Multimodal biometrics for improvised security

Arora, Garima; Pavani, P. Lakshmi; Kohli, Rashi; Bibhu, Vimal

doi:10.1109/iciccs.2016.7542312

Cited by 6 publications

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The text component uses the pre-trained BERT model defined in the Hugginface library. 2 This model consists of a bidirectional Transformer pre-trained on a large corpus using a combination of masked linguistic modelling target and next-sentence prediction tasks. In the current experimental setup, the model was fine-tuned using an output linear layer that transforms the 768 outputs into 10, corresponding to the number of classes (cities) of the problem at hand.…”

Section: City Classifiermentioning

confidence: 99%

See 1 more Smart Citation

Detecting and locating trending places using multimodal social network data

Lucas

Tomás

García-Rodríguez

2022

Multimed Tools Appl

View full text Add to dashboard Cite

This paper presents a machine learning-based classifier for detecting points of interest through the combined use of images and text from social networks. This model exploits the transfer learning capabilities of the neural network architecture CLIP (Contrastive Language-Image Pre-Training) in multimodal environments using image and text. Different methodologies based on multimodal information are explored for the geolocation of the places detected. To this end, pre-trained neural network models are used for the classification of images and their associated texts. The result is a system that allows creating new synergies between images and texts in order to detect and geolocate trending places that has not been previously tagged by any other means, providing potentially relevant information for tasks such as cataloging specific types of places in a city for the tourism industry. The experiments carried out reveal that, in general, textual information is more accurate and relevant than visual cues in this multimodal setting.

show abstract

Section: City Classifiermentioning

confidence: 99%

“…In the area of cybersecurity, the paper presented in [2] shows a 4-fold security system using fusion of facial recognition, retina pattern and fingerprint pattern along with the personal password, which is recognised using keystroke dynamics. Their goal was to reduce the probability of false acceptance rates and false rejection rates.…”

Section: Introductionmentioning

confidence: 99%

Detecting and locating trending places using multimodal social network data

Lucas

Tomás

García-Rodríguez

2022

Multimed Tools Appl

View full text Add to dashboard Cite

show abstract

“…Podemos citar alguns exemplos de seu uso: caixas eletrônicos; smartphones, tablets e notebooks; controle ISSN: 1137-3601 (print), 1988-3064 (on-line) c IBERAMIA and the authors eletrônico de acesso aos locais de trabalho; identificação de usuários de planos de saúde. Em alguns casos, para aumentar ainda mais as taxas de correta identificação individual e o grau de confiabilidade no sistema que identificará o indivíduo, são utilizados sistemas biométricos multimodais, envolvendo mais de uma característica biométrica na identificação, como por exemplo, um sistema capturar a imagem dá ıris e da impressão digital e utilizar ambas para a identificação [3] [4][5] [6].…”

Section: Introductionunclassified

“…http://scikit-learn.org/ Acessado em 22/02/2019 4. https://pypi.org/project/PyWavelets/ Acessado em 22/02/2019…”

unclassified

Untitled

Brito¹,

Ling²

2019

View full text Add to dashboard Cite

This paper describes in detail different hand vein recognition methods based on Wavelet-SVM, Wavelet-ANN and Image Registration. A new image segmentation and processing algorithm is proposed to efficiently locate vein regions and suitable for feature extraction (wavelet coefficients and normalized vein imagens) and classification (SVM, ANN and Image Registration). For real time recognition and high recognition rate, we proposed an integrated system which combines three above mentioned classification methods. The simulation results reveal that the proposed integrated system achieves 1% false rejection rate (FRR) and 0.02% false acceptance rate (FAR). Resumo Este artigo descreve em detalhes diferentes métodos para o reconhecimento biométrico com base nas veias das mãos. Os métodos têm como base Wavelets com SVM, Wavelets com Rede Neural Artificial e Registro de Imagens. Um novo método de segmentação de imagensé proposto para localizar de modo adequado e eficiente a região das veias, permitindo a extração de características (com Wavelets ou a partir de uma imagem normalizada) e a classificação (com SVM, Rede Neural e Registro de Imagens). Para uma alta taxa de reconhecimento realizada em tempo real nós propomos um sistema híbrido que combina os três métodos de classificação mencionados. Os resultados de testes revelam que o sistema híbrido fornece uma taxa de falsa rejeição (FRR) de 1% para uma taxa de falsa aceitação (FAR) de 0.02%.

show abstract

A Nascent Approach for Noise Reduction via EMD Thresholding

Kohli

Gupta

2019

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

Multimodal biometrics for improvised security

Cited by 6 publications

References 14 publications

Detecting and locating trending places using multimodal social network data

Detecting and locating trending places using multimodal social network data

Untitled

A Nascent Approach for Noise Reduction via EMD Thresholding

Contact Info

Product

Resources

About