检索结果-内蒙古大学图书馆

Implementation of computer Vision of Jakarta Weather Image Categorization Using ResNet

Procedia computer science 2023年 227卷 813-822页

作者： Kartika Purwandari Join W.C. Sigalingging Alam Ahmad Hidayat Tjeng Wawan Cenggoro Bens Pardamean Computer Science Department School of Computer Science Bina Nusantara University Jakarta Indonesia 11480 Computer Science Department BINUS Graduate Program – Master of Computer Science Bina Nusantara University Jakarta Indonesia 11480

At the moment, weather data is crucial for supporting neighborhood activities. The economy and trade are both centered in Jakarta, which is also Indonesia's capital. Therefore, it is crucial to have access to weather information so that these activities don't get disrupted, which would then hinder commercial and trade activity. Social media has been a very popular tool for spreading information recently. Particularly on Instagram, where users favor taking images and sharing the information they encounter. @jktinfo is the Instagram account that posts information about the situation in Jakarta and the area, including the current weather. The @jktinfo account is utilized in this project to gather data. Utilizing a variety of techniques, the collected photographs of sunny, cloudy, and wet situations were.

关键词： computer vision weather image categorization data augmentation deep transfer learning residual network

来源：评论

学校读者我要写书评

暂无评论

EmoSync: Facial Emotion Detection for Adaptive Music Playback 3

EmoSync: Facial Emotion Detection for Adaptive Music Playbac...

引用

3rd International Conference on Artificial Intelligence For Internet of Things, AIIoT 2024

作者： Devarajan, Devikanniga Rajkumar, K. GITAM Deemed to be University Computer Science and Engineering Nagadenahalli Doddabalpur Karnataka Bangalore561203 India SRM Institute of Science and Technology Department of Data Science and Business Systems School of Computing Tamil Nadu Kattankulathur603 203 India

ISBN: (纸本)9798350372120

This work investigates using Convolutional Neural Networks (CNN) and VGGNet architecture, for realtime emotion detection. Raw data from the FER-2013 dataset is transformed from CSV format into organized image directories, segmented by distinct emotions. The resulting model, equipped with convolutional, pooling, and dense layers, effectively classifies emotions from 48x48 grayscale images. In application, the model overlays detected emotions onto a webcam stream. An integrated music player module synchronizes music with the subject's emotions. Overall, this research explores a comprehensive interactive system focused on recognizing and responding to human emotions. Without further training data, our model, to the best of our understanding, achieves modern single-network accuracy of 74.57% on FER2013 © 2024 IEEE.

关键词： Python

来源：评论

学校读者我要写书评

暂无评论

Study of Cross-Site Request Forgery on Web-Based Application: Exploitations and Preventions

引用

Procedia computer science 2023年 227卷 92-100页

作者： Crisdeo Nuel Siahaan Mario Rufisanto Raymond Nolasco Said Achmad Chrisando Ryan Pardomuan Siahaan Cyber Security Program Computer Science Department School of Computer Science Bina Nusantara University Jakarta 11480 Indonesia Computer Science Department School of Computer Science Bina Nusantara University Jakarta 11480 Indonesia

Cross-Site Request Forgery (CSRF) is a prominent web exploit that continues to pose significant security risks, even on highly ranked websites. This research focuses on identifying the underlying vulnerability, understanding the techniques employed, and proposing effective preventive measures. The hybrid method which is systematic report review method and lab-based scenarios method is conducted to identify CSRF exploitation and prevention mechanisms. Using a scenario-based testing approach, this study investigated the cause and effect of CSRF attacks, uncovering numerous techniques for exploiting this vulnerability. Based on the report review, 35 CSRF vulnerability reports from the Hacker one Bug Bounty Platform are analyzed. The result shows there is a lack of awareness and implementation of primary defenses, resulting in a high violation of end-user data integrity. The result also extensively describes some of the profound causes and effects of CSRF attacks and uncovers a broad range of techniques for exploiting this vulnerability. Furthermore, this study provides actionable recommendations to enhance system security and raise awareness among developers and users. The findings from this research serve as a valuable resource for improving security practices and mitigating CSRF attacks.

关键词： cross-site request forgery web exploitation vulnerability

来源：评论

学校读者我要写书评

暂无评论

Improved random features for dot product kernels

The Journal of Machine Learning Research

引用

The Journal of Machine Learning Research 2024年第1期25卷 11403-11477页

作者： Jonas Wacker Motonobu Kanagawa Maurizio Filippone Data Science Department EURECOM France Statistics Program KAUST Saudi Arabia

Dot product kernels, such as polynomial and exponential (softmax) kernels, are among the most widely used kernels in machine learning, as they enable modeling the interactions between input features, which is crucial in applications like computer vision, natural language processing, and recommender systems. We make several novel contributions for improving the efficiency of random feature approximations for dot product kernels, to make these kernels more useful in large scale learning. First, we present a generalization of existing random feature approximations for polynomial kernels, such as Rademacher and Gaussian sketches and TensorSRHT, using complex-valued random features. We show empirically that the use of complex features can significantly reduce the variances of these approximations. Second, we provide a theoretical analysis for understanding the factors affecting the efficiency of various random feature approximations, by deriving closed-form expressions for their variances. These variance formulas elucidate conditions under which certain approximations (e.g., TensorSRHT) achieve lower variances than others (e.g., Rademacher sketches), and conditions under which the use of complex features leads to lower variances than real features. Third, by using these variance formulas, which can be evaluated in practice, we develop a data-driven optimization approach to improve random feature approximations for general dot product kernels, which is also applicable to the Gaussian kernel. We describe the improvements brought by these contributions with extensive experiments on a variety of tasks and datasets.

关键词： random features randomized sketches dot product kernels polynomial kernels large scale learning

来源：评论

学校读者我要写书评

暂无评论

Accelerated Malaria Diagnosis with Deep Convolutional Neural Network 1

Accelerated Malaria Diagnosis with Deep Convolutional Neural...

引用

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

作者： Krishna Kumar, V. Geetha Devasana, M.S. Saravana Balaji, B. Sri Ramakrishna Engineering College Department of Computer Science and Engineering Coimbatore India De Montfort University Kazakhstan Department of Business Data Analytics Almaty Kazakhstan

ISBN: (纸本)9798350359756

The disease known as malaria is a communicable illness that is spread by the bites of mosquitoes. There are already diagnostic approaches that include manually counting the amount of red blood cells (RBC) that are contaminated. This is accomplished by doing a microscopic examination of the stained blood cells of the individual who is suffering from the disease. However, this is a hard job that requires careful attention both visually and intellectually. In order to accomplish this, a comprehensive manual examination must be undertaken by an expert who is informed about the subject matter. It is possible that the implementation of deep learning algorithms will simplify the process of disease diagnosis and make this analysis more straightforward. Furthermore, because they are diagnosing malaria, they are expected to manage a vast quantity of data, which includes photographs of microscopic blood smears. This is because they are responsible for finding the disease. In order to get a better level of accuracy, it is necessary to train models that are more in-depth. This requires fast computing since it requires a substantial amount of computer resources. An accelerated convolutional neural network technique that is based on a graphics processing unit (GPU) is utilized by the Accelerated Malaria Diagnosis System (AMDS) that has been created in order to diagnose malaria. © 2023 IEEE.

关键词： Diseases

来源：评论

学校读者我要写书评

暂无评论

Automated Detection of Poultry Farms from Aerial Images for Actionable AI System toward Biosecurity Applications

Automated Detection of Poultry Farms from Aerial Images for ...

引用

2023 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2023

作者： Li, Yu Das, Rana Duong, Cuong Lim, Teng-Teeh Haithcoat, Timothy Ersoy, Ilker Shyu, Chi-Ren University of Missouri Department of Electrical Engineering and Computer Science ColumbiaMO United States University of Missouri Biological Engineering ColumbiaMO United States University of Missouri Plant Science and Technology Extension ColumbiaMO United States University of Missouri Institute for Data Science and Informatics Columbia United States University of Missouri Institute for Data Science and Informatics Electrical Eng. and Computer Science Dept. ColumbiaMO United States

ISBN: (纸本)9798350359527

Poultry productions have shifted towards larger farms and often cluster in certain regions. However, many of the smaller farms with a considerable amount of production are not considered concentrated animal feeding operation under regulation in many regions, and therefore are not required to obtain a permit. As a result, information of these operations is not being recorded and monitored, contributing to an increased risk for biosecurity issues. The development in computer vision and availability of high-resolution remote sensing imagery allows the possibility for automatic localization and operation estimation of the unregistered farms, filling in the missing information needed for biosecurity risk management. The farm information extracted from images can be incorporated with other available geospatial data, such as road networks, watersheds, etc., to build a comprehensive knowledge base for actionable plans. In the current study, we present a framework for automated localization, analysis, and visualization of poultry farms from remote sensing images. A robust machine learning method was developed to computationally link the visual features from remote sensing imagery to Missouri department of Natural Resources Concentrated Animal Feeding Operation data, to identify the animal farm type, location, and potential capacity in the region. The developed process can be performed automatically on images of different timestamps to analyze temporal and spatial development of poultry production facilities of a region. This research contributes to the effective management and regulation of poultry and potentially other animal farms, providing insights into farm characteristics and capacities while facilitating informed decision-making. © 2023 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Fuel Monitoring in Bike Vehicles Based on IoT

Fuel Monitoring in Bike Vehicles Based on IoT

引用

2023 Intelligent Computing and Control for Engineering and Business Systems, ICCEBS 2023

作者： Yuvaraj, G. Nithya, M. Shiny, A. LakshmiNarayanan, S. Vinothini, A. Selva Vignesh, S.M. Sri Sairam Engineering College Department of Computer Science and Engineering Chennai India Veltech Multitech Dr.Rangarajan Dr.Sakunthala Engineering College Department of Artificial Intelligence & Data Science Avadi India

ISBN: (纸本)9798350394580

In this work, we explore the feasibility of automatically dispensing gasoline using a prepaid RFID card. Individual RFID cards loaded with predetermined quantities will be provided to each user. The gasoline station's dispenser has an RFID reader that can scan the card and update the available balance on an LCD screen. The user inputs the required fuel quantity via keypad, and the system then determines how long the electrical fuel pump needs to run for and begins distributing the gasoline. When the system detects that the target level has been reached, it will immediately turn off the pump. It utilizes a Raspberry Pi module to alert the user through their smartphone. In addition, there is a fire sensor and a fuel purity sensor in this system to prevent fires and ensure clean fuel. As a result, the gasoline bunk management system will be decentralized and need fewer people to operate. © 2023 IEEE.

关键词： ADC 0804 buzzer fuel level sensor GSM/GPRS module Raspberry pi sim808

来源：评论

学校读者我要写书评

暂无评论

Hybrid GCN-LSTM Framework for Enhanced Schizophrenia Diagnosis: Integrating Spatial and Temporal Insights from Brain Connectivity data 1

Hybrid GCN-LSTM Framework for Enhanced Schizophrenia Diagnos...

引用

1st International Conference on Innovations in Communications, Electrical and computer Engineering, ICICEC 2024

作者： Anitha, R. Hemalatha Prathyusha Engineering College Department of Artificial Intelligence and Data Science Tamil Nadu Tiruvallur India R.M.D. Engineering College Department of Computer Science and Engineering Tamil Nadu Kavaraipettai601206 India

ISBN: (数字)9798350376517

ISBN: (纸本)9798350376517

A new hybrid approach for diagnosing schizophrenia combines graph convolutional networks (GCNs) with long short-term memory (LSTM) networks, leveraging GCNs for spatial connectivity and LSTMs for temporal modeling. This integration allows a simultaneous examination of spatial and sequential brain connectivity patterns, offering a comprehensive view of schizophrenia's complex dynamics. By extracting critical features from brain connectivity data, the hybrid GCN-LSTM model enhances research insights and improves clinical interpretability. Testing and validation with real-world data demonstrate the model's superior performance over existing methods, advancing clinical decision support in psychiatry. Additionally, this approach not only improves our understanding of schizophrenia's mechanisms but also shows potential for broader applications, such as early and accurate dementia diagnosis. The study marks significant progress in applying advanced machine learning to clinical research. © 2024 IEEE.

关键词： Clinical research

来源：评论

学校读者我要写书评

暂无评论

Faster Region Based Convolutional Neural Network for Skin Lesion Segmentation

引用

Intelligent Automation & Soft Computing 2023年第5期36卷 2099-2109页

作者： G.Murugesan J.Jeyapriya M.Hemalatha S.Rajeshkannan Department of Computer Science and Engineering St.Joseph’s College of EngineeringChennaiTamil NaduIndia Department of Computer Science and Engineering B.S.Abdur Rahman Crescent Institute of Science and TechnologyChennaiTamil NaduIndia Department of Artificial Intelligence and Data Science Saveetha Engineering College(Autonomous)ChennaiTamil NaduIndia Department of Electronics and Communication Engineering St.Joseph’s College of EngineeringChennaiTamil NaduIndia

The diagnostic interpretation of dermoscopic images is a complex task as it is very difﬁcult to identify the skin lesions from the *** the accurate detection of potential abnormalities is required for patient monitoring and effec-tive *** this work,a Two-Tier Segmentation(TTS)system is designed,which combines the unsupervised and supervised techniques for skin lesion *** comprises preprocessing by the medianﬁlter,TTS by Colour K-Means Clustering(CKMC)for initial segmentation and Faster Region based Con-volutional Neural Network(FR-CNN)for reﬁned *** CKMC approach is evaluated using the different number of clusters(k=3,5,7,and 9).An inception network with batch normalization is employed to segment mel-anoma regions *** loss functions such as Mean Absolute Error(MAE),Cross Entropy Loss(CEL),and Dice Loss(DL)are utilized for perfor-mance evaluation of the TTS *** anchor box technique is employed to detect the melanoma region *** TTS system is evaluated using 200 dermoscopic images from the PH2 *** segmentation accuracies are analyzed in terms of Pixel Accuracy(PA)and Jaccard Index(JI).Results show that the TTS system has 90.19%PA with 0.8048 JI for skin lesion segmentation using DL in FR-CNN with seven clusters in CKMC than CEL and MAE.

关键词： Skin cancer melanoma diagnosis clustering convolution neural network unsupervised segmentation deep learning

来源：评论

学校读者我要写书评

暂无评论

Adaptive gradient methods with dynamic bound of learning rate 7

Adaptive gradient methods with dynamic bound of learning rat...

引用

7th International Conference on Learning Representations, ICLR 2019

作者： Luo, Liangchen Xiong, Yuanhao Liu, Yan Sun, Xu MOE Key Lab of Computational Linguistics School of EECS Peking University China College of Information Science and Electronic Engineering Zhejiang University China Department of Computer Science University of Southern California United States Center for Data Science Beijing Institute of Big Data Research Peking University China

Adaptive optimization methods such as ADAGRAD, RMSPROP and ADAM have been proposed to achieve a rapid training process with an element-wise scaling term on learning rates. Though prevailing, they are observed to generalize poorly compared with SGD or even fail to converge due to unstable and extreme learning rates. Recent work has put forward some algorithms such as AMSGRAD to tackle this issue but they failed to achieve considerable improvement over existing methods. In our paper, we demonstrate that extreme learning rates can lead to poor performance. We provide new variants of ADAM and AMSGRAD, called ADABOUND and AMSBOUND respectively, which employ dynamic bounds on learning rates to achieve a gradual and smooth transition from adaptive methods to SGD and give a theoretical proof of convergence. We further conduct experiments on various popular tasks and models, which is often insufficient in previous work. Experimental results show that new variants can eliminate the generalization gap between adaptive methods and SGD and maintain higher learning speed early in training at the same time. Moreover, they can bring significant improvement over their prototypes, especially on complex deep networks. The implementation of the algorithm can be found at https://***/Luolc/AdaBound. © 7th International Conference on Learning Representations, ICLR 2019. All Rights Reserved.

关键词： Gradient methods

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：