Filters








74 Hits in 8.7 sec

KutralNet: A Portable Deep Learning Model for Fire Recognition [article]

Angel Ayala, Bruno Fernandes, Francisco Cruz, David Macêdo, Adriano L. I. Oliveira, Cleber Zanchettin
2020 arXiv   pre-print
Additionally, we propose a portable approach for fire recognition and the use of modern techniques such as inverted residual block, convolutions like depth-wise, and octave, to reduce the model's computational  ...  In this work, we propose a new deep learning architecture that requires fewer floating-point operations (flops) for fire recognition.  ...  was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior -Brasil (CAPES) -Finance Code 001, Fundação de Amparo a Ciência e Tecnologia do Estado de Pernambuco (FACEPE), and  ... 
arXiv:2008.06866v1 fatcat:qm6ium66gbgojim6v7rvqcqy7m

Convolution Optimization in Fire Classification

Angel Ayala, Bruno Fernandes, Francisco Cruz, David Macedo, Cleber Zanchettin
2022 IEEE Access  
In this work, we propose the KutralNext architecture, an efficient model for single-and multi-label fire and smoke recognition tasks.  ...  For this research, we designed a novel specific-purpose model for fire and smoke recognition using still images and the study of state-of-the-art convolution techniques to improve the trade-off between  ...  MODEL EFFICIENCY TECHNIQUES BACKGROUND With the success of deep convolutional neural networks, efficient techniques appeared with newly proposed models.  ... 
doi:10.1109/access.2022.3151660 fatcat:wnp7db3nh5hdjng42qepmpexty

Prediction of Sound Insulation Using Artificial Neural Networks—Part II: Lightweight Wooden Façade Structures

Mohamad Bader Bader Eddin, Nikolaos-Georgios Vardaxis, Sylvain Ménard, Delphine Bard Bard Hagberg, Jean-Luc Kouyoumji
2022 Applied Sciences  
A prediction model based on artificial neural networks is adapted to forecast the acoustic performance of airborne sound insulation of various lightweight wooden façade walls.  ...  For each façade wall, geometric and physical information (material type, dimensions, thicknesses, densities, and more) are used as input parameters.  ...  In building acoustics, a convolutional neural networks model was used to classify inter-floor noise by recording different noise sources for 24 h in a household [43] .  ... 
doi:10.3390/app12146983 fatcat:pmk5zhdwjvamtgvqqsvhi2faum

A Survey on Deep Domain Adaptation and Tiny Object Detection Challenges, Techniques and Datasets [article]

Muhammed Muzammul, Xi Li
2021 arXiv   pre-print
In part 3), To obtain knowledge-able findings, we discussed different object detection methods, i.e., convolutions and convolutional neural networks (CNN), pooling operations with trending types.  ...  In the future, OD methods and models can be analyzed for real-time object detection, tracking strategies.  ...  Faults tolerance in Convolutional Neural Networks: FT-CNN used for finding spots during object recognition processing.  ... 
arXiv:2107.07927v1 fatcat:pgwxu5tnvzhj7ln3ccndmpilsi

Iris Recognition Development Techniques: A Comprehensive Review

Jasem Rahman Malgheet, Noridayu Bt Manshor, Lilly Suriani Affendey, Alfian Bin Abdul Halin, Rosa M. Lopez Gutierrez
2021 Complexity  
Among authentication techniques, iris recognition systems have received attention very much due to their rich iris texture which gives robust standards for identifying individuals.  ...  In addition, the researchers discuss the advantages and disadvantages of previous techniques as well as the limitations and benefits of both the traditional and deep learning approaches of iris recognition  ...  [204] introduced two networks to process the iris segmentation problem, namely, multi-scale fully convolutional networks (MFCNs) and hierarchical convolutional neural networks (HCNNs), to be used for  ... 
doi:10.1155/2021/6641247 fatcat:xil2fsokf5eetc2cx4br7bcquy

Machine Learning and Deep Learning frameworks and libraries for large-scale data mining: a survey

Giang Nguyen, Stefan Dlugolinsky, Martin Bobák, Viet Tran, Álvaro López García, Ignacio Heredia, Peter Malík, Ladislav Hluchý
2019 Artificial Intelligence Review  
It also provides an overview of massive parallelism support that is capable of scaling computation effectively and efficiently in the era of Big Data.  ...  While the number of Machine Learning algorithms is extensive and growing, their implementations through frameworks and libraries is also extensive and growing too.  ...  The authors would like to thanks to all colleagues, especially for Ján Astaloš for knowledge sharing and teamwork.  ... 
doi:10.1007/s10462-018-09679-z fatcat:ueffoypwlva4ndo35g5gzfrpcy

Deep Neural Networks for Marine Debris Detection in Sonar Images [article]

Matias Valdenegro-Toro
2019 arXiv   pre-print
Our results show that for the evaluated tasks, DNNs are a superior technique than the corresponding state of the art. There are large gains particularly for the matching and detection proposal tasks.  ...  Submerged marine debris threatens marine life, and for shallow coastal areas, it can also threaten fishing vessels [I\~niguez et al. 2016, Renewable and Sustainable Energy Reviews].  ...  Network (FCN) 27 for efficient evaluation of objectness computation 27 Jonathan Long, Evan Shelhamer, and Trevor Darrell. Fully convolutional networks for semantic segmentation.  ... 
arXiv:1905.05241v1 fatcat:5t5qm54zjvfdfm3yi5mdgsgbyu

KutralNext: An Efficient Multi-label Fire and Smoke Image Recognition Model

Angel Ayala, David Macêdo, Cleber Zanchettin, Francisco Cruz, Bruno Fernandes
2021 Anais Estendidos da XXXIV Conference on Graphics, Patterns and Images (SIBRAPI Estendido 2021)   unpublished
We proposed the KutralNext architecture, an efficient model with reduced number of layers and computacional resources for singleand multi-label fire and smoke recognition tasks.  ...  In this work, is studied the trade-off between accuracy and complexity of the inverted residual block and the octave convolution techniques, which reduces the model's size and computation requirements.  ...  financed in part by the Coordenac ¸ão de Aperfeic ¸oamento de Pessoal de Nível Superior -Brasil (CAPES) -Finance Code 001, Fundac ¸ão de Amparo a Ciência e Tecnologia do Estado de Pernambuco (FACEPE), and  ... 
doi:10.5753/sibgrapi.est.2021.20007 fatcat:riqk7jxvnragjjfcvs4ponhesa

Smart Homes: How Much Will They Support Us? A Research on Recent Trends and Advances

Adam Zielonka, Marcin Wozniak, Sahil Garg, Georges Kaddoum, Md. Jalil Piran, Ghulam Muhammad
2021 IEEE Access  
In such systems, Convolutional Neural Networks and their derivatives are most reported as those with the highest efficiency [42] .  ...  The idea was to use a sensor network for fast detection of fire symptoms and if detected to immediately inform residents and firefighters.  ...  He is currently working on numerical methods, particularly, by applying computational intelligence, swarm intelligence algorithms and smart environments for Internet of Things.  ... 
doi:10.1109/access.2021.3054575 fatcat:waqnhnxdirellnok3lne5ranhy

Capturing high-frequency phenomena using a bandwidth-limited sensor network

Ben Greenstein, Christopher Mar, Alex Pesterev, Shahin Farshchi, Eddie Kohler, Jack Judy, Deborah Estrin
2006 Proceedings of the 4th international conference on Embedded networked sensor systems - SenSys '06  
With VANGO we have developed new applications: the first acoustic collection system for motes responsive to changing environmental conditions and user interests, and the first neural spike acquisition  ...  VANGO provides a cross-platform library for data transformation, measurement, and classification; a fast and low-jitter data acquisition system for motes; and a mechanism to control mote and microserver  ...  ACKNOWLEDGMENTS We would like to thank our reviewers and our shepherd, Sam Madden, for valuable feedback; Mohammad Rahimi for hardware support; and the TinyOS and Emstar communities for software contributions  ... 
doi:10.1145/1182807.1182835 dblp:conf/sensys/GreensteinMPFKJE06 fatcat:yduq3cb5mjeelao3yqobvv6ss4

27th Annual Computational Neuroscience Meeting (CNS*2018): Part One

2018 BMC Neuroscience  
Networking Fund of the Helmholtz Association and the Helmholtz Portfolio theme "Supercomputing and Modeling for the Human Brain" and the European Union Seventh Framework Programme (FP7/2007-2013) under  ...  the neural recordings of locust.  ...  To meet our goal, we treated burst detection as an image-recognition problem, and trained a deep convolutional neural network to detect a specific activation pattern of spikes occurring across the cortical  ... 
doi:10.1186/s12868-018-0452-x pmid:30373544 pmcid:PMC6205781 fatcat:xv7pgbp76zbdfksl545xof2vzy

16th Sound and Music Computing Conference SMC 2019 (28–31 May 2019, Malaga, Spain)

Lorenzo J. Tardón, Isabel Barbancho, Ana M. Barbancho, Alberto Peinado, Stefania Serafin, Federico Avanzini
2019 Applied Sciences  
The 16th Sound and Music Computing Conference (SMC 2019) took place in Malaga, Spain, 28–31 May 2019 and it was organized by the Application of Information and Communication Technologies Research group  ...  The SMC 2019 TOPICS OF INTEREST included a wide selection of topics related to acoustics, psychoacoustics, music, technology for music, audio analysis, musicology, sonification, music games, machine learning  ...  Committee and other collaborators.  ... 
doi:10.3390/app9122492 fatcat:tcacoupffjewnpjhpw4oy7x6h4

Review on Vehicle Detection Technology for Unmanned Ground Vehicles

Qi Liu, Zirui Li, Shihua Yuan, Yuzheng Zhu, Xueyuan Li
2021 Sensors  
Environmental perception technology is the foundation of UGVs, which is of great significance to achieve a safer and more efficient performance.  ...  This article firstly introduces commonly used sensors for vehicle detection, lists their application scenarios and compares the strengths and weakness of different sensors.  ...  The framework of YOLO was a deep convolutional neural network (DCNN) and full convolutional neural network (FCNN), where DCNN was used to extract image features and greatly reduce its resolution to improve  ... 
doi:10.3390/s21041354 pmid:33672976 fatcat:ammlsccxbbhgpkx6r5vod7ciuy

In-materio neuromimetic devices: Dynamics, information processing and pattern recognition [article]

Dawid Przyczyna, Piotr Zawal, Tomasz Mazur, Pier Luigi Gentili, Konrad Szaciłowski
2020 arXiv   pre-print
with more sophisticated implementations, including signal processing, speech recognition and data security.  ...  This feature article is devoted to various in materio implementation of neuromimetic processes, including neuronal dynamics, synaptic plasticity, and higher-level signal and information processing, along  ...  Scheme of the emulated neural network (b).  ... 
arXiv:2002.07712v1 fatcat:s7fgywfnrbhcrdndggeb7rulfu

Machine Learning and Data Analytics for Design and Manufacturing of High-Entropy Materials Exhibiting Mechanical or Fatigue Properties of Interest [article]

Baldur Steingrimsson, Xuesong Fan, Anand Kulkarni, Michael C. Gao, Peter K. Liaw
2020 arXiv   pre-print
In case that an artificial neural network (ANN) is deemed suitable for the applications at hand, it is suggested to employ custom kernel functions consistent with the underlying physics, for the purpose  ...  The main focus is on alloys and composites with large composition spaces for structural materials.  ...  For in-depth analysis of sample-complexity of convolutional neural networks (CNNs) and recurrent neural networks (RNNs), refer to [104] .  ... 
arXiv:2012.07583v1 fatcat:nppdy5ivendddpdzihyz65wrsq
« Previous Showing results 1 — 15 out of 74 results