105,811 Hits in 3.8 sec

AVN: An Adversarial Variation Network Model for Handwritten Signature Verification

Huan Li, Ping Wei, Ping Hu
2021 IEEE transactions on multimedia  
The proposed model is trained in an adversarial way with a min-max loss function, by which the three modules cooperate and compete to enhance the entire model's ability and therefore the signature verification  ...  features of handwritten signatures, the discriminator aims to make verification decisions based on the extracted features, and the variator is designed to actively generate signature variants for constructing  ...  ACKNOWLEDGMENT This research was supported by the grants National Natural Science Foundation of China No. 61876149 and China Postdoctoral Science Foundation 2018M643657.  ... 
doi:10.1109/tmm.2021.3056217 fatcat:2g3uonr3e5br7onkt37nhjp2q4

SynSig2Vec: Learning Representations from Synthetic Dynamic Signatures for Real-world Verification [article]

Songxuan Lai, Lianwen Jin, Luojun Lin, Yecheng Zhu, Huiyun Mao
2019 arXiv   pre-print
relative and fine-grained signature similarities.  ...  Then, given the templates, we construct a lightweight one-dimensional convolutional network to learn to rank the synthesized samples, and directly optimize the average precision of the ranking to exploit  ...  Average Precision Optimization Given one genuine signature and its synthetic samples, we compute and rank their similarities and incorporate the AP of the ranking into the loss function for optimization  ... 
arXiv:1911.05358v2 fatcat:xmdarsxamrdy7d7ujwvvqwzuku

SynSig2Vec: Learning Representations from Synthetic Dynamic Signatures for Real-World Verification

Songxuan Lai, Lianwen Jin, Luojun Lin, Yecheng Zhu, Huiyun Mao
relative and fine-grained signature similarities.  ...  Then, given the templates, we construct a lightweight one-dimensional convolutional network to learn to rank the synthesized samples, and directly optimize the average precision of the ranking to exploit  ...  Average Precision Optimization Given one genuine signature and its synthetic samples, we compute and rank their similarities and incorporate the AP of the ranking into the loss function for optimization  ... 
doi:10.1609/aaai.v34i01.5416 fatcat:xcx2w45xsjgwxhww6omf2texaq

A Deep Learning Model to Extract Ship Size From Sentinel-1 SAR Images

Yibin Ren, Xiaofeng Li, Huan Xu
2021 IEEE Transactions on Geoscience and Remote Sensing  
We design a custom loss function named mean scaled square error (MSSE) to optimize the DNN-based model. The DNN-based model is concatenated with the SSD-based model to form the integrated SSENet.  ...  Index Terms-Custom loss function, deep learning (DL), deep neural network (DNN) regression, ship size extraction, synthetic aperture radar (SAR) image.  ...  Calculating MSSE Loss and Optimizing SSENet We design a new loss function, MSSE, as the loss function of the DNN-based ship size regression model.  ... 
doi:10.1109/tgrs.2021.3063216 fatcat:xs7vpko4vrcg5fbsyes3wowq7e

The radar information channel and system uncertainty

John A. Malas, John A. Cortese
2010 2010 IEEE Radar Conference  
The radar information channel is developed as a theoretical model for the study of uncertainty within the design, development, and research of radar signature exploitation systems.  ...  Information measures are developed which characterize sources of uncertainty and propagate the associated impacts to system performance.  ...  These effects limit the exploitation of physics-based features and result in a loss in information that can be extracted from signature measurements.  ... 
doi:10.1109/radar.2010.5494635 fatcat:adnp64hyhbhorf4mnvwujzs7sq

Deep Learning Methods for Signature Verification [article]

Zihan Zeng, Jing Tian
2019 arXiv   pre-print
We also improve Path Signature Features by encoding temporal information in order to enlarge the discrepancy between genuine and forgery signatures.  ...  Signature is widely used in human daily lives, and serves as a supplementary characteristic for verifying human identity. However, there is rare work of verifying signature.  ...  We make use of standard cross entropy loss, and select Adam optimizer to minimize the selected objective function. Learning rate plays an vital role in the deep learning applications.  ... 
arXiv:1912.05435v1 fatcat:w4rfsulgyvg7dg27dbhc2xmroi

A Novel Image Watermarking Scheme Based on Wavelet Transform and Genetic Algorithm

Nagraju Mood, Venkata Konkula
2018 International Journal of Intelligent Engineering and Systems  
The proposed watermarking scheme accomplishes RWT and SVD for feature extraction and the GA for optimization.  ...  The Genetic Algorithm proposed in this approach adopts intelligent property and the signature process adopts more security.  ...  Compared to the conventional wavelet transforms, the RWT reduces the information loss in the extracted watermark.  ... 
doi:10.22266/ijies2018.0630.27 fatcat:7byit2dwunfnvpt6un2jywpcbe

Neural Style Transfer Enhanced Training Support For Human Activity Recognition [article]

Shelly Vishwakarma, Wenda Li, Chong Tang, Karl Woodbridge, Raviraj Adve, Kevin Chetty
2021 arXiv   pre-print
The proposed network extracts environmental effects such as noise, multipath, and occlusions effects directly from the measurement data and transfers these features to our clean simulated signatures.  ...  This results in more realistic-looking signatures qualitatively and quantitatively.  ...  ACKNOWLEDGMENTS This work is part of the OPERA project funded by the UK Engineering and Physical Sciences Research Council (EPSRC), Grant No: EP/R018677/1.  ... 
arXiv:2107.12821v1 fatcat:viis5xt7vzbu7nnrzllg4twnh4

Offline Signature Identification and Verification Based on Capsule Representations

Dilara Gumusbas, Tulay Yildirim
2020 Cybernetics and Information Technologies  
Since signature samples per user are limited and feature orientations in signature samples are highly informative, this paper first aims to evaluate the capability of Capsule Network for signature identification  ...  AbstractOffline signature is one of the frequently used biometric traits in daily life and yet skilled forgeries are posing a great challenge for offline signature verification.  ...  Two genuine (first two rows) and one forgery signature (last row) samples from CEDAR, GPDS and MCYT databases, respectively [2] [3] [4] Many research has been devoted to extracting the most informative  ... 
doi:10.2478/cait-2020-0040 fatcat:wp2h4yx2jbha3g6wy4vohkfsqe

Dependence-Analysis-Based Data-Refinement in Optical Scatterometry for Fast Nanostructure Reconstruction

Dong, Chen, Wang, Shi, Jiang, Liu
2019 Applied Sciences  
Our experiments demonstrated the capability of the proposed method in an optimized selection of a subset of measurement wavelengths that contained sufficient information for profile reconstruction and  ...  We propose a method based on dependence analysis to identify and then eliminate the measurement configurations with redundant information.  ...  Moreover, the employment of the redundant information, which is insensitive to the measurands in solving inverse problems, leads to the loss in precision as well.  ... 
doi:10.3390/app9194091 fatcat:dcma53ij2zev7l3mxffscjheai

BlackMarks: Blackbox Multibit Watermarking for Deep Neural Networks [article]

Huili Chen, Bita Darvish Rouhani, Farinaz Koushanfar
2019 arXiv   pre-print
To extract the WM, the remote model is queried by the WM key images and the owner's signature is decoded from the corresponding predictions according to the designed encoding scheme.  ...  BlackMarks takes the pre-trained unmarked model and the owner's binary signature as inputs and outputs the corresponding marked model with a set of watermark keys.  ...  Here, we use Hamming Distance as the loss function L W M to measure the difference between the extracted signature (obtained by decode predictions) and the true signature b.  ... 
arXiv:1904.00344v1 fatcat:5tsxan644fcxrbsarnolegf2v4

3D Deep Affine-Invariant Shape Learning for Brain MR Image Segmentation [article]

Zhou He, Siqi Bao, Albert Chung
2019 arXiv   pre-print
More recently, some works have presented approaches to incorporate shape information.  ...  Experiments on human brain MRI segmentation demonstrate that our approach can achieve a lower Hausdorff distance and higher Dice coefficient than the state-of-the-art approaches.  ...  Concretely, the shape loss measures the difference in shape signature, while shape signature extracted by a network trained to minimize the difference in shape signature between two affine pairs of the  ... 
arXiv:1909.06629v2 fatcat:6s2ecljv3jgrrcgz53ahgfie74


P.N .Ganorkar .
2014 International Journal of Research in Engineering and Technology  
It also referred as static and dynamic. Signature verification also used to provide authentication to user.  ...  The proposed paper present slope based method to identify the signature using various parameters like speed, time, pressure, movement of signature, accuracy and matching percentage of two signature etc  ...  To achieve more reliable information for verification and identification we should use something that really recognize given person signature verification is one in that .Signature verification techniques  ... 
doi:10.15623/ijret.2014.0308059 fatcat:li6j53j7tvcpdhxewgr3una7nq

Predicting ionizing radiation exposure using biochemically-inspired genomic machine learning

Jonathan Z.L. Zhao, Eliseos J. Mucaki, Peter K. Rogan
2018 F1000Research  
Optimal signatures were derived by backward, complete, and forward sequential feature selection using Support Vector Machines (SVM), and validated using k-fold or traditional validation on independent  ...  Certain multi-class murine signatures have sufficient granularity in dose estimation to inform eligibility for cytokine therapy (assuming these signatures could be translated to humans).  ...  In the FS Misclass. and FS Log Loss columns, one value is always N/A because signatures are derived by optimizing either misclassification or log loss, but never both.  ... 
doi:10.12688/f1000research.14048.2 fatcat:wftijuj3azhhvbag5dgrkvbdvq

Quality-Optimized and Secure End-to-End Authentication for Media Delivery

Qibin Sun, John Apostolopoulos, Chang Wen Chen, Shih-Fu Chang
2008 Proceedings of the IEEE  
Conventional data authentication can not be directly applied for streaming media when an unreliable channel is used and packet loss may occur.  ...  By applying conventional cryptographic hashes and digital signatures to the media packets, the system security is similar to that achievable in conventional data security.  ...  The experimental results presented in this paper are merely for illustrative purposes, and more detailed and rigorous test results are given in [20] [21] [22] [39] [40] [43] .  ... 
doi:10.1109/jproc.2007.909926 fatcat:jqhgrv43dnb6tiezksw5l2heoa
« Previous Showing results 1 — 15 out of 105,811 results