342 Hits in 6.6 sec

Selfie Periocular Verification using an Efficient Super-Resolution Approach [article]

Juan Tapia, Andres Valenzuela, Rodrigo Lara, Marta Gomez-Barrero, Christoph Busch
2022 arXiv   pre-print
To that end, the method implements a loss function based on the Sharpness metric used to evaluate iris images quality.  ...  We propose an Efficient Single Image Super-Resolution algorithm, which takes into account a trade-off between the efficiency and the size of its filters.  ...  [15] in a model called Enhanced Deep Residual Networks for Single Image Super Resolution (EDSR).  ... 
arXiv:2102.08449v2 fatcat:gptykhzgxfedthhy6lqhnet5ny

Selfie Periocular Verification Using an Efficient Super-Resolution Approach

Juan E. Tapia, Andres Valenzuela, Rodrigo Lara, Marta Gomez-Barrero, Christoph Busch
2022 IEEE Access  
To that end, the method implements a loss function based on the Sharpness metric used to evaluate iris images quality.  ...  We propose an Efficient Single Image Super-Resolution algorithm, which takes into account a trade-off between the efficiency and the size of its filters.  ...  [15] in a model called Enhanced Deep Residual Networks for Single Image Super Resolution (EDSR).  ... 
doi:10.1109/access.2022.3184301 fatcat:akdk3fdkmnhb5i2gnvl3gmphcm

Deep GAN-Based Cross-Spectral Cross-Resolution Iris Recognition

Moktari Mostofa, Salman Mohamadi, Jeremy Dawson, Nasser M. Nasrabadi
2021 IEEE Transactions on Biometrics Behavior and Identity Science  
We have developed two different novel techniques using the conditional generative adversarial network (cGAN) as a backbone architecture for cross-spectral iris matching.  ...  However, matching iris images acquired at different spectral bands (i.e., matching a visible (VIS) iris probe to a gallery of near-infrared (NIR) iris images or vice versa) shows a significant performance  ...  Hence, we have also used the perceptual loss, introduced in [58] , for style transfer and super-resolution.  ... 
doi:10.1109/tbiom.2021.3102736 fatcat:73ngxa4dzzaxpgwsqbqxf7eokq

Deep Learning Based Super Resolution and Classification Applications for Neonatal Thermal Images

Fatih M. Senalp, Murat Ceylan
2021 Traitement du signal  
In this way, super-resolution applications have been carried out on the deep network model developed based on generative adversarial networks (GAN) by using three different datasets.  ...  In addition, healthy - unhealthy classification application was carried out by means of a classifier network developed based on convolutional neural networks (CNN) to evaluate the super-resolution images  ...  The thermal images used in this study were obtained in project studies supported by the Scientific and Technological Research Council of Turkey (TUBITAK, project number: 215E019).  ... 
doi:10.18280/ts.380511 fatcat:rily2yulmzatznav3igcvej7bu

Enhanced Robotic Vision System Based on Deep Learning and Image Fusion

E. A. Alabdulkreem, Ahmed Sedik, Abeer D. Algarni, Ghada M. El Banby, Fathi E. Abd El-Samie, Naglaa F. Soliman
2022 Computers Materials & Continua  
Hence, the proposed system can be considered as an efficient solution for the robotic vision problem with multi-modality images.  ...  The basic objective of this system is to fuse visual and IR images for efficient feature extraction from the captured images.  ...  The concept of image super resolution can be used to obtain images with much details based on some sort of dictionary techniques and single image super resolution algorithms. Liu et al.  ... 
doi:10.32604/cmc.2022.023905 fatcat:2df7fjhvorharhqsfjydp6usx4

Single Image Super-Resolution with Arbitrary Magnification Based on High-Frequency Attention Network

Jun-Seok Yun, Seok-Bong Yoo
2022 Mathematics  
In addition, the importance of single image super-resolution at arbitrary magnification is emphasized for tasks such as object recognition and satellite image magnification.  ...  Among various developments in the field of computer vision, single image super-resolution of images is one of the most essential tasks.  ...  Related Works Conventional Single Image Super-Resolution The purpose of super-resolution is to use a low-resolution image as input and predict a corresponding high-resolution image.  ... 
doi:10.3390/math10020275 fatcat:t2n7eewhn5bw3m5sodro27bx6e

Real-World Super-Resolution using Generative Adversarial Networks

Haoyu Ren, Amin Kheradmand, Mostafa El-Khamy, Shuangquan Wang, Dongwoon Bai, Jungwon Lee
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
Our method involves deploying an ensemble of generative adversarial networks (GANs) for robust real-world SR. The ensemble deploys different GANs trained with different adversarial objectives.  ...  Robust real-world super-resolution (SR) aims to generate perception-oriented high-resolution (HR) images from the corresponding low-resolution (LR) ones, without access to the paired LR-HR ground-truth  ...  Introduction Image super-resolution (SR) generates a high-resolution (HR) image from a given low-resolution (LR) image by attempting to recover the missing information.  ... 
doi:10.1109/cvprw50498.2020.00226 dblp:conf/cvpr/RenKEWBL20 fatcat:jarencdbejcpdjljks5tzsol5m

FMPN: Fusing Multiple Progressive CNNs for Depth Map Super-Resolution

Shuaihao Li, Bin Zhang, Weiping Zhu, Xinfeng Yang
2020 IEEE Access  
This method enables simple structured Neural Networks with high accuracy, high efficiency and relatively simple network training for depth map super-resolution.  ...  The Convolution Neural Network (CNN) is widely used in the super-resolution task of depth map.  ...  In recent years, many CNN-based neural networks [22] - [26] have been used for image super-resolution tasks.  ... 
doi:10.1109/access.2020.3024650 fatcat:svkkznmiobedfb7gmid4d3mzwq

ProfileSR-GAN: A GAN based Super-Resolution Method for Generating High-Resolution Load Profiles [article]

Lidong Song, Yiyan Li, Ning Lu
2022 arXiv   pre-print
Thus, in this paper, we propose ProfileSR-GAN: a Generative Adversarial Network (GAN) based load profile super-resolution (LPSR) framework for restoring high-frequency components lost through the smoothing  ...  When training the ProfileSR-GAN generator network, to make the generated profiles more realistic, we introduce two new shape-related losses in addition to conventionally used content loss: adversarial  ...  Load Profile Super Resolution Problem Formulation Super-resolution algorithms originate from the image processing domain for processing 2-dimensional images.  ... 
arXiv:2107.09523v2 fatcat:atskp3ok7zbzbdse7d6i3w7gca

SIP-SegNet: A Deep Convolutional Encoder-Decoder Network for Joint Semantic Segmentation and Extraction of Sclera, Iris and Pupil based on Periocular Region Suppression [article]

Bilal Hassan, Ramsha Ahmed, Taimur Hassan, Naoufel Werghi
2020 arXiv   pre-print
Finally, the semantic segmentation of sclera, iris and pupil is achieved using the densely connected fully convolutional encoder-decoder network.  ...  To address these issues, SIP-SegNet begins with denoising the pristine image using denoising convolutional neural network (DnCNN), followed by reflection removal and image enhancement based on contrast  ...  We have used five subsets of the CASIA-IrisV4 database to perform this study [80].  ... 
arXiv:2003.00825v1 fatcat:gqsus26nnfccvd5sr6z7pl3ocm

Learned deconvolution using physics priors for structured light-sheet microscopy [article]

Philip Wijesinghe, Stella Corsetti, Darren J.X. Chow, Shuzo Sakata, Kylie Dunning, Kishan Dholakia
2021 bioRxiv   pre-print
Here, we introduce a deep learning method that can deconvolve and super-resolve structured light-sheet images using such fields without the need for paired experimental data.  ...  We make use of the known physics of light propagation by constraining a generative adversarial network with estimated, simulated image data.  ...  ACKNOWLEDGEMENTS We would like to acknowledge Mirna Merkler for preparing the excised mouse brain section.  ... 
doi:10.1101/2021.05.26.445797 fatcat:ppcholn3xraarj35hgrth22sga

Image Compression Using Neural Networks: A Review

Haval Tariq Sadeeq, Thamer Hassan Hameed, Abdo Sulaiman Abdi, Ayman Nashwan Abdulfatah
2021 International Journal of Online and Biomedical Engineering (iJOE)  
The most important studies are highlighted and future trends even envisaged in relation to image coding topics using neural networks.  ...  In particular, the end-to-end frames based on neural networks are reviewed, revealing fascinating explorations of frameworks/standards for next-generation image coding.  ...  It also achieves remarkable results, for example super-resolution and compression artifact reductions, also in many low-level computer vision tasks.  ... 
doi:10.3991/ijoe.v17i14.26059 fatcat:eyhsq2c55bbmjjqamccwi3lgn4

Artificial Intelligence-based Semantic Segmentation of Ocular Regions for Biometrics and Healthcare Applications

Rizwan Ali Naqvi, Dildar Hussain, Woong-Kee Loh
2020 Computers Materials & Continua  
In this paper, to address the accurate segmentation of multiple eye regions in unconstrainted scenarios, a lightweight outer residual encoder-decoder network suitable for various sensor images is proposed  ...  Comprehensive experiments were performed, and optimal performance was achieved using SBVPI and UBIRIS.v2 datasets containing images of the eye region.  ...  The proposed network can segment the input eye image into four main classes corresponding to the iris, sclera, pupil, and background region using a single model.  ... 
doi:10.32604/cmc.2020.013249 fatcat:orqnykocjzcidpf2jc7fuwo3li

Verification system based on long-range iris and Graph Siamese Neural Networks [article]

Francesco Zola, Jose Alvaro Fernandez-Carrasco, Jan Lukas Bruse, Mikel Galar, Zeno Geradts
2022 arXiv   pre-print
For this reason, in this work, we propose a novel approach that uses long-range (LR) distance images for implementing an iris verification system.  ...  More specifically, we present a novel methodology for converting LR iris images into graphs and then use Graph Siamese Neural Networks (GSNN) to predict whether two graphs belong to the same person.  ...  Super-resolution convolutional neural networks (SRCNNs) for increasing the iris resolution from selfie images and the final gender classification are used in [48] .  ... 
arXiv:2208.00785v1 fatcat:er3nc2n3bjhtrbqw76fmekcvva

Heuristic Algorithm for Efficient Data Retrieval Scheduling in the Multichannel Wireless Broadcast Environments

A. Porselvi, S.Brindha Devi
2015 International Journal of Computer Applications Technology and Research  
Wireless data broadcast is an efficient way of disseminating data to users in the mobile computing environments.  ...  A Security use of decryption with decryption key stored in RFID tags embedded within ID cards.  ...  High-frequency contents in the Fourier transform are responsible for Edges and sharp transitions in an image. Smooth areas of image appear due to low frequency contents of Fourier transform.  ... 
doi:10.7753/ijcatr0405.1001 fatcat:7xpnnpep5vd7hcrmcjr4fvpnq4
« Previous Showing results 1 — 15 out of 342 results