检索结果-内蒙古大学图书馆

Innovative Approaches in Smart Technologies (ISAS), International Symposium on

作者： Ritu Aggarwal Eshaan Aggarwal Anurag Jain Tanupriya Choudhury Ketan Kotecha Computer Science and Engineering Seth Jai Prakash Mukand lal Institute of Engineering and Technology Radaur India Computer Science and Engineering (MMDU) Maharishi Markandeshwar Engineering College Mullana MMU India School of Computer Science University of Petroleum and Energy Studies (UPES) Dehradun Uttarakhand India CSE Dept. Symbiosis Institute of Technology Symbiosis International (Deemed University) Pune Maharashtra India Symbiosis Centre for Applied Artificial Intelligence Symbiosis Institute of Technology Symbiosis International (Deemed University)India Pune Maharashtra India

Cotton crop disease detection and classification based on images of leaves is a significant objective in agriculture. Cotton crops play a vital role in India. Every year, due to the attack of diseases, cotton production is decreasing. The main reasons for disease in plants are the pests, insects, and pathogens used; if they are not controlled on time, they affect productivity. In advancing digital image processing technologies, machine learning and other techniques that identify early disease detection in plants are proposed. This paper focuses on improving disease detection in cotton crop plants and leaves. Machine learning is used to identify and predict cotton plant disease using images and leaves collected in an uncontrolled environment. This study uses machine learning to review the different problems identified regarding plant disease. Various experimental configurations are investigated to analyze the impacts of different leaf classes, plant disease combinations, and their categories. Cotton crops are detected to classify the various types of cotton plants. Cotton plant diseases have a wide range of illnesses, from bacterial deficiency to bacterial, fungal, viral, and vitamin and nutrient deficiency. The proposed approach indicates the cotton crop plant in the leaf disease by their implementation results.

关键词：

来源：评论

学校读者我要写书评

暂无评论

USED: Universal Speaker Extraction and Diarization

arXiv

引用

arXiv 2023年

作者： Ao, Junyi Yıldırım, Mehmet Sinan Tao, Ruijie Ge, Meng Wang, Shuai Qian, Yanmin Li, Haizhou Shenzhen Research Institute of Big Data School of Data Science The Chinese University of Hong Kong Shenzhen518172 China Department of Electrical and Computer Engineering National University of Singapore Singapore119077 Singapore Saw Swee Hock School of Public Health National University of Singapore Singapore117549 Singapore Shenzhen Research Institute of Big Data Shenzhen518172 China Auditory Cognition and Computational Acoustics Lab Department of Computer Science and Engineering China MoE Key Laboratory of Artificial Intelligence AI Institute Shanghai Jiao Tong University Shanghai200240 China

Speaker extraction and diarization are two enabling techniques for real-world speech applications. Speaker extraction aims to extract a target speaker’s voice from a speech mixture, while speaker diarization demarcates speech segments by speaker, annotating ‘who spoke when’. Previous studies have typically treated the two tasks independently. In practical applications, it is more meaningful to have knowledge about ‘who spoke what and when’, which is captured by the two tasks. The two tasks share a similar objective of disentangling speakers. Speaker extraction operates in the frequency domain, whereas diarization is in the temporal domain. It is logical to believe that speaker activities obtained from speaker diarization can benefit speaker extraction, while the extracted speech offers more accurate speaker activity detection than the speech mixture. In this paper, we propose a unified model called Universal Speaker Extraction and Diarization (USED) to address output inconsistency and scenario mismatch issues. It is designed to manage speech mixtures with varying overlap ratios and variable number of speakers. We show that the USED model significantly outperforms the competitive baselines for speaker extraction and diarization tasks on LibriMix and SparseLibriMix datasets. We further validate the diarization performance on CALLHOME, a dataset based on real recordings, and experimental results indicate that our model surpasses recently proposed approaches. Copyright © 2023, The Authors. All rights reserved.

关键词： Frequency domain analysis

来源：评论

学校读者我要写书评

暂无评论

Joint-Motion Mutual Learning for Pose Estimation in Videos

arXiv

引用

arXiv 2024年

作者： Wu, Sifan Chen, Haipeng Yin, Yifang Hu, Sihao Feng, Runyang Jiao, Yingying Yang, Ziqi Liu, Zhenguang College of Computer Science and Technology Jilin University Changchun China A*STAR Singapore Georgia Institute of Technology Atlanta United States School of Artificial Intelligence Jilin University Changchun China The State Key Laboratory of Blockchain and Data Security Zhejiang University Hangzhou China Key Laboratory of Symbolic Computation and Knowledge Engineering Ministry of Education Jilin University Changchun China Institute of Blockchain and Data Security Hangzhou China

Human pose estimation in videos has long been a compelling yet challenging task within the realm of computer vision. Nevertheless, this task remains difficult because of the complex video scenes, such as video defocus and self-occlusion. Recent methods strive to integrate multi-frame visual features generated by a backbone network for pose estimation. However, they often ignore the useful joint information encoded in the initial heatmap, which is a byproduct of the backbone generation. Comparatively, methods that attempt to refine the initial heatmap fail to consider any spatiotemporal motion features. As a result, the performance of existing methods for pose estimation falls short due to the lack of ability to leverage both local joint (heatmap) information and global motion (feature) dynamics. To address this problem, we propose a novel joint-motion mutual learning framework for pose estimation, which effectively concentrates on both local joint dependency and global pixel-level motion dynamics. Specifically, we introduce a context-aware joint learner that adaptively leverages initial heatmaps and motion flow to retrieve robust local joint feature. Given that local joint feature and global motion flow are complementary, we further propose a progressive joint-motion mutual learning that synergistically exchanges information and interactively learns between joint feature and motion flow to improve the capability of the model. More importantly, to capture more diverse joint and motion cues, we theoretically analyze and propose an information orthogonality objective to avoid learning redundant information from multi-cues. Empirical experiments show our method outperforms prior arts on three challenging benchmarks. Copyright © 2024, The Authors. All rights reserved.

关键词： Motion estimation

来源：评论

学校读者我要写书评

暂无评论

ML-Based Feature Selection Technique for Imbalanced data Streams

ML-Based Feature Selection Technique for Imbalanced Data Str...

引用

International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT)

作者： Ayushi Pillay Monika Arya Sumit Kumar Sar Bhupesh Kumar Dewangan Tanupriya Choudhury Ketan Kotecha Sanjana Dewangan Bhilai Institute of Technology Durg Chhatisgarh India Department of Computer Science and Engineering School of Engineering OP Jindal University Raigarh India Symbiosis Institute of Technology Symbiosis International (Deemed University) Pune India Symbiosis Centre for Applied Artificial Intelligence Symbiosis Institute of Technology Symbiosis International (Deemed University) Pune India Pune India School of Science OP Jindal University Raigarh India

In the data stream classification domain, efficient feature selection (FS) plays an ornamental role in improving the overall performance of machine learning (ML) frameworks. Also, the imbalanced data significantly affects the classifier's performance. This study proposes an efficient ML framework with a hybrid FS based on the filter and wrapper approach. Another primary goal of this rigorous study is to show the efficacy of ML using the proposed FS in the classification of imbalanced data. First, feature subsets are rapidly selected using statistical (filter) FS methods. Modified correlation-based feature subset selection with greedy stepwise search and chi-square method has been implemented as filter-based methods. Then the selected feature subset is fine-tuned using the wrapper approach recursive feature elimination (RFE) and an ensemble of classifiers. The framework is built and tested on imbalanced datasets from the healthcare domain. The proposed framework has been compared to three datasets used in disease prediction. The framework effectively achieves better diagnostic and prediction performance with the optimal number of features without computational overheads.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Adaptive Convolutional Neural Network for Image Super-resolution

arXiv

引用

arXiv 2024年

作者： Tian, Chunwei Zhang, Xuanyu Wang, Tao Zhang, Yongjun Zhu, Qi Lin, Chia-Wen The School of Software Northwestern Polytechnical University Xi’an710129 China The National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology Xi’an710129 China The Yangtze River Delta Research Institute Northwestern Polytechnical University Taicang215400 China The School of Computer Science Northwestern Polytechnical University Xi’an710129 China The College of Computer Science and Technology Guizhou University Guiyang550025 China The School of Artificial Intelligence Nanjing University of Aeronautics and Astronautics Nanjing210016 China The Department of Electrical Engineering The Institute of Communications Engineering National Tsing Hua University Taiwan

Convolutional neural networks can automatically learn features via deep network architectures and given input samples. However, the robustness of obtained models may face challenges in varying scenes. Bigger differences in network architecture are beneficial to extract more diversified structural information to strengthen the robustness of an obtained superresolution model. In this paper, we proposed a adaptive convolutional neural network for image super-resolution (ADSRNet). To capture more information, ADSRNet is implemented by a heterogeneous parallel network. The upper network can enhance relation of context information, salient information relation of a kernel mapping and relations of shallow and deep layers to improve performance of image super-resolution. That can strengthen adaptability of an obtained super-resolution model for different scenes. The lower network utilizes a symmetric architecture to enhance relations of different layers to mine more structural information, which is complementary with a upper network for image super-resolution. The relevant experimental results show that the proposed ADSRNet is effective to deal with image resolving. Codes are obtained at https://***/hellloxiaotian/ADSRNet. Copyright © 2024, The Authors. All rights reserved.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Incorporating Transfer Learning in Engine Health Assessment for Accurate Remaining Useful Life Prediction

Incorporating Transfer Learning in Engine Health Assessment ...

引用

Sentiment Analysis and Deep Learning (ICSADL), International Conference on

作者： C. Karpagavalli Nelson A R. Raja Saranya E Manzoore Elahi M. Soudagar Christu Paul Ramaian Department of Artificial Intelligence and Data Science Ramco Institute of Technology Rajapalayam Tamil Nadu India Department of Mathematics Panimalar Engineering College Chennai Tamil Nadu India Department of Electronics and Communication Engineering Pandian Saraswathi Yadav Engineering College Sivagangai Tamil Nadu India Department of Computer Science and Business Systems Sri Eshwar College of Engineering Coimbatore Tamil Nadu India Department of VLSI Microelectronics Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Chennai Tamil Nadu India Department of Automobile Engineering Hindustan Institute of Technology and Science Chennai Tamil Nadu India

ISBN: (数字)9798350396157

ISBN: (纸本)9798350396164

Aircraft engines are critical, therefore any move to make them safer, more reliable, and fuel-efficient is highly encouraged. To solve the challenges of flight safety and maintenance expense during aircraft engine operation, a prognostics and health management system is deployed. This system is primarily concerned with flaw detection, health evaluation, and life prediction. Making decisions about aircraft engine operation and maintenance is strongly reliant on estimating the remaining useful life (RUL). This study provides an ensemble transfer learning model for RUL prediction using the C-MAPSS turbofan engine dataset. Pre-processing procedures such as filtering and normalization are used to improve the dataset. The proposed ensemble model for RUL prediction is then trained on the processed data. Some of the well-known transfer learning models that are compared include Xception, ResNet-50, and VGG-16. The ensemble model outperforms the other models, with the highest R2 value of 0.9901. Furthermore, the ensemble model is very effective; it can predict RUL in just 37 seconds, which suggests that it could be useful for keeping an eye on engine health and making decisions about aircraft.

关键词： Training Accuracy Atmospheric modeling Computational modeling Transfer learning Predictive models Maintenance

来源：评论

学校读者我要写书评

暂无评论

Comparison of Wearable and computer Vision Based Approaches to Knee Flexion Angle Measurement

Comparison of Wearable and Computer Vision Based Approaches ...

引用

IST-Africa

作者： Antony Gitau Victor Kulankash Gachathi Wanjema Ciira Wa Maina Centre for Data Science and Artificial Intelligence Dedan Kimathi University of Technology. P.O. BOX - PRIVATE BAG Dedan Kimathi - Nyeri Kenya Dept of Electrical & Electronic Engineering BSc Biomedical Engineering Kenyatta University Nairobi Kenya School of Health Sciences Dedan Kimathi University of Technology Nyeri Kenya

This paper outlines the development of a wearable device to measure knee joint flexion angles and compares its performance to a computer vision based human pose model. The objective is to advance techniques used in measuring joint flexion angles, which have several drawbacks such as relying heavily on the clinician for accuracy while also being invasive as they require physical contact between the patient and orthopaedic specialist. The proposed implementations integrate digital innovation in orthopaedic healthcare while being affordable and easy to deploy, making them marketable to clinicians and patients in developing countries. To evaluate the workability of the implementations, knee flexion angle measurements of a test subject were obtained using both techniques which yielded consistent and comparable flexion angle readings. Further development of this work can have a significant impact on the accessibility and affordability of innovative healthcare monitoring technologies, especially for patients in developing countries.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A data Completion Algorithm Based on Low-Rank Prior Knowledge for data-Driven Applications

引用

IEEE Transactions on Network and Service Management 2024年

作者： Tang, Jun Guo, Bing Shen, Yan Garg, Sahil Kaddoum, Georges Hossain, M. Shamim Sichuan University College of Computer Science Chengdu China Chengdu University of Information Technology School of Computer Science Chengdu China École de Technologie Supérieure Electrical Engineering Department MontréalQCH3C 1K3 Canada Chitkara University Institute of Engineering and Technology Chitkara University Centre for Research Impact and Outcome Punjab Rajpura140401 India Lebanese American University Artificial Intelligence and Cyber Systems Research Center Beirut03797751 Lebanon King Saud University College of Computer and Information Sciences Department of Software Engineering Riyadh12372 Saudi Arabia

Low rank tensor ring based data recovery algorithms have been widely used in data-driven consumer electronics to recover missing data entries in the collecting data pre-processing stage for providing stable and reliable service. However, traditional recovery methods often fail to utilize the abundant prior knowledge of data and the non-local self-similarity of the data, thus leading to the failure to effectively capture the spatial relationships within high-dimensional data to recover them accurately. To address these problems, we present a novel Non-local Self-similarity and Low-rank Prior Knowledge based tensor ring completion method. Firstly, we incorporate the BM3D denoising operator within a Plug-and-Play framework to exploit the self-similarity in the data. Then a logarithmic determinant function is integrated to distinguish singular values in the cyclic unfolding matrix of the tensor and adopts a tensor ring completion approach based on weighted nuclear norms. Finally, in order to evaluate the effectiveness of our proposed method, we conducted a series of experiments by using the missing image dataset and the missing traffic data dataset respectively, and the experimental results show that our method achieves the highest level in terms of data recovery accuracy. © 2024 IEEE.

关键词： data assimilation

来源：评论

学校读者我要写书评

暂无评论

Automatic Text Summarization Method Based on Improved Textrank Algorithm and K-Means Clustering

SSRN

引用

SSRN 2023年

作者： Liu, Wenjun Sun, Yuyan Yu, Bao Wang, Hailan Peng, Qingcheng Hou, Mengshu Guo, Huan Wang, Hai Liu, Cheng Chengdu611731 China School of Computer and Software Engineering XiHua University Chengdu610039 China School of Big Data and Artificial Intelligence Chengdu Technological University Chengdu611730 China Science and Technology on Communication Security Laboratory The 30th Research Institute of China Electronics Technology Group Corporation Chengdu610041 China

Automatic text summarization is to obtain a summary by compressing the text while retaining its important information, and then users can obtain the important content of the text by reading the summary. In the research literatures, the extraction method is widely used and is also one of the main research methods of most abstracts. However, this summary method still has some problems. The selection of the initial cluster center has not been carefully determined, and in articles with complex sentences, the sentence redundancy summarized is high. In order to solve the above problems, this paper proposes an automatic text summarization method based on improved TextRank algorithm and K-Means clustering. This method uses the improved BM25 model to calculate the BM25 similarity between sentences, and then uses the TextRank model to obtain the TR score of sentences. Then, the TR score is used to select the initial center of K-Means clustering. Finally, the summary is obtained through the class score and sentence score. The experimental results show that the proposed BM25Rank algorithm has a better effect on DUC2004 dataset. In conclusion, the improved TextRank and K-Means method proposed in this paper improve the accuracy of automatic text summarization and reduce the redundancy. © 2023, The Authors. All rights reserved.

关键词： K-means clustering

来源：评论

学校读者我要写书评

暂无评论

De-Modulation Techniques for Non-Coherent UWB Radio Systems

De-Modulation Techniques for Non-Coherent UWB Radio Systems

引用

International Conference on Computing and Networking Technology (ICCNT)

作者： G. Ravivarman Amritpal Sidhu K. K. Senthilkumar Saumya Goyal P S Raghavendra Rao Suvarna R. Bhagwat Department of Electrical and Electronics Engineering Karpagam Academy of Higher Education Coimbatore Centre of Research Impact and Outcome Chitkara University Rajpura Punjab India Department of Electronics and Communication Engineering Prince Shri Venkateshwara Padmavathy Engineering College Chennai Quantum University Research Center Quantum University School of Engineering and Technology Mechaical Engineering JAIN (Deemed to be University) Bangalore Karnataka Department of Artificial Intelligence & Data Science Vishwakarma Institute of Technology Pune India

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

Extremely Wideband (UWB) radio conversation systems provide the potential of high statistics fees and occasional strength intake from a small range of transceivers. The use of UWB for non-coherent packages is, in particular, attractive because of its low complexity and value. However, sign demodulation is a difficult challenge due to the presence of inter-symbol interference and the requirement of particular timing. Numerous demodulation strategies were proposed to address both the presence of inter-symbol interference and the want for precise timing. It includes time redundancy coding, frequency hopping, spreading, and linear codes. This paper critiques the one-of-a-kind demodulation techniques for Non-Coherent UWB Radio systems and goes over the key capabilities and blessings of each method. Time Redundancy Coding uses an encoding scheme that introduces temporal redundancy into the signal, allowing for premier demodulation at the receiver. Frequency Hopping employs a binary segment shift keying (BPSK) modulation for the UWB signal and utilizes unexpectedly converting carrier frequencies to randomize the interference in the channel. Spreading utilizes some form of spread-spectrum modulation for the UWB sign, which expands the bandwidth of the sign and enables the unfolding of the energy across an extensive frequency range.

关键词： Termination of employment Interference Frequency shift keying Frequency conversion Encoding Timing Complexity theory Wideband Demodulation Radiofrequency interference

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：