检索结果-内蒙古大学图书馆

2025 International Conference on Pervasive Computational Technologies, ICPCT 2025

作者： Kulkarni, Supritha Suhaas Rao, S. Chopra, Tushar Kumar Yeshwanth, S.P. Bhanushree, K.J. Suneetha, K.R. Bangalore Institute of Technology Dept. of Computer Science and Engineering Bengaluru India

ISBN: (纸本)9798331508685

Kidney stones are a common and impactful urological condition, with timely and accurate diagnosis being critical to prevent complications such as renal damage and avoid invasive treatments. This study explores the use of advanced deep learning models, including Vision Transformers (ViT), Swin Transformer and Convolutional Neural Networks (CNNs) - specifically XResNet-50, VGG-16, MobileNetV2 and AlexNet - to enhance kidney stone detection from CT scan images. Leveraging a robust dataset of annotated CT scans in JPG format, the approach employs transfer learning which leverages features learnt from the ImageNet dataset to optimize model performance, focusing on classifying cases as either "Kidney Stone Present"or "Kidney Stone Absent". To further improve diagnostic reliability, the Grad-CAM (Gradient-weighted Class Activation Mapping) identifies and localizes kidney stones within CT images, enhancing understanding of the model's decision-making and supporting interpretability. Addressing challenges related to dataset imbalance, model interpretability, and generalizability across healthcare settings. By refining these models for kidney stone detection, the main objective of this work is to build a reliable diagnostic tool that aids in clinical judgement and treatment planning, ultimately enhancing patient outcomes and encouraging broader clinical adoption. © 2025 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Automail AI: Intelligent Voice-Controlled Email Generator 3

Automail AI: Intelligent Voice-Controlled Email Generator

引用

3rd International Conference on Intelligent and Innovative Technologies in Computing, Electrical and Electronics, IITCEE 2025

作者： Jayaram, Kavitha Mandagere, Samarth Gopisetti, Sampriti Sahit, S. Bnm Institute of Technology Dept. of Computer Science and Engineering Bengaluru India

ISBN: (纸本)9798331515911

The complexity of contemporary communication further emphasizes the need to automate monotonous work to increase efficiency and effectiveness. This paper introduces a new advance, voice-controlled Automail AI, in the form of an email generation tool that automatically sends an email according to end-user-specified keywords. Automail AI utilizes NLP, ML, and Speech Recognition to differentiate formal emails from informal ones and draft appropriate responses as the situation demands. It has been tuned with Naive Bayes Classifier that yields a better level of accuracy as compared to models such as Decision Tree, SVM, and Logistic Regression. Setting the benchmark for Automated Communication tools, Automail AI combines diverse technologies like SMTP to guarantee secure Email delivery, Google Text-to-Speech or GTTS for auditory feedback, and a module providing an interface for speech recognition that offers the possibility to use the tool with no hands. © 2025 IEEE.

关键词： Automail AI email generation GTTS machine learning Naive Bayes natural language processing SMTP speech recognition

来源：评论

学校读者我要写书评

暂无评论

Advancing Pediatric Appendicitis Diagnosis: Harnessing CNN and Hybrid Models with Ultrasound Imaging

Advancing Pediatric Appendicitis Diagnosis: Harnessing CNN a...

引用

2025 International Conference on Pervasive Computational Technologies, ICPCT 2025

作者： Kulkarni, Supritha Sharma, Sanchit Manavi, J. Vachana, U. Bhargavi, M.S. Bangalore Institute of Technology Dept. of Computer Science and Engineering Bengaluru India

ISBN: (纸本)9798331508685

Pediatric appendicitis, an acute inflammatory disease, arises from the obstruction of the appendix, often due to inflammation or a fecalith. This common abdominal emergency in children presents diagnostic challenges due to its varied clinical manifestations, which can lead to misdiagnosis or higher perforation rates stemming from atypical symptoms. This research investigates the use of deep learning, more especially Convolutional Neural Networks (CNNs), by utilizing ultrasonic imaging data, which is less expensive, non-invasive, and quicker than the other approaches now in use. Several variants are evaluated including a basic CNN and more advanced hybrid models such as CNN-Support Vector Regression, CNN-Transformer, and Hierarchical Deep CNN, to identify the most optimal model to significantly enhance diagnostic accuracy, for young patients facing this condition. The study's main goal is to develop a highly accurate and dependable diagnostic tool for evaluating the severity of pediatric appendicitis by differentiating between simple and complex cases and identifying whether surgery is required to inform treatment plans. Various challenges that need to be tackled are managing significant dataset imbalances, enhancing interpretability and transparency for real-world clinical settings, and ensuring the robustness and adaptability of all these models across various healthcare environments. © 2025 IEEE.

关键词： Pediatrics

来源：评论

学校读者我要写书评

暂无评论

IOT Based Multicrop Soil Moisture Control and Monitoring System 1

IOT Based Multicrop Soil Moisture Control and Monitoring Sys...

引用

1st International Conference on AIML-Applications for Engineering and technology, ICAET 2025

作者： Barhate, Madhusudan Desai, Devam Mishra, Gaurav Phadke, Gargi Das, Siuli Bhuran, Supriya Yogesh Ramrao Adik Institute of Technology D.Y. Patil Deemed to be University Dept. of Instrumentation Engg Dept. of Computer Engg Dept. of Computer Science and Engg Navi Mumbai India

ISBN: (纸本)9798350355611

Agriculture is crucial for the economy, with China and India leading in production. Conventional irrigation methods require precise timing and quantity, varying for different crops. This paper proposes an IoT-enabled intelligent irrigation system to enhance crop productivity. The system uses a moisture sensor to activate the pump motor, and IoT technology enables remote monitoring through a web interface or mobile app. The system stores moisture data customized to specific crop requirements, reducing human intervention while ensuring optimal water supply. © 2025 IEEE.

关键词： Moisture control

来源：评论

学校读者我要写书评

暂无评论

SINR-Delay Constrained Node Localization in RIS-Assisted Time-Varying IoT Networks Using ML Frameworks

引用

IEEE Transactions on Network and Service Management 2025年第2期22卷 1544-1557页

作者： Bhardwaj, Vikash Kumar Mundada, Gagan Sahoo, Omm Prakash Shukla, Mahendra K. Pandey, Om Jee Varanasi Dept. of Electronics Engineering 221005 India University of California San Diego Computer Science and Engineering Dept United States Varanasi Dept. of Electrical Engineering 221005 India ABV-IIITM Gwalior Dept. of Information Technology India

Node localization in time-varying Internet of Things (IoT) networks is an essential problem due to increased delay and poor Signal-to-Interference plus Noise Ratio (SINR) at the Base Station (BS). To improve the received signal strength at the BS, Reconfigurable Intelligent Surface (RIS) has recently been used between transmitter and receiver. Additionally, novel phase prediction methods and optimal weight assignment frameworks have been proposed over RIS and BSs, respectively. Nevertheless, these methods suffer from poor performance due to their heuristic approach, resulting in more time consumption and poor SINR. Motivated by the aforementioned challenges, we propose a novel node localization method over a RIS-assisted time-varying IoT network using Machine Learning (ML) frameworks in this work. Firstly, the method computes the optimal phase configuration over the RIS corresponding to each element using coeff2phaseNN, which has been trained on channel coefficients among the transmitter, receiver, and RIS. Subsequently, the weight of the individual antenna element at the BS is optimized using the proposed VectorSync model. The results confirm that the coeff2phaseNN method demonstrates a reduction of 89.79% in total MSE loss compared to the Artificial Neural Network-RIS (ANN-RIS) method. Additionally, it demonstrates a 71.04% reduction in the absolute RIS phase prediction deviation from the optimal phase compared to the ANN-RIS method. Moreover, the proposed VectorSync method attains a 79.28% and 92.29% reduction in time required for optimal weight assignment compared to the Bartlett and Capon methods, respectively. Finally, the Localization Error(LR) using the proposed method is compared to conventional methods in a time-varying experimental scenario and found to be the minimum, i.e., 6.156%. © 2004-2012 IEEE.

关键词： Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

Enhanced Realtime Sign Language Detection and Multilingual Translation System

Enhanced Realtime Sign Language Detection and Multilingual T...

引用

2025 International Conference on Computing for Sustainability and Intelligent Future, COMP-SIF 2025

作者： Kakkar, Madhur Daniel, John Sudha, D. Sathyabama Institute of Science and Technology Dept. of Computer Science and Engineering Chennai India

ISBN: (纸本)9798331538538

Sign language is the communication medium that is utilized by the hearing-impaired community. The medium is visual and it uses hand signs and symbols for communication. Over 300 different types of sign languages present in the world today. Sign language uses gestures and hand movements for the majority of the words and finger spellings for the rest. Nonverbal speakers face a lot of challenges while interacting with verbal speakers. Even non-verbal speakers who speak different sign languages face difficulties communicating with each other. This project proposes an enhanced real-time system which detects gestures and finger spellings of multiple sign languages, and translates them into multiple spoken languages. The proposed system aims to utilize CNN network for finger spelling recognition and Long Short-Term Memory (LSTM) network for gesture interpretation. The system will use computer vision for accurate gesture detection and Google Translator API for recognizing signs which will be translated into multiple spoken languages. Initial tests with isolated datasets indicate promising accuracy for both finger spelling and gesture recognition. The goal of the completed system is to help aid communication between verbal and non-verbal speakers and to provide an efficient mode of communication between the two communities. © 2025 IEEE.

关键词： Interactive computer systems

来源：评论

学校读者我要写书评

暂无评论

LeXNet++: Layer-wise eXplainable ResUNet++ framework for segmentation of colorectal polyp cancer images

引用

Neural Computing and Applications 2025年第1期37卷 213-229页

作者： Das, Surajit Khan, Soumya Suvra Sengupta, Diganta Debashis, De Dept. of Information Technology Meghnad Saha Institute of Technology Behind Urbana Complex Near Ruby General Hospital Anandapur Rd Uchhepota Kolkata700150 India Dept. of Computer Science & Engineering Meghnad Saha Institute of Technology Behind Urbana Complex Near Ruby General Hospital Anandapur Rd Uchhepota Kolkata700150 India Dept. of Computer Science & Engineering Heritage Institute of Technology Chowbaga Road West Bengal Anandapur Kolkata700107 India Dept. of Computer Science & Engineering Maulana Abul Kalam Azad University of Technology West Bengal NH-12 Simhat Haringhata West Bengal Nadia741249 India

Colorectal polyps are benign lesions that develop in the colon and can progress to cancer if left untreated. Clinical observations from medical images are often preferred over computational results due to the lack of trust in the machine learning models, thereby posing serious challenge for the explainability of the results. In order to computationally diagnose colorectal polyps from cancerous images and explain the results, we propose a Layer-wise eXplainable ResUNet++ (LeXNet++) framework for segmentation of the cancerous images, followed by layer-wise explanation of the results. We utilize a publicly accessible dataset that contains of 612 raw images with a resolution of 256×256×3 and an additional 612 clinically annotated and labeled images with a resolution of 256×256×1, which includes the infected region. The LeXNet++ framework comprises of three components—encoder, decoder and the bridge. The encoder and the decoder components each comprise of four layers. Each of the four layers in the encoder and the decoder comprises of 14 and 11 internal sub-layers, respectively. Among the sub-layers of the encoder and the decoder, there are three 3×3 convolutional layers with an additional 3×3 convolution-transpose layer in the decoder. The output of each of the sub-layers has been explained through heatmap generation after each iteration which have been further explained. The encoder and the decoder are connected by the bridge which comprises of three sub-layers. The results obtained from these three sub-layers have also been explained to inculcate trust in the findings. In this study, we have used three models to segment the images, namely UNet, ResUNet, and proposed LeXNet++. LeXNet++ exhibited the best result among the three models in terms of performance;hence, only LeXNet++ was explained layer-wise. Apart from explanation of the results fetched in this study, the performance of the proposed explainable model has been observed to be 2% greater than the existing poly

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Real-Time Object Detection from Surveillance using Deep Learning 3

Real-Time Object Detection from Surveillance using Deep Lear...

引用

3rd International Conference on Intelligent and Innovative Technologies in Computing, Electrical and Electronics, IITCEE 2025

作者： Tanisha, A. Tanisha, N. Chaitra, M. Tejasri, P.R. Bnm Institute of Technology Vtu Dept. of Computer Science and Engineering Bangalore India

ISBN: (数字)9798331515911

ISBN: (纸本)9798331515911

Object detection in surveillance systems leverages advanced deep learning techniques to enhance security measures through real-time analysis of dynamic video feeds. This project integrates the YOLOv5 model for detecting weapons in both pre-recorded videos and live camera feeds. The model, trained on a dataset with 4000 images labeled for handguns and knives, utilizes image preprocessing steps such as resizing, normalization, and augmentation to improve detection accuracy. Implementing the system with OpenCV and Tkinter, the application processes video streams, identifying potential threats with high precision rate of 0.85 and recall rate of 0.90. The integration of a graphical user interface ensures user-friendly operation for security personnel. Key outcomes demonstrate the efficacy of deep learning models in real-time object detection, significantly contributing to proactive surveillance and enhanced situational awareness. © 2025 IEEE.

关键词： Graphical user interfaces

来源：评论

学校读者我要写书评

暂无评论

TurfIt: An Instant Turf Booking Application

TurfIt: An Instant Turf Booking Application

引用

2025 IEEE International Conference on Computational, Communication and Information technology, ICCCIT 2025

作者： Rajeswari, A.M. Tamilselvan, V. Abishek, M.R.P. Dept. of Computer Science and Engineering Velammal College of Engineering and Technology Madurai India

ISBN: (纸本)9798331512965

As sports grow in popularity need for effective online turf booking applications has also increased. Although some existing applications are faced with numerous challenges such as;restrictions on access, ineffectiveness in scheduling, and over-bookings, leading to frustrations and operational problems. This paper focuses on those problems by the presentation of an online turf booking application to streamline the booking process. The solution offers real-time updates, automatic payment, and safe transactions based on QR code. The more user-friendly the application is, the more efficient it will be. Firebase would be used for real-time data synchronization and the responsiveness of React interface ensures the smooth and conflict-free bookings. This solution would lead to increasing customer satisfaction as well as increasing the booking rates and effective utilization of sports facilities. © 2025 IEEE.

关键词： Customer satisfaction

来源：评论

学校读者我要写书评

暂无评论

Fake Job Post Detection Using Machine Learning

Fake Job Post Detection Using Machine Learning

引用

2025 International Conference on Computing for Sustainability and Intelligent Future, COMP-SIF 2025

作者： Pradeep Kumar, B.P. Bhagya, D.J. Gagana, G. Mani, N. Dept. of Computer Science and Engineering Atria Institute of Technology Karnataka Bangalore India

ISBN: (数字)9798331538538

ISBN: (纸本)9798331538538

By examining patterns in the text data and metadata linked to job adverts, The use of learning algorithms has grown in popularity in the detection of fraudulent job postings. For this, Research Machine emphasises the application of a number of algorithms, such as Naive Bayes and Random Forest, Together with models for deep learning, like CNN and Bidirectional LSTM, and AdaBoost. To extract useful features These techniques depend on text preparation techniques such as lemmatisation, tokenisation, and TF-IDF vectorisation. To differentiate between authentic and fraudulent posts, models are instructed in datasets and evaluated using measures such as accuracy, accuracy, memory, and F1-score. Advanced classifiers and group techniques have proven to be quite successful in enhancing detection performance while lowering false positives and negatives. © 2025 IEEE.

关键词： Random forests

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：