch representation in the correlation coefficient calculation process to sustain the sharp edges and texture details in the resulting HR face images. Experiments on standard datasets and locally captured real-world faces demonstrate the superior performance of the proposed NNMEPR algorithm over the existing state-of-the-art techniques. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Motion estimation

来源：评论

学校读者我要写书评

暂无评论

MTUNet + + : explainable few-shot medical image classification with generative adversarial network

引用

Multimedia Tools and Applications 2025年第11期84卷 9219-9239页

作者： Titoriya, Ankit Kumar Singh, Maheshwari Prasad Singh, Amit Kumar Department of Computer Science and Engineering National Institute of Technology Patna India

Medical imaging, a cornerstone of disease diagnosis and treatment planning, faces the hurdles of subjective interpretation and reliance on specialized expertise. Deep learning algorithms show improvements in automating medical image analysis, reducing radiologists' burden, and potentially enhancing patient outcomes. However, these algorithms require substantial quantities of high-quality labelled data for effective training and refinement. This paper proposes an innovative approach that harnesses few-shot learning (FSL) and generative adversarial networks (GANs) to overcome conventional methods' limitations in medical image classification. FSL, capable of learning from limited labelled examples, holds promise for scenarios where labelled data is scarce. However, the lack of interpretability in existing FSL models impedes their clinical adoption. To tackle this, this paper proposes a explainable FSL network, "MTUNet + +," which integrates an attention mechanism to emphasize relevant regions in medical images. Furthermore, integrating a generative adversarial network, enhances the performance of MTUNet + + by generating synthetic medical images. Systematically eliminating misleading synthetic images improves the reliability and accuracy of medical image classification. Empirical evaluation on benchmark datasets underscores the effectiveness of the approach, achieving 85.19% and 69.28% accuracy for the HAM10000 and Kvasir datasets, respectively. This paper contributes to advancing AI-driven solutions in clinical practice, facilitating enhanced patient care and streamlined workflows within real-world healthcare settings. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

A verifiable multi-secret image sharing scheme based on DNA encryption

引用

Multimedia Tools and Applications 2025年第4期84卷 1967-1983页

作者： Chattopadhyay, Arup Kumar Saha, Sanchita Nag, Amitava Singh, Jyoti Prakash Department of Computer Science and Engineering Central Institute of Technology Kokrajhar Kokrajhar India Department of Computer Science and Engineering National Institute of Technology Patna Bihar Patna India

A multi-secret image sharing (MSIS) scheme facilitates the secure distribution of multiple images among a group of participants. Several MSIS schemes have been proposed with a (n, n) structure that encodes secret input images into n share images. This encoding ensures that if a subset of participants collude, no information about the secret images can be revealed. The secret images can only be recovered if all participants cooperate. Most of these schemes are built using basic Boolean operations, primarily XOR operations. MSIS schemes that rely on Boolean logic offer various benefits compared to other methods, such as the ability to recover data without any loss, no increase in pixel size, and efficient computation. This article presents a verifiable MSIS scheme that relies on a secure two-variable one-way function and a secure hash function. The scheme allows for independent verification of both the dealer and the participants to detect any cheating attempts. The proposed scheme utilizes the Deoxyribonucleic acid (DNA) encryption technique to leverage the inherent benefits of DNA computing, such as enhanced security, fast computation, minimal storage needs, and low power usage. The share images produced by the proposed scheme exhibit an average entropy of 7.99, indicating that the generated 8-bit share images are completely random. Additionally, the correlation coefficient is approximately 0.005, suggesting that neighboring pixels in the images are not related to each other. The proposed scheme can detect any instance of cheating, whether it is done by the dealer or by a participant. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： DNA

来源：评论

学校读者我要写书评

暂无评论

A brief review on Quantum Key Distribution Protocols

引用

Multimedia Tools and Applications 2025年 1-40页

作者： Kumar, Mandeep Mondal, Bhaskar Department of Computer Science & Engineering National Institute of Technology Patna Bihar Patna800005 India

Development in Quantum computing paves the path to Quantum key distribution (QKD) by using the principles of quantum physics. QKD enables two remote parties to produce and share secure keys while removing all computing constraints on an adversary. The basic physics laws are used to identify any outside parties eavesdropping on the key exchange. In recent years many revolutionary developments in the field of QKD have been developed to overcome security and networking constraints. This survey provides an overview of the QKD protocol’s evolution and quantum network architecture. The paper also demonstrates QKD deployment techniques and elements of the QKD network. It also highlights ongoing design challenges by considering security and error estimation and correction, in contrast to studies concentrating on optical channels and equipment. Finally, this paper examines the possible directions for future research and offers design principles to guide the development of QKD and its related area. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Quantum computers

来源：评论

学校读者我要写书评

暂无评论

Maximizing Emotion Recognition Accuracy with Ensemble Techniques on EEG Signals

引用

Recent Advances in computer science and Communications 2024年第5期17卷 24-36页

作者： Jha, Sonu Kumar Suvvari, Somaraju Kumar, Mukesh Department of Computer Science and Engineering National Institute of Technology Patna Patna800005 India

Background: Emotion is a strong feeling such as love, anger, fear, etc. Emotion can be recognized in two ways, i.e., External expression and Biomedical data-based. Nowadays, various research is occurring on emotion classification with biomedical data. Aim: One of the most current studies in the medical sector, gaming-based applications, education sector, and many other domains is EEG-based emotion identification. The existing research on emotion recognition was published using models like KNN, RF Ensemble, SVM, CNN, and LSTM on biomedical EEG data. In general, only a few works have been published on ensemble or concatenation models for emotion recognition on EEG data and achieved better results than individual ones or a few machine learning approaches. Various papers have observed that CNN works better than other approaches for extracting features from the dataset, and LSTM works better on the sequence data. Methods: Our research is based on emotion recognition using EEG data, a mixed-model deep learning methodology, and its comparison with a machine learning mixed-model methodology. In this study, we introduced a mixed model using CNN and LSTM that classifies emotions in valence and arousal on the DEAP dataset with 14 channels across 32 people. Result and Discussion: We then compared it to SVM, KNN, and RF Ensemble, and concatenated these models with it. First preprocessed the raw data, then checked emotion classification using SVM, KNN, RF Ensemble, CNN, and LSTM individually. After that with the mixed model of CNN-LSTM, and SVM-KNN-RF Ensemble results are compared. Proposed model results have better accuracy as 80.70% in valence than individual ones with CNN, LSTM, SVM, KNN, RF Ensemble and concatenated models of SVM, KNN and RF Ensemble. Conclusion: Overall, this paper concludes a powerful technique for processing a range of EEG data is the combination of CNNs and LSTMs. Ensemble approach results show better performance in the case of valence at 80.70% and 78.24

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Trajectory-Prediction Techniques for Unmanned Aerial Vehicles (UAVs): A Comprehensive Survey

引用

IEEE Communications Surveys and Tutorials 2025年第3期27卷 1867-1910页

作者： Shukla, Pushpak Shukla, Shailendra Kumar Singh, Amit Motilal Nehru National Institute of Technology Allahabad Department of Computer Science and Engineering Prayagraj211004 India National Institute of Technology Patna Computer Science and Engineering Department Patna800005 India

Unmanned Aerial Vehicles (UAVs) have witnessed remarkable significance in diverse sectors, ranging from environmental monitoring, infrastructure inspection, disaster response, wildlife conservation, surveillance, and reconnaissance missions. It is crucial to predict their future states to enable UAVs’ safe and efficient operation in dynamic environments. UAV trajectory planning is a crucial aspect of UAV operations, as it determines how the drone will navigate, perform tasks, and avoid obstacles. UAVs can be operated with varying degrees of autonomy, and they can be controlled by humans or autonomously via onboard autopilot software. While existing research has extensively focused on trajectory planning methodologies for UAVs, there is a noticeable gap in the literature concerning the integration of predictive capabilities into trajectory planning, highlighting the need for a comprehensive review of methodologies in UAV trajectory prediction connected with the associated realm of trajectory planning. This article provides a comprehensive and comparative analysis of trajectory prediction methods tailored for autonomous UAVs. Beginning with a precise problem definition and algorithm categorization, our study delves into evaluating methodologies rooted in conventional mathematical models, classical machine learning, deep learning, and reinforcement learning models. © 1998-2012 IEEE.

关键词： Trajectory Autonomous aerial vehicles Data models Predictive models Internet of Things Drones Wireless sensor networks Trajectory planning Safety Optimization

来源：评论

学校读者我要写书评

暂无评论

Exploring ensemble optimized voting and stacking classifiers through Cross-validation for early detection of suicidal ideation

引用

Journal of Intelligent and Fuzzy Systems 2024年第5-6期47卷 335-349页

作者： Shukla, Shiv Shankar Prasad Singh, Maheshwari Prasad Department of Computer Science Engineering National Institute of Technology Patna Bihar Patna India

Detecting behavioral changes associated with suicidal ideation on social media is essential yet complex. While machine learning and deep learning hold promise in this regard, current studies often lack generalizability due to single dataset reliance. Traditional embedding techniques struggle with semantic analysis,leading to challenges in achieving high accuracy models and conventional validation methods have data drift limitations. To address these challenges, this study proposes a novel evaluation approach using natural language processing across diverse platforms like Twitter and Reddit. By integrating BERT embedding, adept at handling semantic nuances, with an optimized Stacked Classifier combining different base classifiers and XGBoost as the meta-classifier, the model excels in swiftly detecting signs of suicidal ideation compared to the Voting Classifier, i.e., the combination of Decision Tree, Random Forest, Gradient Boost and XGBoost and several machine learning models. Additionally, the study explores advanced embedding techniques like MUSE and LLM, and deep learning models including Bi-LSTM, Bi-GRU, and Text-CNN for *** ensemble approach aims to create a model that is not only interpretable but also robust, reducing computational complexity and enhancing resilience against noisy data - common challenges faced in text classification tasks. Through K-fold validation, which involves partitioning the dataset into k equal-sized subsets or "folds"and training the model k times, using k-1 folds for training and one-fold for testing each time, the proposed model achieves impressive accuracy rates of 97% on Reddit and 96% on Twitter datasets, underscoring its effectiveness in identifying suicidal ideation across social media platforms. © 2024 - IOS Press. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Deep feature-support vector machine based hybrid model for multi-crop leaf disease identification in Corn, Rice, and Wheat

引用

Multimedia Tools and Applications 2025年第8期84卷 4751-4771页

作者： Bhola, Amit Kumar, Prabhat Department of Computer Science and Engineering National Institute of Technology Patna Bihar India

Corn, Rice, and Wheat serve as primary staple foods globally, playing a pivotal role in the economies of numerous countries. Despite their paramount importance, these cereal crops face susceptibility to various diseases, posing a significant threat to production. Timely and accurate disease diagnosis is imperative to mitigate these risks. In response, a range of Machine Learning (ML) and Deep Learning (DL) models have been devised to automate the identification of diseases impacting cereal crops. While DL excels with abundant data, traditional ML may outperform DL with limited data. This study proposes a hybrid model integrating deep transfer learning and ML techniques for identifying crop diseases. The model utilizes pre-trained DenseNet201 weights to extract features, which are subsequently fed into a Support Vector Machine (SVM) for identification. Experimental results show that the hybrid model outperforms several benchmark counterparts. The DenseNet201 deep feature combined with SVM exhibits superior performance, achieving an impressive accuracy score of 87.23% with just 20.2M parameters. The model underwent comprehensive performance assessment using precision, recall, F1 score, and overall accuracy metrics. It effectively classifies healthy and diseased categories for each crop, achieving accuracies of 99.82%, 98.75%, and 84.15% for Corn, Wheat, and Rice, respectively. Moreover, the model’s remarkable performance and lightweight nature make it a practical and efficient real-time crop disease detection solution, emphasizing its relevance and significance for farmers and researchers in the agricultural community. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Antenna Optimization using Machine Learning Algorithms and their Applications: A Review

引用

Journal of engineering science and technology Review 2024年第2期17卷 128-144页

作者： Pandey, Anil Kumar Singh, M.P. Department Of Computer Science & Engineering National Institute Of Technology Patna India

Antenna optimization using machine learning is a rapidly evolving field that leverages the power of artificial intelligence to design and improve antenna systems. Antenna optimization is a process of modifying antenna parameters to achieve desired performance metrics, such as gain, bandwidth, radiation pattern, and impedance matching. This paper presents a review of the most advanced development in antenna design and optimization by using machine learning techniques. The aim of this survey is to focus on different machine learning optimization techniques and their optimization capability with efficiency challenges. A deep outline from literature survey on optimization of antennas using machine learning are presented and listing various optimization algorithms and procedures that are applied to produce desired antenna characteristics and specifications. Firstly, a brief introduction of machine learning and its algorithms, later a quick explanation of antenna optimization process followed by an arranged introduction of different types of printed antenna designs using machine learning algorithm are reported. The methods emphasized in this survey have probably an effect on the imminent advancement of antennas for a variety of wireless applications. © (2024), (International Hellenic University – School of science). All Rights Reserved.

关键词： Microstrip antennas

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：