A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Filters
Deep Frequency Filtering for Domain Generalization
[article]
2022
arXiv
pre-print
Preserved Fulltext
In this paper, we propose Deep Frequency Filtering (DFF) for learning domain-generalizable features, which is the first endeavour to explicitly modulate frequency components of different transfer difficulties ...
Improving the generalization capability of Deep Neural Networks (DNNs) is critical for their practical uses, which has been a longstanding challenge. ...
Conclusion In this paper, we first conceptualize Deep Frequency Filtering (DFF) and point out that such a simple frequency-domain feature filtering operation can significantly facilitate domain generalization ...
arXiv:2203.12198v1
fatcat:hry4y5cypvak3owccpm467lgr4
A Deep Convolutional Neural Network Inspired by Auditory Perception for Underwater Acoustic Target Recognition
2019
Sensors
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
In the ADCNN model, inspired by the frequency component perception neural mechanism, a bank of multi-scale deep convolution filters are designed to decompose raw time domain signal into signals with different ...
Inspired by the plasticity neural mechanism, the parameters of the deep convolution filters are initialized randomly, and the is n learned and optimized for UATR. ...
Raw time-domain ship radiated noise signal The local amplification of frequency domain Frequency domain of ship radiated noise signal Frequency domain of outputs of filters The outputs of deep convolution ...
doi:10.3390/s19051104
fatcat:xciemqfga5fi5nsq5jecztz6qu
Efficient Convolutional Auto-Encoding via Random Convexification and Frequency-Domain Minimization
[article]
2016
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
domain via coordinate descent (CD). ...
To help alleviating these limitations, we propose an efficient learning strategy for layer-wise unsupervised training of deep CNNs on conventional hardware in acceptable time. ...
Acknowledgments This work is supported by the Agency for Innovation by Science and Technology in Flanders (IWT) -PhD grant nr. 131814, the VUB Interdisciplinary Research Program through the EMO-App project ...
arXiv:1611.09232v1
fatcat:5bdyj5f4qfgilhrn447nrcc6hi
Image Denoising with Control over Deep Network Hallucination
2022
IS&T International Symposium on Electronic Imaging Science and Technology
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
With our framework, the user can control the fusion of the two components in the frequency domain. ...
In this framework, we exploit the outputs of a deep denoising network alongside an image convolved with a reliable filter. ...
For the first type, we carry out the fusion in the frequency domain to smoothly blend the structure of both images. We considered two domain transforms, the DWT and the DCT for the fusion. ...
doi:10.2352/ei.2022.34.14.coimg-217
fatcat:5lfhpcw5tvhjrn6w2unwrh7o3u
Performance of Iterative Soft Decision Feedback Equalizers for Single-Carrier Transmission
2017
Journal of Electrical Engineering and Technology
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
The results shows that frequency-domain sDFE performs better than time-domain one and also that considerable gain can be obtained especially when the channel has deep nulls. ...
Specifically, we consider both time domain and frequency-domain sDFE and compare their performances. ...
Generic structure of iterative detection and decoding for single-carrier transmission through ISI channel operating in time-domain and the other in frequency domain. ...
doi:10.5370/jeet.2017.12.3.1280
fatcat:xrwb7hzxobbfdgel4fomi2afcm
STFNets: Learning Sensing Signals from the Time-Frequency Perspective with Short-Time Fourier Neural Networks
2019
The World Wide Web Conference on - WWW '19
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2019 pre-print arXiv:1902.07849v1 fatcat:2y4a77dnn5bizhnltg5lcd2ili
of signal frequencies, offering better features in the frequency domain. ...
A STFNet, therefore, demonstrates superior capability as the fundamental building block of deep neural networks for IoT applications for various sensor inputs. ...
Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation here on. ...
doi:10.1145/3308558.3313426
dblp:conf/www/YaoPJZSLLLWHS0A19
fatcat:5p3a2qbmbjbdxlcji2v5rw32v4
Dual-domain Deep Convolutional Neural Networks for Image Demoireing
2020
2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
We develop deep convolutional neural networks (CNNs) for moiré artifacts removal by exploiting the complex properties of moiré patterns in multiple complementary domains, i.e., the pixel and frequency ...
Next, we develop a dynamic filter generation network that learns dynamic blending filters. ...
The frequency network processes DCT coefficients to remove moiré artifacts in the frequency domain. Subsequently, the dynamic filter generation network learns the dynamic blending filters. ...
doi:10.1109/cvprw50498.2020.00243
dblp:conf/cvpr/AnPL20
fatcat:ollmzuelgzbnnhiuvwt5bgkbua
An Explainable Neural Network for Fault Diagnosis with a Frequency Activation Map
2021
IEEE Access
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
To generate the frequency activation map, the proposed model structure for learning the 1D vibration signals is designed to filter the frequency components of the 1D vibration signals using a 1D convolutional ...
Since the 1D vibration signal for monitoring the normal and faulty states of equipment is easy to interpret in the frequency domain, the frequency activation map provides the user with a specific frequency ...
frequency domain. ...
doi:10.1109/access.2021.3095565
fatcat:zchwtucjwfb6hbuczaf5qzeaca
Convolutional Neural Filtering for Intelligent Communications Signal Processing in Harsh Environments
2021
IEEE Access
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
INDEX TERMS Linear filter, convolution neural network, neural filtering, model-driven deep learning. ...
Aiming at utilizing artificial neural networks to enhance intelligent filtering for interfered wireless communication signal in harsh environments, a new method named convolutional neural filtering is ...
FIGURE 9 . 9 Frequency domain characteristics of learned filter for signal with different bandwith. ...
doi:10.1109/access.2021.3049950
fatcat:7uhdfj57h5h3lkhwsomga4j4pm
Design of an Optimal High Pass Filter in Frequency - Wave Number (F-K) Space for Suppressing Dispersive Ground Roll Noise from Onshore Seismic Data
2017
Universal Journal of Physics and Application
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
The filters were then applied to the entire seismic data ensemble for the filtering operation. ...
The results obtained after the two filtering operations were compared to ascertain the optimal filter which was most effective for the suppression of the unwanted ground roll noise. ...
Acknowledgements We are grateful to Shell Petroleum Development Company (SPDC) for providing the seismic datasets used for the study. ...
doi:10.13189/ujpa.2017.110502
fatcat:pxhplojuzzd5rlyqrdf4ni7mhq
Multi-features taxi destination prediction with frequency domain processing
2018
PLoS ONE
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
So, we import image frequency domain processing to taxi destination prediction to reduce noise and sparsity, then propose multi-features taxi destination prediction with frequency domain processing (MTDP-FD ...
Firstly, we transform the spatial domain trajectory image into frequency-domain representation by fast Fourier transform and reduce the noise of the trajectory images. ...
Acknowledgments This work was supported by the Fundamental Research Funds for the Central Universities (2017XKQY078). Thanks to the UCI Machine Learning Repository for Porto taxi trajectory data. ...
doi:10.1371/journal.pone.0194629
pmid:29566042
pmcid:PMC5864052
fatcat:mu6sceswvff7zc7afjfhkytnlq
Orthogonally Regularized Deep Networks For Image Super-resolution
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Aiming for faster inference and more efficient solutions than solving the SR problem in the spatial domain, we propose a novel network structure for learning the SR mapping function in an image transform ...
Deep learning methods, in particular trained Convolutional Neural Networks (CNNs) have recently been shown to produce compelling state-of-the-art results for single image Super-Resolution (SR). ...
In this paper, we begin by exploring a DCT domain deep SR method. ...
arXiv:1802.02018v1
fatcat:oilslv7wpnbdzonz6mdyqx52zy
Sampling Operator to Learn the Scalable Correlation Filter for Visual Tracking
2019
IEEE Access
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
The scalable filter encodes the sparse frequency representation to reconstruct a larger filter with zeros outside of the object in the spatial domain. ...
However, the number of available training samples is limited to the filter size, and the lack of samples leads to poor generalization. Moreover, spectral leakage degrades the filter quality. ...
The windowing operation filters the band-limited signal in the spatial domain and generates a smooth boundary for scalable frequency representation. ...
doi:10.1109/access.2019.2892429
fatcat:ppwaownm25d55clrx3od4dy2lu
Signal processing techniques for motor imagery brain computer interface: A review
2019
Array
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
In recent studies, the researchers are using deep neural networks for the classification of motor imagery tasks. ...
Authors discuss existing challenges in the domain of motor imagery brain-computer interface and suggest possible research directions. ...
The multichannel EEG signal is passed into bandpass filter for selecting the frequency. After frequency filtering, spatial filtering is performed that uses spatial filters and FIR filters. ...
doi:10.1016/j.array.2019.100003
fatcat:tlkzqreshzgfpeusxub3f5h4bq
Low-latency Monaural Speech Enhancement with Deep Filter-bank Equalizer
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
network with a deep learning-based shortened digital filter mapping network. ...
To improve the performance of traditional low-latency speech enhancement algorithms, a deep filter-bank equalizer (FBE) framework was proposed, which integrated a deep learning-based subband noise reduction ...
In the second stage, a neural filter namely Filter Approximation network (FA-Net) was utilized to generate low-order time-domain filter coefficients to approximate the FBE subband-domain response. ...
arXiv:2202.06764v1
fatcat:qdee4t5a3nhdjgbgofrgcdbt4i
« Previous
Showing results 1 — 15 out of 166,563 results