检索结果-内蒙古大学图书馆

An effective object detection via BS2ResNet and LTK-Bi-LSTM

Multimedia Tools and Applications 2025年 1-20页

作者： V, Arulalan V, Premanand Kumar, Dhananjay Department of Computing Technologies SRM Institute of Science and Technology Chengalpattu District Tamil Nadu Kattankulathur603203 India Department of Computer Science and Engineering Vellore Institute Technology Chennai Campus Tamil Nadu Chennai India Department of Information Technology Anna University MIT Campus Tamil Nadu Chennai India

Video analytics faces complex challenges in object detection and classification. Deep learning based approaches have achieved remarkable success in past decade. However, existing object identification models that utilize backbone’s core features still present challenges due to their lack of semantic information. To address these issues, a novel object detection and classification framework utilizing Batch normalization and Softswish activation adapted ResNet (BS2ResNet) and Logistic Tanh Kaiming Bi-directional Long Short Term Memory (LTK-Bi-LSTM) techniques was proposed. The framework employs frame conversion, noise removal, and contrast elevation during frame pre-processing, followed by background subtraction using the Supreme Distance-centered Fuzzy C-Means (SD-FCM) clustering algorithm, and edge detection using the Hyperbolic Tangent Kernel Canny Edge Detector (HTKCED). BS2ResNet is then used for object detection, and features are extracted and passed to the LTK-Bi-LSTM neural network for object classification. The proposed system was found to improve object detection and classification accuracy, outperforming existing techniques. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Recent Advances in Analysis and Detection of Tuberculosis System in Chest X-Ray Using Artificial Intelligence (AI) Techniques: A Review

引用

Current Materials science 2023年第1期16卷 43-51页

作者： Suchitra, Shanmugam Ibrahim, S. Jafar Ali Sathya, Mariappan Sahini, Varsha Chakravarthy, N. Surya Kalyan Kumar, Vaneet Saruchi Department of Computer Science and Business Systems SRM Institute of Science and Technology Kattankulathur Chennai India Department of IOT SCOPE Vellore Institute of Technology Tamilnadu Vellore India AMC Engineering College Karnataka Bengaluru India Department of Computer Science Engineering CT Institute of Engineering Management and Technology Punjab Jalandhar India Data Science QIS College of Engineering and Technology Andhra Pradhesh Ongole India Department of Applied Sciences CT Institute of Engineering Management and Technology Shahpur Campus Punjab Jalandhar India Department of Biotechnology CT Institute of Pharmaceutical Sciences Shahpur Campus Punjab Jalandhar India

Mycobacterium tuberculosis causes tuberculosis (TB), a bacterial illness. Although germs are most typically found in the lungs, they can affect other sections of the body as well. Tuberculosis is one of the primary causes of mortality in both developed and developing nations, necessitating worldwide attention. Even though TB may be prevented in the majority of instances if discovered and treated early, the number of deaths caused by the disease is quite high. There has been a significant increase in interest and research activity in TB detection in recent years. The new advancement in the field of AI technology may be able to assist them in overcoming these development gaps. computer-Aided Detection and Diagnosis (CADD) aids in the diagnosis of diseases by analysing symptoms and X-ray images of patients. Many solutions are currently be-ing developed to improve the effectiveness of TB diagnosis classification using AI and DL ap-proaches. Although a variety of TB detection techniques have been developed, there is no common-ly acknowledged method. The purpose of this study is to give a survey on Tuberculosis Detection. It also emphasises the difficulty and complexity of the Tuberculosis Detection System's design. © 2023 Bentham science Publishers.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

T5LSTM-RNN based Text Summarization Model for Behavioral Biology Literature

引用

Procedia computer science 2023年 218卷 585-593页

作者： Shivangi Chaurasia Debalay Dasgupta Rajeshkhannan Regunathan School of Computer Science and Engineering Vellore Institute of Technology Vellore Professor School of Computer Science and Engineering Vellore Institute of Technology Vellore

Behavioral biology is one of the crucial and trending topics these days which needs proper attention by scholars for rapid development in this field. Therefore, the purpose of this paper is to ease the process of collecting all kinds of relevant and vital data available on the internet regarding this topic in different forms of media, such as news articles, research papers, and YouTube lecture videos, all into one place into a single document in a proper summarized form. For proper training of the LSTM model, the lengthy video and journal datasets are pre-processed using the T5 transformer model to generate a uniform training dataset. So, in this work, a comprehensive approach is proposed based on an abstractive form of text summarization using the seq2seq encoder-decoder model combined with a stacked LSTM layer with an attention mechanism and a T5 transformer model pre-processor. Therefore, a proper hybrid model, T5LSTM-RNN, is implemented to generate the summarized data.

关键词： Seq2Seq Text Summarization Abstractive Summary T5 Transformer Behavior Biology LSTM RNN

来源：评论

学校读者我要写书评

暂无评论

Insider Attack Detection Using Deep Belief Neural Network in Cloud Computing

引用

computer Systems science & engineering 2022年第5期41卷 479-492页

作者： A.S.Anakath R.Kannadasan Niju P.Joseph P.Boominathan G.R.Sreekanth School of Computing E.G.S.Pillay Engineering CollegeNagapattinam611002TamilnaduIndia Department of Software Systems School of Computer Science and Engineering(SCOPE)Vellore Institute of Technology UniversityVellore632014TamilnaduIndia Department of Computer Science CHRIST(Deemed to be University)Bengaluru560029India Department of Information Security School of Computer Science and Engineering(SCOPE)Vellore Institute of Technology UniversityVellore632014TamilnaduIndia Department of Computer Science and Engineering Kongu Engineering CollegeErode638060TamilnaduIndia

Cloud computing is a high network infrastructure where users,owners,third users,authorized users,and customers can access and store their information *** use of cloud computing has realized the rapid increase of information in every field and the need for a centralized location for processing *** cloud is nowadays highly affected by internal threats of the *** applications such as banking,hospital,and business are more likely affected by real user *** intruder is presented as a user and set as a member of the *** becoming an insider in the network,they will try to attack or steal sensitive data during information sharing or *** major issue in today's technological development is identifying the insider threat in the cloud *** data are lost,compromising cloud users is *** and security are not ensured,and then,the usage of the cloud is not *** solutions are available for the external security of the cloud ***,insider or internal threats need to be *** this research work,we focus on a solution for identifying an insider attack using the artificial intelligence *** insider attack is possible by using nodes of weak users’*** will log in using a weak user id,connect to a network,and pretend to be a trusted ***,they can easily attack and hack information as an insider,and identifying them is very *** types of attacks need intelligent solutions.A machine learning approach is widely used for security *** date,the existing lags can classify the attackers *** information hijacking process is very absurd,which motivates young researchers to provide a solution for internal *** our proposed work,we track the attackers using a user interaction behavior pattern and deep learning *** usage of mouse movements and clicks and keystrokes of the real user is stored in a *** deep belief neural

关键词： Cloud computing security insider attack network security privacy user interaction behavior deep belief neural network

来源：评论

学校读者我要写书评

暂无评论

Aviation Safe Landing Prediction Utilizing Long-Short Term Memory Algorithm 6

Aviation Safe Landing Prediction Utilizing Long-Short Term M...

引用

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

作者： Krishna, Raji Sariki, Tulasi Prasad Sigamsetty, Siri Mahalakshmi Kumar, G. Bharadwaja Benil, T. Vellore Institute of Technology School of Electrical Engineering Chennai Campus Chennai India Vellore Institute of Technology School of Computer Science Engineering Chennai Campus Chennai India

ISBN: (纸本)9798350394702

The aviation industry is significantly dependent on sophisticated technology, such as engines, avionics, and air traffic control systems, to assure the safety and efficiency of its operations. A significant proportion of commercial aviation accidents may have been averted by implementing a flying procedure wherein an incoming aircraft discontinues its landing attempt and reenters the queue for landing. The timely implementation of this approach can substantially impact mitigating the overall accident rate within the aviation industry. Hence, the main aim of this article is to effectively forecast flight landings, improve safety measures and enable informed decision-making in complex situations for both autonomous and human-operated aircraft using a Deep Learning algorithm such as long short term memory (LSTM). The performance of LSTM compared with machine learning (ML) technique like Support Vector Machines (SVM) and logistic regression algorithm in terms of sensitivity and specificity © 2023 IEEE.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Pothole Detection Using YOLOv3 Model

Pothole Detection Using YOLOv3 Model

引用

2023 IEEE International Students' Conference on Electrical, Electronics and computer science, SCEECS 2023

作者： Sathya, R. Saleena, B. Prakash, B. School of Computer Science and Engineering Vellore Institute of Technology Chennai Campus Chennai India Srm Institute of Science and Technology Department of Computer Science and Engineering Ramapuram Campus Chennai India

ISBN: (纸本)9798350398748

When it comes to moving people and goods around a country, road transports are like the heart and arteries. It is essential that roads need be kept in good shape at all times. Most of the times, potholes on the road surface causes major damages and they must be repaired for maintaining the roads in usable condition. Countries like India, where roads span millions of kilometers and require constant upkeep, it may be difficult to locate potholes. Since this is the case, rapid, accurate, and real-time automated pothole detection is essential. The primary motivation of this proposed work is to explain the pothole prediction models that are created using object detection algorithm called YOLOv3. The YOLOv3 model is trained on a dataset of potholes, and the results are analyzed and compared with variants of YOLO by computing the model's precision, recall, and mAP. When compared to other models, the mAP of this one is improved while the computing cost is reduced significantly. © 2023 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Accessing the Efficiency of Machine learning Models for Crop Categorization Based on Regional Attributes in India

引用

Procedia computer science 2025年 258卷 2554-2562页

作者： Hemanth Thulasiraman Aryan Jain Tanmay Kukreja Rajkumar Rajasekaran School of Computer Science and Engineering Vellore Institute of Technology Vellore Tamil Nadu India

This comprehensive model comparison demonstrates assessment of various machine learning methods designed for identifying the different crops grown in India and categorizing them, tailoring the classification to the specific attributes of each State. Some drawbacks observed in previously implemented systems and studies arise from their singular focus on a specific parameter—be it moisture, temperature, soil pH, etc.—for predicting the adequacy of crop growth. This project focuses on a multitude of factors like soil quality indicated by parameters such as pH, Nitrogen (N), Phosphorus (P), Potassium (K) content, in addition to temperature and rainfall. The primary models utilized and compared are Decision Tree, K-Nearest Neighbors, Random Forest, Extreme Gradient Booster (XGBoost), and Naïve Bayes. K Nearest Neighbors emerges as the best model with an f1-score of 97% and a accuracy of 94%. The research investigates the connection between model complexity and accuracy, demonstrating how the choice of a particular model influences practical outcomes in crop prediction. The models underwent training and validation using a dataset structured in CSV format, comprising a total of 1,000,000 crop records. This training was conducted within a supervised learning framework. Comparison metrics for these models were derived from standardized parameters, encompassing losses, accuracy, and other pertinent metrics

关键词： Deep learning machine learning crop yield prediction

来源：评论

学校读者我要写书评

暂无评论

Multimodal Deep Learning for Cardiovascular Risk Stratification: Integrating Retinal Biomarkers and Cardiovascular Signals for Enhanced Heart Attack Prediction

引用

IEEE Access 2025年 13卷 99672-99689页

作者： K. Sathya G. Magesh School of Computer Science Engineering and Information Systems Vellore Institute of Technology Vellore India

Cardiovascular diseases, particularly myocardial infarction, have remained a significant cause of death around the world. Therefore, dedicated non-invasive risk prediction frameworks are required. Conventional measures for risk prediction among most heterogeneous patients, such as the Framingham Risk Score and ASCVD calculator, are often less precise and poorly generalizable as they fail to capture individual variations in the mechanisms. This study seeks to present a new multimodal deep learning model developed for cardiovascular risk stratification by fusing retinal microvascular features with cardiovascular physiological signals. The pipeline includes a Hierarchical Retinal Vessel Graph Transformer (HRV-GT) for graph-based vascular representation, Spectro-Temporal Cardiovascular Transformer (STC-T) for capturing short- and long-term signal variability, Sparse Manifold Multimodal Fusion (SMM Fusion) for joint alignment of features, Evolutionary Feature Selection with Clinical Prior Constraints (EFS-CP), and a Contrastive Hierarchical Risk Classifier (CHRC-Net) for decision modeling structured. The model was evaluated across the UK Biobank and SEED datasets (n = 55,000), achieving an AUC of 0.97, sensitivity of 91%, and specificity of 89%, with more than 35% better performance than existing models in reducing misclassification. Clinically interpretable biomarkers that combine high computational efficiency (inference time $\approx \; 37$ ms) affirm its design for real-world deployment in preventive cardiovascular screening processes.

关键词： Feature extraction Deep learning Biomarkers Biological system modeling Transformers Predictive models Imaging Heart rate variability Accuracy Retinal vessels

来源：评论

学校读者我要写书评

暂无评论

Framework for Face recognition and Scene Description using Deep Learning for Visually Challenged people 1

Framework for Face recognition and Scene Description using D...

引用

1st IEEE International Conference on Emerging Research in Computational science, ICERCS 2023

作者： Sharon, Felicia Sellamuthu, Suseela Nachiyappan, S. Akila, V. Vellore Institute of Technology School of Computer Science and Engineering Tamil Nadu Chennai India Gokaraju Rangaraju Institute of Engineering and Technology School of Computer Science and Engineering Telangana Hyderabad India

ISBN: (纸本)9798350359756

In today's world, independent living becomes increasingly crucial for those who are visually impaired, as they encounter numerous challenges. Visually impaired people are at a disadvantage since they need manual help to compel information about their surroundings. Since many tasks require visual information, visually impaired persons find it difficult to complete many of life's essential chores, such as going for a stroll, eating, and conversing with a group of people. As technology has advanced, helping the blind is now possible. This project offers an advanced solution to assist visually challenged individuals, taking advantage of recent advancements in Artificial Intelligence (AI) Technologies and Deep Learning frameworks. This project proposes a sophisticated system to help visually challenged people. The system uses a novel strategy of combining Deep Learning Faster R-CNN ResNet-50 FPN for object detection, CNN ResNet-50 for Face recognition, and CNN ResNet-50 and LSTM for Image Captioning and adding some more objects to the existing dataset to customize it for visually impaired people. Then a voice message is generated to assist the Visually Impaired person. Furthermore, each of the models used in the system is compared with two other models to determine the most effective model. Of all the three models used for Face recognition, Image Captioning, and Object detection, the proposed system performs the best with 80.6%, 0.493(BLEU-1), and 92.7% respectively. © 2023 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Automatic Emotion Recognition System: A cross Culture Study between Tamil and Russian Speaking Children 3

Automatic Emotion Recognition System: A cross Culture Study ...

引用

3rd International Conference on Advanced Research in Computing, ICARC 2023

作者： Nersisson, Ruban Bhuyan, Prithiraj Mekala, A. Mary Lyakso, Elena Vellore Institute of Technology School of Electrical Engineering Vellore India Vellore Institute of Technology School of Computer Science and Engineering Vellore India St. Petersburg State University The Child Speech Research Group Russia

ISBN: (纸本)9798350347371

Speech based communication contains both linguistic and paralinguistic information. It conveys emotions not only through its verbal communication, but also through voice characteristics such as pitch, volume, and stress. The objective of this study is to convey the child's emotion to a caretaker. To achieve this, a cross-cultural study of four emotions "joy - neutral (calm state) - sadness - anger"in children aged 8-12 years is considered. The data collection is carried out in multiple modes like acting speech in Russian and spontaneous and acting speech in Tamil language. SVM and MLP classitiers were used by the Russian specialist to train and validate speech data of both languages. The accuracy achieved by the Russian team is 84.1% and the Indian specialists used CNN and obtained the training and validation accuracy of 90% and >68% over 75 Epoch and used 7-layer CNN and obtained the training and validation accuracy of 95% and > 65% over 50 Epoch. The goal of the study is to see how well Russian and Indian specialists can recognise the emotional condition of Russian and Indian children based on their speech, and the results are correlated with computer-based emotion classification algorithms with various machine learning and deep net architectures. © 2023 IEEE.

关键词： Emotion Recognition

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：