no code implementations • 21 Jun 2023 • Ziliang Hong, Emadeldeen Hamdan, Yifei Zhao, Tianxiao Ye, Hongyi Pan, A. Enis Cetin
This paper surveys different publicly available neural network models used for detecting wildfires using regular visible-range cameras which are placed on hilltops or forest lookout towers.
no code implementations • 25 May 2021 • Hongyi Pan, Diaa Badawi, Erdem Koyuncu, A. Enis Cetin
We consider a family of vector dot products that can be implemented using sign changes and addition operations only.
no code implementations • 12 Nov 2020 • Diaa Badawi, Agamyrat Agambayev, Sule Ozev, A. Enis Cetin
Our results show that we can have an accurate and smooth drift estimate even when the observed sensor signal is very noisy.
no code implementations • 14 May 2018 • T. Ceren Deveci, Serdar Cakir, A. Enis Cetin
Deep learning has made significant improvements at many image processing tasks in recent years, such as image classification, object recognition and object detection.
no code implementations • 9 Feb 2017 • Arman Afrasiyabi, Ozan Yildiz, Baris Nasir, Fatos T. Yarman Vural, A. Enis Cetin
This "product" is used to construct a vector product in $R^N$.
no code implementations • 16 Mar 2015 • Mohammad Tofighi, Onur Yorulmaz, A. Enis Cetin
Therefore, the set of images with a prescribed FT phase can be used as a constraint set in blind deconvolution problems.
no code implementations • 22 Oct 2014 • Osman Gunay, Cem Emre Akbas, A. Enis Cetin
If the gradient does not exist at a given vector, it is possible to use the subgradients and the normal producing the smallest angle between the two vectors is used to compute the similarity measure.
no code implementations • 9 Jul 2014 • Alican Bozkurt, Pinar Duygulu, A. Enis Cetin
Recognizing fonts has become an important task in document analysis, due to the increasing number of available digital documents in different fonts and emphases.
no code implementations • 10 Jun 2014 • A. Enis Cetin, Mohammad Tofighi
In sparsity based denoising methods, it is assumed that the original signal is sparse in some transform domains such as the wavelet domain and the wavelet subsignals of the noisy signal are projected onto L1-balls to reduce noise.
no code implementations • 10 Feb 2014 • Mohammad Tofighi, Kivanc Kose, A. Enis Cetin
A new signal processing framework based on making orthogonal Projections onto the Epigraph Set of a Convex cost function (PESC) is developed.