检索结果-内蒙古大学图书馆

PyFCS: A new Python library to create and manipulate Fuzzy Color Spaces

学校读者我要写书评

暂无评论

PyFCS: A new Python library to create and manipulate Fuzzy C...

IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)

作者： Rafael Vazquez-Conejo María Tejada-Casado Luis Javier Herrera Razvan Ghinea Jose Manuel Soto-Hidalgo Dept. Optics University of Granada Spain Dpt. Optics University of Granada Spain Dpt. Computer Engineering Automation and Robotics University of Granada Spain

ISBN: (数字)9798350319545

ISBN: (纸本)9798350319552

Color, as a fundamental element in digital imagery, plays a vital role across various domains, from art and design to scientific imaging. However, its representation and manipulation are challenged by the inherent subjectivity of human color perception and the limitations of traditional color naming and categorization methods. Addressing these challenges, this paper introduces PyFCS, a novel Python library developed to create and manipulate fuzzy color spaces, offering a more nuanced and realistic approach to color modeling. PyFCS leverages fuzzy logic and the conceptual spaces theory to allow flexible and accurate representation of color categories, overcoming the constraints of crisp boundaries. Building upon the groundwork laid by the Java-based JFCS software, PyFCS harnesses Python's strengths in computation and its rich ecosystem of libraries for enhanced data processing and image analysis capabilities. The library's user-friendly and scalable design, coupled with its integration with Python's computational resources, makes it accessible to a broad range of users, from beginners to advanced researchers. The efficacy of PyFCS is demonstrated through a case study involving the creation of fuzzy color spaces from a paint catalog and its subsequent application in mural painting, exemplifying its practical utility in real-world scenarios. Furthermore, this paper details how PyFCS seamlessly integrates with other libraries to offer a comprehensive color analysis, exemplified by linguistic description of colors present in an urban mural painting. This study introduces a sophisticated color space tool and enhances scientific collaboration by providing open-source access to PyFCS on GitHub.

关键词： Fuzzy logic Accuracy Image color analysis Color Machine learning Media Libraries

Microcontroller Based Intelligent Chinese-Speech Keywords Detector by Transferring the Mid-level Features of Deep Speech 3

学校读者我要写书评

暂无评论

Microcontroller Based Intelligent Chinese-Speech Keywords De...

3rd International Conference on Communication Technologies, ComTech 2021

作者： Muzamal, Junaid Hussain Asghar, Muhammad Kwong, Anne Raza, Usman Ahmed Riphah International University Dpt. of Computer Science Lahore Pakistan Lahore Leads University Dpt. of Computer Science Lahore Pakistan Digital Galaxy Technologies Limited Dpt. Research and Development Guangdong Sheng China

ISBN: (纸本)9781665419017

Speech Keywords Detection (SKD) can be described as a task of finding keywords in audio streams, while only a few samples of the keyword are available. SKD has immense applicability in terms of making the temporal signature of audio, to implement the word clouds, voice-command-based appliances, and intelligent voice-based agents. Existing SKD applications suffer from high power requirement constraints, false-positive results due to low training data, and missing keywords in languages like Chinese. Besides, SKD is an enormously imbalanced classification delinquent. In this work, we proposed to implement a lower power consumer and low memory occupying solution, which is built on microcontroller-based applications. We addressed the false-positive and data imbalancing problem by employing a novel combination of objective functions metric loss and Prototypical loss with the mid-level feature engineering of Deep Speech. The extensive experiments on real-time microcontroller showed that our technique is state of the art and provide comparable results with 11% improvement in F1 score. © 2021 IEEE.

关键词： Deep neural networks

Self-Supervised Graph Representation Learning for In-The-Wild Wearable and Smartphone based Emotion Recognition

学校读者我要写书评

暂无评论

Self-Supervised Graph Representation Learning for In-The-Wil...

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Ioannis Ziogas Leontios J. Hadjileontiadis Ahsan H. Khandoker Aamna Al Shehhi Dpt. of Biomedical Eng. and Biotechnology Khalifa University of Science and Technology Abu Dhabi UAE Dpt. of Electrical and Computer Eng Aristotle University of Thessaloniki Thessaloniki Greece

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Wearable and smartphone-based emotion recognition (WER) remains a challenging setting in affective computing, due to the notorious difficulty and bias associated with in-thewild label collection. The high inter-and intra-subject emotional variability motivates us to explore WER modeling through graph node classification in a limited resources learning scheme powered by Self-Supervised Learning (SSL) graph masking augmentation tasks. We employ a subgraph sampling approach during training, utilizing labeled and unlabeled data, along with supervised, semi-supervised, and SSL mechanisms in a multi-task inductive graph neural network architecture. Our evaluations on K-EmoPhone through leave-one-group-out cross-validation in the binary arousal and valence tasks yield average accuracy gains of 4.3% and 7.8%, compared to the full resource setting, utilizing only 20% and 25% of the labels, respectively. Our model analysis sheds light on the relation of SSL graph augmentations to emotional arousal and valence and justifies the approach of SSL-driven subgraph training for in-the-wild WER.

关键词： Training Representation learning Emotion recognition Time series analysis Speech recognition Self-supervised learning Signal processing Biomedical monitoring Wearable devices Speech processing

Home Automation System based on Intelligent Transducer Enablers

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Suárez-Albela, Manuel Fraga-Lamas, Paula Fernández-Caramés, Tiago M. Dapena, Adriana González-López, Miguel Dpt. Electronics and Systems Faculty of Computer Science Universidade da Coruña A Coruña15071 Spain

This paper presents a novel home automation system named HASITE (Home Automation System based on Intelligent Transducer Enablers), which has been specifically designed to identify and configure transducers easily and fast. These features are especially useful in situations where many transducers are deployed, since their setup becomes a cumbersome task that consumes a significant amount of time and human resources. HASITE simplifies the deployment of a home automation system by using wireless networks, and both self-configuration and self-registration protocols. Thanks to the application of such three elements, HASITE is able to add new transducers by just powering them up. According to tests performed in different realistic scenarios, a transducer is ready to be used in less than 13 seconds. Moreover, all HASITE functionalities can be accessed through an API, which also allows for the integration of third-party systems. As an example, an Android application based on the API is presented. Remote users can use it to interact with transducers by just using a regular smartphone or a tablet. © 2024, CC BY.

关键词： Transducers

Transportation mode recognition based on low-rate acceleration and location signals with an attention-based multiple-instance learning network

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Siargkas, Christos Papapanagiotou, Vasileios Delopoulos, Anastasios The Multimedia Understanding Group Dpt. of Electrical and Computer Engineering Faculty of Engineering Aristotle University of Thessaloniki Greece The IMPACT research group Dpt. of Biosciences and Nutrition Karolinska Institutet Stockholm Sweden

Transportation mode recognition (TMR) is a critical component of human activity recognition (HAR) that focuses on understanding and identifying how people move within transportation systems. It is commonly based on leveraging inertial, location, or both types of signals, captured by modern smartphone devices. Each type has benefits (such as increased effectiveness) and drawbacks (such as increased battery consumption) depending on the transportation mode (TM). Combining the two types is challenging as they exhibit significant differences such as very different sampling rates. This paper focuses on the TMR task and proposes an approach for combining the two types of signals in an effective and robust classifier. Our network includes two sub-networks for processing acceleration and location signals separately, using different window sizes for each signal. The two sub-networks are designed to also embed the two types of signals into the same space so that we can then apply an attention-based multiple-instance learning classifier to recognize TM. We use very low sampling rates for both signal types to reduce battery consumption. We evaluate the proposed methodology on a publicly available dataset and compare against other well known *** Codes 68T07 Copyright © 2024, The Authors. All rights reserved.

关键词： Location

A Machine Learning-assisted Game Theoretical Service Selection Strategy in UAV Networks

学校读者我要写书评

暂无评论

A Machine Learning-assisted Game Theoretical Service Selecti...

Emerging Technologies (GCET), Global Congress on

作者： Abdessalam Mohammed Hadjkouider Chaker Abdelaziz Kerrache Yesin Sahraoui Asma Adnane Carlos T. Calafate Dpt. of Computer Science and IT Kasdi Merbah University of Ouargla Ouargla Algeria Dpt. of Computer Science Université Amar Telidji de Laghouat Laghouat Algeria National Higher School of AI Algiers Algeria Computer Science Department Loughborough University Loughborough UK Computer Engineering Department (DISCA) Universitat Politècnica de València Valencia Spain

ISBN: (数字)9798331542603

ISBN: (纸本)9798331542610

The Unmanned Aerial Vehicle (UAV) has emerged as a transformative technology application in healthcare, environmental monitoring, and search and rescue operations, with the growing complexity of their services, Dynamic Environments, and limited resources. This paper introduces a novel approach that integrates machine learning (ML) with game theory to revolutionize the decision-making process in UAV service selection. Our method focuses on enhancing user experience by employing ML algorithms to analyze user requests and predict the most suitable UAV services. Concurrently, game theory is integrated to evaluate and ensure efficient selection regarding delay, throughput, packet loss ratio, and residual energy within UAV *** synergy aims to benefit both users and service providers by optimizing service delivery and user satisfaction. Experimental results demonstrate the efficacy of our model, highlighting its superior performance in terms of accuracy, precision, recall, and F-score.

关键词： Machine learning algorithms Packet loss Machine learning Medical services Autonomous aerial vehicles Throughput Prediction algorithms User experience Game theory Vehicle dynamics

Assessing Fidelity in XAI post-hoc techniques: A Comparative Study with Ground Truth Explanations Datasets

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Miró-Nicolau, Miquel Jaume-I-Capó, Antoni Moyà-Alcover, Gabriel UGiVIA Research Group University of the Balearic Islands Dpt. of Mathematics and Computer Science Palma07122 Spain University of the Balearic Islands Dpt. of Mathematics and Computer Science Palma07122 Spain

The evaluation of the fidelity of eXplainable Artificial Intelligence (XAI) methods to their underlying models is a challenging task, primarily due to the absence of a ground truth for explanations. However, assessing fidelity is a necessary step for ensuring a correct XAI methodology. In this study, we conduct a fair and objective comparison of the current state-of-the-art XAI methods by introducing three novel image datasets with reliable ground truth for explanations. The primary objective of this comparison is to identify methods with low fidelity and eliminate them from further research, thereby promoting the development of more trustworthy and effective XAI techniques. Our results demonstrate that XAI methods based on the backpropagation of output information to input yield higher accuracy and reliability compared to methods relying on sensitivity analysis or Class Activation Maps (CAM). However, the backpropagation method tends to generate more noisy saliency maps. These findings have significant implications for the advancement of XAI methods, enabling the elimination of erroneous explanations and fostering the development of more robust and reliable XAI. © 2023, CC BY.

关键词： Sensitivity analysis

A novel approach to generate datasets with XAI ground truth to evaluate image models

学校读者我要写书评

暂无评论

arXiv 2023年

With the increased usage of artificial intelligence (AI), it is imperative to understand how these models work internally. These needs have led to the development of a new field called eXplainable artificial intelligence (XAI). This field consists of on a set of techniques that allows us to theoretically determine the cause of the AI decisions. One main issue of XAI is how to verify the works on this field, taking into consideration the lack of ground truth (GT). In this study, we propose a new method to generate datasets with GT. We conducted a set of experiments that compared our GT with real model explanations and obtained excellent results confirming that our proposed method is correct. © 2023, CC BY.

关键词： Artificial intelligence

Predicting Cryptocurrencies Market Phases through On-Chain Data Long-Term Forecasting

学校读者我要写书评

暂无评论

Predicting Cryptocurrencies Market Phases through On-Chain D...

IEEE International Conference on Blockchain and Cryptocurrency (ICBC)

作者： Bruno Casella Lorenzo Paletto Computer Science Department University of Turin Turin Italy Computer Science Dpt. University of Turin Turin Italy

Blockchain, the underlying technology of Bitcoin and several other cryptocurrencies, like Ethereum, produces a massive amount of open-access data that can be analyzed, providing important information about the network's activity and its respective token. The on-chain data have extensively been used as input to Machine Learning algorithms for predicting cryptocurrencies' future prices; however, there is a lack of study in predicting the future behaviour of on-chain data. This study aims to show how on-chain data can be used to detect cryptocurrency market regimes, like minimum and maximum, bear and bull market phases, and how forecasting these data can provide an optimal asset allocation for long-term investors.

关键词：