21 Hits in 9.8 sec

Object Recognition Based on Dual Tree Complex Wavelet Transform

S. Elakkiya, S. Audithan
2014 Research Journal of Applied Sciences Engineering and Technology  
In this study, an approach for object recognition based on Dual Tree Complex Wavelet Transform (DTCWT) is proposed.  ...  The proposed approach attempts to extract the detailed information of objects from the multi scale representation by DTCWT. The proposed system is tested on Columbia Object Image Library (COIL-100).  ...  METHODOLOGY The proposed approach for object recognition is built based on Dual Tree Complex Wavelet Transform (DTCWT).  ... 
doi:10.19026/rjaset.7.842 fatcat:s6eauzloyff3vldfy4cyde6z74

Texture feature extraction in the spatial-frequency domain for content-based image retrieval [article]

Nadia Baaziz, Omar Abahmane, Rokia Missaoui
2010 arXiv   pre-print
Gabor wavelets, dual-tree complex wavelet and contourlets.  ...  In particular, it gives an overview of statistical methodologies and techniques employed for texture feature extraction using most popular spatial-frequency image transforms, namely discrete wavelets,  ...  Complex Wavelets for Texture Retrieval The Dual-Tree Complex Wavelet Transform The dual-tree complex wavelet transform (DT-CWT) as introduced by Kingsbury [34] has been found to be a useful tool for  ... 
arXiv:1012.5208v1 fatcat:ew5orkvrx5gvbjpym4e7d2oyne

Wavelet-based Scale Saliency [article]

Anh Cat Le Ngo, Kenneth Li-Minn Ang, Jasmine Kah-Phooi Seng, Guoping Qiu
2013 arXiv   pre-print
The paper discusses different wavelet descriptors: discrete wavelet transform (DWT), wavelet packet transform (DWPT), quaternion wavelet transform (QWT) and best basis quaternion wavelet packet transform  ...  While PSS uses pixel-value descriptors, WSS treats wavelet sub-bands as basis descriptors.  ...  The shift-variance of real-wavelet transform can be avoided by complex wavelet transform de- sign; for instances, recently developed dual-tree complex wavelet transform (DTCWT) [35], Quaternion wavelet  ... 
arXiv:1301.2884v1 fatcat:akled7atjfatxdjrxgrs76wtna

Texture Retrieval via the Scattering Transform [article]

Alexander Sagel, Dominik Meyer, Hao Shen
2015 arXiv   pre-print
the Scattering transform subbands, due to the complex modulus performed on each one of them.  ...  Moreover, we adopt a common approach of measuring the similarity of textures by comparing the subband histograms of a filterbank transform.  ...  Additionally, we implemented a method based the Dual-Tree Complex Wavelet Transform (DT-CWT), inspired by [15] . The DT-CWT was performed by the DT-CWT Transform Pack 3 , version 4.3.  ... 
arXiv:1501.02655v4 fatcat:3qst4glhafgwdpi2nuucpuvfpy

Data Streams: Algorithms and Applications

S. Muthukrishnan
2005 Foundations and Trends® in Theoretical Computer Science  
Some of the methods rely on metric embeddings, pseudo-random computations, sparse approximation theory and communication complexity.  ...  Algorithms have to work with one or few passes over the data, space less than linear in the input size or time significantly less than the input size.  ...  Acknowledgements My sincere thanks to the referees for correcting the mistakes without curbing my style. Also, thanks to the many coauthors and coresearchers who have led the way.  ... 
doi:10.1561/0400000002 fatcat:ozoe572tfnappg2zcpgtzdeasm

Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges [article]

Michael M. Bronstein, Joan Bruna, Taco Cohen, Petar Veličković
2021 arXiv   pre-print
Remarkably, the essence of deep learning is built from two simple algorithmic principles: first, the notion of representation or feature learning, whereby adapted, often hierarchical, features capture  ...  The last decade has witnessed an experimental revolution in data science and machine learning, epitomised by deep learning methods.  ...  In a way, we hope to have presented "all you need to build the architectures that are all you need"-a play on words inspired by Vaswani et al. (2017) .  ... 
arXiv:2104.13478v2 fatcat:odbzfsau6bbwbhulc233cfsrom

Signature Methods in Machine Learning [article]

Terry Lyons, Andrew D. McLeod
2022 arXiv   pre-print
The examples presented in this article are intended to bridge this communication gap and provide tractable working examples drawn from the machine learning context.  ...  Signature-based techniques give mathematical insight into the interactions between complex streams of evolving data.  ...  A truncated signature is extracted from each dyadic path • (c) The collection of the signatures of each dyadic path is transformed to the dyadic path-tree signature representation • (d) Tree-based convolution  ... 
arXiv:2206.14674v1 fatcat:3zbhnbkeojh4tmbqooolccsevq

Real-time embedded age and gender classification in unconstrained video

Ramin Azarmehr, Robert Laganiere, Won-Sook Lee, Christina Xu, Daniel Laroche
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
Also, we introduce several novel improvements such as face alignment using the nose, and an illumination normalization method for unconstrained environments using bilateral filtering.  ...  On a multi-resolution feature vector we have achieved up to 99.5% compression ratio for training data storage, and a maximum performance of 20 frames per second on an embedded Android platform.  ...  Ullah et al. [123] tried LBP and Dyadic Wavelet Transform (DyWT) which is a multi-scale image transformation technique for gender recognition.  ... 
doi:10.1109/cvprw.2015.7301367 dblp:conf/cvpr/AzarmehrLLXL15 fatcat:5bxiwufnr5e4hfwssmomuqxr5a

GPGPU-based Gaussian Filtering for Surface Metrological Data Processing

Yang Su, Zhijie Xu, Xiangqian Jiang
2008 2008 12th International Conference Information Visualisation  
The algorithms employed in this thesis included various filtering techniques, discrete wavelet transform, and the fast Fourier Transform which cover the common operations implemented in most LTI systems  ...  Thirdly, this thesis devised methods for carrying out result visualization directly on GPU by storing processed data in local GPU memory through making use of GPU's rendering device features to achieve  ...  The discrete wavelet transform (DWT) uses a specific subset of scale and translation values where the chosen scale and translation are based on the powers of two, which is the so-called dyadic scales and  ... 
doi:10.1109/iv.2008.14 dblp:conf/iv/SuXJ08 fatcat:lpagxjxstjbj5lcdumztlpgolu

Directional multiresolution image representations

Minh N. Do
Recently, it has become evident that commonly used separable transforms (such as wavelets) are not necessarily best suited for images.  ...  The first method constructs multiresolution, local and directional image expansions by using non-separable filter banks.  ...  invariant.  ... 
doi:10.5075/epfl-thesis-2500 fatcat:75jtyyvb55f3je6fkbkd4epcci

29th Annual Computational Neuroscience Meeting: CNS*2020

2020 BMC Neuroscience  
Hence, feedback from downstream visual cortical areas to V1 for better decoding (recognition), through analysis-by-synthesis, should query for additional information and be mainly directed at the foveal  ...  It suggests a massive loss of non-selected information from V1 downstream along the visual pathway.  ...  Acknowledgments: The research reported here was supported, in part, by NSF grant #1544383 and in part by AFOSR grant #FA9550-16-1-0410.  ... 
doi:10.1186/s12868-020-00593-1 pmid:33342424 fatcat:edosycf35zfifm552a2aogis7a

Nonlinear and stochastic dynamics in the heart

Zhilin Qu, Gang Hu, Alan Garfinkel, James N. Weiss
2014 Physics reports  
The transition from a normal rhythm to an arrhythmia is a transition from regular electrical wave conduction to irregular or turbulent wave conduction in the heart, and thus this medical problem is also  ...  dynamics from the molecular (microscopic) scale, to the organelle (mesoscopic) scale, to the cellular, tissue, and organ (macroscopic) scales.  ...  We are grateful for grant support from National Institutes of Health and the American Heart Association (to A.G., J.W., and Z.Q.), National Natural Science Foundation of China (to G.H.), and Laubisch and  ... 
doi:10.1016/j.physrep.2014.05.002 pmid:25267872 pmcid:PMC4175480 fatcat:iqsua6tjunhknglo7yupo73re4

New Developments in Science and Technology - 60th Anniversary Special Issue, Beihang University (BUAA)

2012 Science  
There are 75 full-time teachers in the school, which annually graduates 150 undergraduates, 180 postgraduates, 50 Ph.D.s, and 10 students from abroad.  ...  processing community with the development of the wavelet transform (WT).  ...  Step 1 established the state-space models of the scaling and wavelet coefficients of the DSCS by using the Haar WT.  ... 
doi:10.1126/science.338.6104.274-c fatcat:iurk5neuwvh2jgbohwevb5dicy

Data Fusion for Multi-Sensor Nondestructive Detection of Surface Cracks in Ferromagnetic Materials [article]

René Heideklang, Humboldt-Universität Zu Berlin, Humboldt-Universität Zu Berlin
However, the theoretical advantages of such directional transforms are not attained in practice with the given data.  ...  Data fusion reduces the rate of false positives is reduced by a factor of six over the best individual sensor at a 10 μm deep groove.  ...  The Dual-Tree Complex Wavelet Transform (DTCoWT) [115] achieves higher directional selectivity by implementing a complex-valued filter bank, which is able to separate positive from negative frequencies  ... 
doi:10.18452/19586 fatcat:i34kxpiqi5fpbfi7ffrzujjrt4

Dagstuhl Reports, Volume 9, Issue 12, December 2019, Complete Issue

For example, by transforming the game levels to a tile-based graph. The associated actions in these abstracted worlds are high-level as well.  ...  CIG tree, different from the sub-trees in which the search started.  ...  License Creative Commons BY 3.0 Unported license © Cindy Marie Grimm Robotic manipulation is a complex combination of four different elements: 1) The kinematics of the manipulator, 2) The sensors, 3) The  ... 
doi:10.4230/dagrep.9.12 fatcat:hebigxkvinhjdb6qlg3j5hw25u
« Previous Showing results 1 — 15 out of 21 results