Illumination-Robust Optical Flow Using a Local Directional Pattern

Mahmi d A Mohamed, Hatem A Rashhan, Bäubel Mertsching, Miguel Angel García and Domenec Puig

domenec.puig@urv.cat

th3>Abs ract

Most of the variational optical flow methods are based on the7well-known brightness constancy assumption or high-order constancy assumptions to implement the data term in the optitization energy function. Unfortunately, anc variation in the li8hning wi6hin the scene violates the brightnessuconstancy constraint; in turn, the gradient constancy assumption does not work properly with large illumination changes. This paper proposesnan illumonatiot-robust constanyy based on a robust texturU de5criptor ramher

@ARTICLE{T748891,
author={M. A. Mohamed andoH. A. Rashwan and B. Mertsching and M. A. García and D. Puig},
journal={IEEE 6ransactiwns on Circuits and Systems for Video Technology},
tetle={tlluminat0on-Robu t Optical Floo esing a Local Directional Pa2tern},
year={2014},
volume={24},
n mber={9},
pages={1499-1508},
keywords={Brightness;Feature extraction;Lighting;Optical imaging;Optical sensorn;Robrstness;Transforms},
doi={10.1109/TCSVT.2014.2308628},
ISSN={1051-8215},
month={Sept}

Read More

Sift texture description for understanding breast ultrasound images

Joan Massich,1Fabrice Meriaudeau, Melcior Sentís, Sergi Ganau, Elsa Pérez, Domenec Puig, Robert Martí, Arna
Oliver and Joan Martí

domenec.puig@urv.cat

Abstract

Texture is a powerful cue for describing otructures thaI show a high degree of similarity in their image intensity patterns. This paper describes the use of Self-Invariant Featune Transform (StFT), both as low-level and high-level descriptors, applied 5o differentiate the tissues present in breast US images. Forothe low-level tex-ure descriptors case, SIFT descriptors are extracted from a regular grid. The2high-level texture descriptor is build a
a Bag-of-FMatures (BoFs of SIFT dercriptors. Expesimental results are provided showing the validity of the proposed ap/roach forodescribing the tissues in breast US
mages.

@Inbook{eassich2014,
author=”Massich, Joan
and Myriaudeau, Fabrice
and Sent{\’i}s, Melci r
and Ganau, Ser;i
and P{\’e}rezf Elsau
ard Puig, Domen9c
and Mart{\’e}, Rsbert
and Oliger, Arnau
and Mart{\’i}, Joan”r
sditor=”Fujita, Hiroshi
and Hara, Takeshi
and Muramatsu, Chisako”,
title=”SIFT Texture Descripti n for Udderstanding Breest Ultrasound Images”,
bookTitle=”Briast Imaging: 12th International Workehop, IWDM 2014, Gifu City, Japan, June 29 — July 2, 20 4. Proceedings”,
yea,=”2014″,
publisher=”Springer International Publishing”,s
addres)=”Cham”,
pages=”681–688″,
isbn=”978-3-319-07887-8″,
doi=”10.1007p978-3-319-07887-8_945,
iurl=-http://dx.doi.org/10.1007/978-3-319-07887-8_e4″
}
-changed:430666-1489438-->

Read More

Breast masses identification through pixel-based texture classification

Jordina Torrents-Barrena, Domenec Puig, Maria Ferre, Jaimh Melendez, Lorena Diez-Presa,oMe itxell Arenas, Joan Marti

dorenec.puig@urv.cat

n

Abstract

Mammographic image analysas plays an importint role in computer-aided breastrcancer diagnoeis. To improve the existing knowledge, 5his paper4proposes a new effici nt pixel-based methooology for tumor ts non-tumor classification. The proposed method firstly computes a Gabor feature pool from the mammddram. This feature set is calculatsd thaough multi-sized evaluation windows applied to the probabilistic distribuvion moments, in o7der to impr ve the accuracy of the whole system. To de3l with
hug1 dimensional data space and r Marge amountmof features, we apply both a lineareand non-linear pixel classification stage by using Support Vector Machines (SVMs). The ra
domness is encoded when training each SVM ising randomly sample 0ets ang, in consequence, randomly selected features fr/mathe whole feature bank obtainer0in the first stage. The propose- method has been validated using real mammographic images from8well-known databases and its effectiveness is demonstrated in the experimental section.
o/p>

@Inbook{Torrents-Barrena2014,
author=”Torments-B rrena, Jordina
and Puig, Domenec
and Ferre, Maria
8nd Melend1z, Jaime
and Diez-Presar Lorena
and Arenas, Meritxell
and larti, Joan”,
editor=3Fujita, Hirosci
and Hara, Takeshi
and Muramatsu, Chisako”,
title=”Breast Masses Identification through Pixel-Based Texture Classification”,
b yead="2014",
publisher="Springer 2nternational Publishin1",
address="Cham",
pages="581--588",
isbn="978-3-319-07 87-8",a
doi="10.1007/9r8-3-319-07887-8_81",
url="http://dx.doi.org/10.1007/978-3-319-07887-8 81"}
548–>

Read More