检索结果-内蒙古大学图书馆

Unveiling deceit: A computer vision-based approach to detect lies from facial expressions

AIP Conference Proceedings 2024年第1期3134卷

作者： Elizabeth Isaac Pristy Paul Thoduparambil M. Raghu Rohitha Anuvindha Ansalam T. P. Akshay Assistant Professor Department of Computer Science and Engineering Mar Athanasius College of Engineering Kothamangalam Kerala India Department of Computer Science and Engineering Mar Athanasius College of Engineering Kothamangalam Kerala India

Polygraph tests monitor heart rate and pulse for truth determination but aren’t fully reliable. Face based deception detection systems are noninvasive and are more convenient. The proposed system uses computer Vision to monitor eye blink rate, gaze direction, lip compression, and pupil dilation to determine truth accuracy in real-time. The number of frames with a threshold value lower than the Eye-Aspect-Ratio (EAR) constant determines the eye blink count. The pupil coordinates are used to track the eye gaze direction. The eye centroid to iris contour maximum distance tracks pupil dilation. Lip compression is determined by measuring the distance between the mean of upper and lower lip landmarks. The weighted sum method is used to find the lie index from the obtained features. The true positive rate (TPR) and false positive rate (FPR) at various threshold values is calculated and an ROC (Receiver Operating Characteristic) curve is plotted using these values. The point on the curve with the highest TPR and lowest FPR determines the optimal threshold for the lie detection system. This method allows customization of the system’s performance to meet specific requirements.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Lip reading using deep learning and convolutional neural network

引用

AIP Conference Proceedings 2024年第1期3134卷

作者： Elizabeth Isaac Sachin K. Tomy Basil Joy Edwin Augustine Thomas Hussain Nizar Ahmed Jacob Raybin Assistant Professor Department of Computer Science & Engineering Mar Athanasius College of Engineering Kothamangalam Kerala India Department of Computer Science & Engineering Mar Athanasius College of Engineering Kothamangalam Kerala India

Lip reading involves deciphering and scrutinizing a speaker’s lip movements to understand spoken words or phrases. The diverse factors like speech pace, intensity, and similar character sequences pose significant challenges to lip reading. This study introduces a lip-reading model tailored for video data containing variable-length sequence frames without audio. Initially, we isolate the lip region from each facial image within the video sequence and amalgamate them into a unified image. Subsequently, a twelve-layer convolutional neural network with two layers of batch normalization is constructed to train the model and extract visual information. This approach aims to mitigate both internal and external disparities encompassing traits such as speaker accent, lighting conditions, picture frames, speaking tempo, and posture. By leveraging extensive data, the model demonstrates the capability to accurately anticipate complete words.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Keyword search application for videos using frames

引用

AIP Conference Proceedings 2024年第1期3134卷

作者： Anu Eldho Linda Sara Mathew B. Rohit Eldhobasil Parackal Balu Felix Assistant Professor Department of Computer Science & Engineering Mar Athanasius College of Engineering Kothamangalam Kerala India Department of Computer Science & Engineering Mar Athanasius College of Engineering Kothamangalam Kerala India

Keyword search in videos is a technique that allows for efficient search and retrieval of specific moments within a video. This is accomplished by using speech recognition technology to convert audio from the video into text. This text can be compared with the visual content of each frame in the video using optical character recognition (OCR) technology. By iterating through each frame and comparing the audio and visual text, matches can be identified and the relevant parts of the video can be quickly accessed using the corresponding timestamps. We use a video analysis tool or library that has the ability to extract timestamps from a video file. These tools can be used to create a list of timestamps for all the frames in a video, which can then be used to quickly access specific moments in the v ideo. This technique can be useful in a variety of applications, such as media search, video editing, and video surveillance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multimodal search in fashion store using CLIP

引用

AIP Conference Proceedings 2024年第1期3134卷

作者： Isaac Elizabeth Anu Mathew Joby Roy Meckamalil Rotney Sagar V. K. Sruthi N. Kartha Sreekanth P. M. Sreelakshmi Assistant Professor Department of Computer Science & Engineering Mar Athanasius College of Engineering Kothamangalam Kerala India Department of Computer Science & Engineering Mar Athanasius College of Engineering Kothamangalam Kerala India

Contemporary computer vision systems are typically trained to recognize a limited set of predetermined object categories. This restricted type of supervision limits their applicability and generalization since additional labeled data is needed to indicate any other visual concept. A potential approach that leverages a broader source of supervision involves learning directly from raw text about images. The proposed solution involves implementing a multimodal search in a fashion store using the Contrastive Language Image Pretraining (CLIP) model, which enables vector-based search rather than syntax-based search. CLIP improves product visibility and assists users in obtaining better search results that match their preferences. It is demonstrated that a simple pre-training task of predicting the corresponding image for a given caption is an effective and scalable technique for learning state-of-the-art image representations from scratch using a dataset of (image, text) pairs collected from the internet. Natural language is employed during pre-training to refer to learned visual concepts or describe new ones, enabling zero-shot transfer of the model to downstream tasks. Integration with clustering techniques on the vector space further enhances CLIP’s capabilities, facilitating efficient and accurate retrieval of items. By grouping similar fashion items based on visual similarity and semantic relationships, the system enables efficient and accurate retrieval of relevant items. The model can generalize effectively to most tasks without requiring any training on a specific dataset and is frequently on par with a fully supervised baseline.

关键词：

来源：评论

学校读者我要写书评

暂无评论

IoT Research on Healthcare Data to Control Illness of Oncology

IoT Research on Healthcare Data to Control Illness of Oncolo...

引用

Advance Computing and Innovative Technologies in engineering (ICACITE), International Conference on

作者： Dhiraj Kapila Prince Tehseen Ganai G.Sreenivasa Raju Lims Thomas Manisha Girase José Luis Arias Gonzales Assistant Professor Department of Computer Science &#x0026 Engineering Lovely Professional University Phagwara India Research Scholar Department of Information Technology Shri Jagdishprasad Jhabarmal Tibrewala University Rajasthan India Assistant Professor Department of Electronics and Communication Engineering Anurag University Assistant Professor Department of Social Work Vimala College (Autonomous) Thrissur Professor &#x0026 Head Dept. Of Microbiology K SK W Arts Science &#x0026 Commerce College CIDCO Nashik Maharashtra Departrnent of business University of British Columbia

Let us begin with the internet of things that empowers software programmers to start increasing prediction without being done with full to do so. With today's data availability, machine learning techniques are being developed to improve performance and maintenance prediction. Cancer, according the Health Organisation, is a global disease and the second leading cause of death worldwide. People with cancer have been identified as having an increased risk of dying in the current worldwide epidemic. With the disease's incidence growing decade after decade, the need for a continuously and then round surveillance apparatus has become crucial. A cancer monitoring system known as the internet of Things (IoT) has the potential to identify cancer-related indications in their initial phases, continue to monitor disease individuals, and screen someone who has been cured through post-treatment operations. This research presents a multi-layered architectural for an IoT-based disease monitoring and control system that might be utilised as a foundation for remotely cancer diagnosis and management.

关键词： Wireless communication Hospitals Surveillance Machine learning Maintenance engineering Oncology Software

来源：评论

学校读者我要写书评

暂无评论

Green materials & construction: A step towards sustainable environment

引用

AIP Conference Proceedings 2024年第1期3050卷

作者： Swedika Sharma Assistant Professor Department of Computer Science and Engineering & USB Chandigarh University Punjab 140431 India

The built environment, energy, and thermal comforts are all interrelated. The remarkable expansion by construction sector, lead to dense pollution due to the population growth and increased thermal comfort needs. Construction, one of India’s primary economic sectors, compared to the 6% global average, is expanding at a 9% yearly pace. Buildings require a lot of energy to construct, operate, and maintain. Greater than 33 % of electricity in India are used by the building sector. The numerous steps made throughout the world to lower building energy use and carbon footprint via the use of green building techniques are addressed in this article. The current status of green building programmes, as well as the grading and certification processes for green buildings in India, are also covered.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A global analysis and outlook for the sustainable materials approach to corporate social responsibility

引用

AIP Conference Proceedings 2024年第1期3050卷

作者： Swedika Sharma Assistant Professor Department of Computer Science and Engineering & USB Chandigarh University Punjab 140431 India

As a result of the globalization process, businesses are compelled to follow CSR and best practises from a sustainable stance towards their stakeholders and society. By evaluating the importance of corporate social responsibility (CSR) and its relationship to sustainability, the explicit purpose is to develop trends and future study fields. The global evolution of this field’s study was looked at from 2001 to 2018. The scientific production of the journals, authors, organisations, and nations that contributed to this study was evaluated by a bibliometric analysis of 1832 publications. There is growing evidence that people are interested in learning more about the connections between sustainability and socially responsible behaviour. The core category is made up of accounting, management, and business. The most successful journal is the Journal of Business Ethics and Sustainability. The United States is the country with the most publications and quotations. France and China are the two countries that collaborate on international projects the most. Global research has been trending upward in recent years, with great publication rates.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Game theory-based optimization for efficient IoT task offloading in 6G network base stations

引用

Measurement: Sensors 2024年 33卷

作者： Keshta, Ismail Soni, Mukesh Deb, Nabamita singh, Shweta Saravanan, K. Raza Khan, Dr Ihtiram Computer Science and Information Systems Department College of Applied Sciences AlMaarefa University Riyadh Saudi Arabia Dr. D. Y. Patil Vidyapeeth Pune Dr. D. Y. Patil School of Science & Technology Tathawade Pune India Assistant Professor Department of Information Technology Gauhati University India Department of Information Technology R.M.D. Engineering College Chennai India Computer Science Department Jamia Hamdard Delhi India Electronics and communication department IES college of technology Bhopal India

This study aims to optimise computing for intricate jobs within the overlapping coverage of 6G network base stations. A multi-access edge computing network model is created by solving the issues of task offloading. This model involves many base stations and IoT devices. It takes into account factors such as task delay, energy consumption, societal impacts, and economic incentives. The objective of joint optimization is to maximize the revenues of base stations and the utilities of IoT devices by considering base station price, IoT device base station selection, and job offloading mechanisms. The IoT device base station selection is addressed using a many-to-one matching game model, while pricing and task offloading interactions are managed by a two-stage approach based on Stackelberg game theory. The suggested Optimal Pricing and Best Response Algorithm (OBGT), which is based on Game Theory, successfully reaches equilibrium solutions. It demonstrates rapid convergence in simulations and improves both base station profits and IoT device utility. This study integrates data-driven mobile computing systems assurance for futuristic smart vertical networks, enhancing the efficiency of task offloading and optimization. © 2024 The Authors

关键词： Game theory

来源：评论

学校读者我要写书评

暂无评论

Data hiding in whatsapp chats using horoscopes

引用

AIP Conference Proceedings 2024年第1期3168卷

作者： Monika Agarwal Assistant Professor Department of Computer Science and Engineering RKGIT Ghaziabad U. P. India

Steganography uses the technique of espionaging a media object in an another innocent looking file so that no one can doubt of its existence. Text steganography is a sub domain of steganography in the sense that it conceals the secret message in a text file. This article proposes a new way of hiding textual messages using steganography. The method uses Lagna chart of horoscope as well as Aadhar number of recipient to hide the message. Each character of the input message is masked to a value using digits in a random lagna Chart and recipient’s Aadhar number. Then this value is hidden under an emoticon using a pre-shared emoji table. Finally, cover message is created and sent via WhatsApp. The proposed approach also compares the results of various parameters like robustness, capacity ratio and security with some existing text steganographic methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Automated data preprocessing and training interface for machine learning applications

引用

AIP Conference Proceedings 2024年第1期2942卷

作者： Anagha Challa Akhila Vutukuri Maithreyi Kenguva Sateesh Kumar Kanagala Department of Electronics and Computer Engineering Sreenidhi Institute of Science and Technology Hyderabad India Assistant Professor Department of Electronics and Computer Engineering Sreenidhi Institute of Science and Technology Hyderabad India

Machine learning is a promising tool for analyzing and predicting data. However, it is very time-consuming and challenging to manually process the data. There are also many missing values in the data that can affect the accuracy of the models. We propose an automated approach to pre-process the data. It efficiently performs the various steps involved in the data cleaning process by converting the categorical values to a label encoded value, sampling the data, and replacing the missing values with the appropriate central tendency. This project takes the raw data as input and provides a clean error-free dataset that is suitable for training machine learning models. It also aims to use a web application to automate data pre-processing and training of classification and regression models. It uses multiple algorithms and normalization strategies to get the optimum model based on the desired metric. It allows users to download the clean dataset as well as the trained models after the models have been trained. Multiple datasets, including binary classification and regression, are used to test this model.

关键词：

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：