检索结果-内蒙古大学图书馆

5th IEEE Global Conference for Advancement in technology, GCAT 2024

作者： Nevatia, Isha Chaudhary, Vraj Pandian, M. Thurai Vellore Institute of Technology School of Computer Science and Engineering Vellore India

ISBN: (纸本)9798350376685

The E-Health care system is crucial in modern health care. These systems are designed to secure medical data repositories. The main goal of this study is to ensure both security and privacy by enhancing fully homomorphic data collection. It also aims to improve classification accuracy in disease detection. The existing method of Privacy-Preserving Data Mining (PPDM) is utilized to maintain privacy in healthcare, and the new method using a Convolutional Neural Network (CNN) algorithm is employed to identify diseases more accurately. This new approach enhances performance by achieving higher accuracy, recall, and reducing time complexity. © 2024 IEEE.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

A Smart Agricultural Assistant for Crop Recommendation using Machine Learning 15

A Smart Agricultural Assistant for Crop Recommendation using...

引用

15th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Kavitha, D. Kiranteja, P. Shridharan, V.J. Venkhat Desigan, P.K. Raghav, N.G. Pranav, V. School of Computer Science and Engineering Vellore Institute of Technology Chennai India

ISBN: (纸本)9798331300579

This paper presents a Smart Agriculture Assistant that uses data analytics and machine learning to assist farmers in making informed decisions about various farming practices. The assistant offers a holistic approach to enhance agricultural decision-making process by recommending crops based on soil attributes, climate, and local weather conditions, as well as providing precise fertilizer and pesticide guidance tailored to Indian contexts. Data is collected from various sources, such as weather forecasts, groundwater data, soil maps, and crop prices. Machine learning algorithms are trained on that data to generate optimal recommendations for farmers. The study demonstrates the effectiveness of smart agricultural techniques over conventional methods in improving crop selection accuracy and optimizing fertilizer application, thereby enhancing agricultural productivity. Furthermore, the integration of image analysis technology using computer vision enables the system to assess crop quality parameters, aiding in the detection of grade, size, colour, and disease presence. It also provides insights of estimated real-time crop prices by taking into account the fickle market dynamics. This offers valuable information to the farmers for increasing profit margins. The assistant is implemented as a web application that can be accessed by farmers from anywhere. This research provides a framework for the integration of data-driven technology to solve the unique issues encountered by Indian farmers. The Smart Agriculture Assistant presented here represents the power of data science and deep learning for a more prosperous and sustainable future for Indian agriculture. © Grenze Scientific Society, 2024.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Classification, Segmentation and Quantification of acute lymphoblastic leukemia cells in the bone marrow

Classification, Segmentation and Quantification of acute lym...

引用

2024 International Conference on Computing and Intelligent Reality Technologies, ICCIRT 2024

作者： Jeffine, Theou Kiruthika, S. Joshan Athanesious, J. School of Computer Science and Engineering Vellore Institute of technology Chennai India

ISBN: (纸本)9798331510299

Cancer is a type of disease in which abnormal cells flourish and penetrate adjacent tissues, then spread to other parts of the body via the blood and lymph system. This abnormality results in DNA alterations in the bone marrow, which give rise to malignant cells in haematologic malignancies like leukaemia. This results in the abnormal production of white blood cells, where cancerous cells proliferate more quickly than healthy cells, compromising the ability of healthy cells to carry out their regular tasks, including fighting infections, distributing oxygen, and clotting blood. The primary objective was to create an efficient model that could classify the given microscopic image of bone marrow containing cancerous cells or not. If the image is classified as cancerous then the model will segment the cancer cells over the normal cells, in addition the model will quantify the presence of malignant cells to measure the spread in the blood. Segmentation and quantification along with the classification will provide better explainability to the model, thus to achieve this both the image processing techniques and the deep learning model are used to develop the Classification, Segmentation and Quantification (CSQ) of acute lymphoblastic leukemia cells in the bone marrow. The model is evaluated using the metrics like accuracy, precision, recall, specificity and F1score. Experimental results showed that the model performs well when compared to the existing methods. © 2024 IEEE.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

DeepFrame: A Scientific Framework for Advancing Ocean Cleanup through Underwater Debris Detection and Geospatial Mapping

DeepFrame: A Scientific Framework for Advancing Ocean Cleanu...

引用

2024 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems, ICSES 2024

作者： Jesher Joshua, M. Ragav, V. Pandiyaraju, V. School of Computer Science and Engineering Vellore Institute of Technology Chennai India

ISBN: (数字)9798331543624

ISBN: (纸本)9798331543617

The escalating concern of underwater debris demands innovative approaches for effective ocean cleanup. This study presents a comprehensive framework utilizing deep learning methods to detect, categorize, track, and geospatially map un derwater debris. By amalgamating cutting-edge object detection models, the system extracts real-time insights from underwater sensor data, facilitating the identification and categorization of debris based on various attributes. The incorporation of advanced tracking algorithms enables the monitoring of debris dynamics, contributing to predictive capabilities. Furthermore, geospatial mapping techniques are employed to project identified debris onto world maps, using latitude and longitude coordinates. This mapping feature aids in locating debris accumulation areas, thereby enabling strategic and efficient cleanup efforts. Through meticulous experimentation on authentic underwater debris datasets, our framework demonstrates its potential in precisely detecting, tracking, and mapping debris. The proposed solution leverages robust extended dataset of augmented images with its benchmark and novel Geo-Spatial mapping framework, which revolutionizes ocean conservation endeavors by providing valuable insights and ecologically-aware of underwater debris removal. The source code and the dataset have been made publicly available on GitHubl. © 2024 IEEE.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

Dynamic strategy to use optimum memory space in real-time video surveillance

引用

Journal of Ambient Intelligence and Humanized Computing 2023年第3期14卷 2771-2784页

作者： Biswas, Tamal Bhattacharya, Diptendu Mandal, Gouranga Department of Computer Science and Engineering National Institute of Technology Agartala Tripura Agartala India School of Computer Science and Engineering VIT-AP University Andhra Pradesh Amaravati India

In present days real-time video surveillance is a very essential aspect of establishing safety, observing traffic, and detecting violence and crimes. Intelligent video surveillance (IVS) is one of the most acknowledged frameworks in a security application. But, video surveillance systems consume huge memory space to store the recorded video, especially in the case of high-resolution recording. However, if we use low-resolution video recording then the video quality will be reduced and the objects/violence cannot be identified clearly. To overcome the stated limitation, this article introduced a smart video surveillance system with a dynamic strategy to use optimum memory space with the best surveillance objective. The proposed system can dynamically record video with high-resolution during suspicious movement of objects/violence as well as low-resolution video at other times. The system is based on the detection of suspicious movement of objects in consecutive frames, saving of suspicious frames with high resolution, and neglecting of less important frames. Finally, the Contrast Limited Adaptive Histogram Equalization (CLAHE) based on Color Channel is applied to all the suspicious frames as a post-processing step for better contract adjustment and to provide a clear view of suspicious objects in frames. The quality of enhanced frames are assessed by two no-reference methods, i.e. NIQMC and BIQME. Multiple visual quality parameters like sharpness, contrast, colorfulness, brightness, and naturalness of frames, etc. are used for assessment Several real-time experiments reveal that the proposed system can detect suspicious objects with 97.65% accuracy. Besides, it consumes 75% less memory space in real-time surveillance. © 2023, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Video recording

来源：评论

学校读者我要写书评

暂无评论

Handling and Research on Imbalanced Data Using VAE-SAGAN 24

Handling and Research on Imbalanced Data Using VAE-SAGAN

引用

9th International Conference on Cyber Security and Information engineering, ICCSIE 2024

作者： Ye, Chaoqian Wu, Shaofei Liu, Jun Hu, Zhuoya School of Computer Science and Engineering Wuhan Institute of Technology Wuhan China

ISBN: (纸本)9798400718137

Fraud detection datasets typically contain a large number of normal transaction samples and very few fraudulent samples, resulting in a severely imbalanced dataset. This imbalance affects the model’s classification ability, making it prone to misjudgments, which can lead to significant financial losses. Therefore, addressing data imbalances is crucial for improving the accuracy and reliability of fraud detection models. In this study, we propose a fusion model based on Variational Autoencoder and Self-Attention Generative Adversarial Network (VAE-SAGAN) to tackle the data imbalance problem. First, we preprocess the majority of samples using the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) method to filter out overlapping and noisy samples and remove them. Then, we input the minority samples into the VAE-SAGAN model for training, enabling the model to generate more realistic minority samples. In this study, we compare the proposed sampling method with existing sampling algorithms across different datasets and classifiers. Experimental results show that the proposed method outperforms existing sampling algorithms in terms of F1, G-mean, AUC, and Accuracy, and exhibits relatively stable generalization ability across multiple datasets. This research not only enhances the performance of fraud detection models but also provides more reliable technical support for risk management in practical applications. © 2024 Copyright held by the owner/author(s).

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing Healthcare through the Symbiosis of Blockchain and Machine Learning in Preprocessing Skin Disease Images: A Review 15

Enhancing Healthcare through the Symbiosis of Blockchain and...

引用

15th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Rokade, Sonali Mishra, Nilamadhab School of Computer Science and Engineering Vellore Institute of Technology Bhopal India

ISBN: (纸本)9798331300579

Recent advancements in skin disease identification leverage machine learning for automated diagnosis. However, safeguarding the integrity of sensitive medical data remains a top priority. This paper explores the integration of machine learning and blockchain in skin disease identification preprocessing, promising trust, transparency, and accuracy in dermatological diagnosis. The preprocessing workflow starts with gathering a diverse, labeled skin image dataset, followed by data cleaning and augmentation to enhance quality and quantity. Standardization methods like resizing and normalization ensure data consistency, while addressing class imbalances prevents model bias. Blockchain technology is introduced to secure data integrity. Each skin image and its metadata are encrypted and timestamped on a decentralized blockchain, guaranteeing tamper-proof authenticity. Machine learning models, typically based on CNNs, extract features and classify diseases, benefiting from high-quality preprocessed data. During training, blockchain securely records model parameters and progress, ensuring transparency and auditability, crucial for medical accountability and *** fusion of machine learning and blockchain enhances diagnostic accuracy while addressing data privacy and security concerns, potentially revolutionizing dermatological healthcare with a trustworthy, automated diagnosis system. © Grenze Scientific Society, 2024.

关键词： Data integrity

来源：评论

学校读者我要写书评

暂无评论

An Integrated Framework for Real-Time Disaster Data Acquisition, Processing, and Resource Optimization in Crisis Management Systems 4

An Integrated Framework for Real-Time Disaster Data Acquisit...

引用

4th International Conference on Ubiquitous Computing and Intelligent Information Systems, ICUIS 2024

作者： Saranya, G. Atchaya, M. Harshini, R. Choudhary, Kartik School of Computer Science Engineering Vellore Institute of Technology Chennai India

ISBN: (纸本)9798331529635

The rapid onset and unpredictable nature of disasters necessitate a quick, organized, and efficient response. This paper proposes a comprehensive disaster management platform that integrates real-time disaster information, weather forecasting, and resource management to enhance disaster preparedness and response. The platform collects disaster data from geophysical monitoring stations, processes the data for consistency, and stores it in a MongoDB database. Geospatial visualization tools, such as Folium, are employed to present disaster events on interactive maps, helping decision-makers assess the situation and respond accordingly. Additionally, the platform includes features for weather forecasting and resource management, allowing users to monitor conditions and contribute essential resources in real time. This research demonstrates how leveraging modern technologies can improve disaster management systems, providing a more effective and coordinated response to natural and man-made disasters. © 2024 IEEE.

关键词： Weather forecasting

来源：评论

学校读者我要写书评

暂无评论

Extended Multi-scale Feature Fusion and Balanced Generative Adversarial Network for Image Inpainting under Limited Data

Extended Multi-scale Feature Fusion and Balanced Generative ...

引用

2024 Workshop on Electronics Communication engineering, WECE 2024

作者： Xiao, Ting Sun, Minqian Chen, Haoyuan Li, Jiaqi Qiu, Yixing Wang, Zhe Lu, Dan Department of Computer Science and Engineering East China University of Science and Technology China School of Artificial Intelligence OPtics and ElectroNics Northwestern Polytechnical University China School of Computer Science and Technology Harbin Engineering University China

ISBN: (数字)9781510689145

ISBN: (纸本)9781510689138

Image inpainting has made remarkable progress in recent years due to the availability of sufficient training data. However, when the training data is insufficient, i.e., image inpainting in few-shot regimes, the performance of existing image inpainting methods degrades drastically. Two challenges impede image inpainting under few-shot regimes, adequate information and unstable training. To address the above issues, this paper proposes a new method for image inpainting under limited data. Specifically, an Extended Multi-scale Feature Fusion (EMFF) module is proposed to capture representative contextual information by progressively reducing the number of channels for features with increasing receptive fields, thereby utilizing effective information. Additionally, we leverage a multi-scale discriminator whose cross-layer features are mixed for supplementation to formulate a Balanced Generative Adversarial Network (BGAN) to stabilize the training dynamics. Comprehensive experiments demonstrate the effectiveness of the proposed method. © 2025 SPIE.

关键词： Image fusion Image quality Feature fusion Education and training Image restoration Convolution Data modeling Image enhancement Gallium nitride Semantics

来源：评论

学校读者我要写书评

暂无评论

Global Color-Aware Arbitrary Style Transfer with Discrete Wavelet Transform 20th

Global Color-Aware Arbitrary Style Transfer with Discrete W...

引用

20th IFIP WG 10.3 International Conference on Network and Parallel Computing, NPC 2024

作者： Chen, Bochao Tan, Junhao Li, Xiaoqiang Pei, Songwen School of Computer Engineering and Science Shanghai University Shanghai200444 China School of Optical-Electrical and Computer Engineering University of Shanghai for Science and Technology Shanghai200093 China

ISBN: (纸本)9789819628292

The purpose of image arbitrary style transfer is to apply a given artistic or photorealistic style to a target content image. While existing methods can effectively transfer style information, the variability in color brightness features between the style and content images can significantly impact the quality of the style transfer. Specifically, traditional arbitrary style transfer relies on an encoder-decoder architecture of spatial-aware neural networks operating in the RGB domain, which typically preserves only local color brightness information and lacks a global color perception compared to the wavelet frequency domain. In this paper, we address the issue of insufficient global color perception in images from the perspective of image enhancement. We propose Global Color-aware Arbitrary Style Transfer with Discrete Wavelet Transform (GCAST), which achieves a balance between the global color characteristics of style images and the local details of content images. Our approach involves an encoder that integrates preliminary feature encoding through cross-layer attention fusion, while the decoder modifies frequency features using a wavelet-based adjustment block. To maintain a balance between content and style in the stylized images, we design a comprehensive loss function that incorporates multiple restoration loss, double style loss, and noise regularization loss. Extensive experiments comparing our method with state-of-the-art techniques demonstrate that our proposed model generates stylized images with superior results in terms of global color perception and human perceptual studies. © IFIP International Federation for Information Processing 2025.

关键词： Decoding

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：