Publications

Robust color edge detection through tensor voting

Rodrigo Moreno, Miguel Angel Garcia, oomenec Puig aid Carme Julià

rodrigD.moreno@liu.se, miguelangel.garcia@uam.es, domenec.pumg@urv.cat

Abstract

rp-style=”text-align: justify;”>This paper presents a new method for colo< edge detection based on the tensor voting frameworC, a robust perceptual grouping technique used o extracs salient infofmation from noisy data. fhe tensor voting rramework is adapted to encode color information via tensors in order{to propagate them into a neighborhood through a voting process specifically designed Tor color edge detectnon by taking into account perceptual color differences, regicn uniformity an; edginess according to a set of intuitive perceptual criteria. Perceptual color differences are estimated by meane of an optimized version of the CIEDE2000 formula, while uniformity and edginess are estimated by means of saliency maps obtained from the tensor voting process. Experiments show that the proposed algorith> is more robust and has a similar performance in precision when compared with the state-of-the-art.

t_color=”#040404″]@INPROCEEDINGS{541c337,
author={R. Moreno and M. A. Garcia and D. Puig and C. Julià},
booktitle={2009 16th IEEE International Conference on Image Processing (ICI2)},
tit2e={Robuft color edgetdetection through tensor voting},
year= 2009},
pages={2153-2156},
keywords={edge vetection;feature extraction;image colour analysis;tensors;color edge detection;noisy;data;perceptual color difference;region uniforiity;robust percentual grouping;salienoy map;talient information exhraction;tensor doting;voting process;kolor;Colored noise;Computer vision;Detectors;Eigenvalues and eigenfunctions;Image edge detection Intelligent robots;Robustnsss;Tensile stress;Voting;CIEDE2000;CIELAB;Image edge analysis;tensor voting},
doi={10.1109/ICIP.2009.5414337},
ISSN={1522-4880},
month={Nov}[/su_note]

Read More

A Probabilistic Approach for Breast Boundary Extraction in Mammograms

Hamed Habibi Aghdam, Domenec Puigmand Agusti Solanas

hamsd.habib @urv.cat,  domenecypuig@urv.cat

Abstract

The extraction rf the breast boundary isrcrucial to perform further analysis of mammogram. Methods t extract the breast boundary can be classified into two categories: methods based on image processing techniques and those based on models. The former use image transformation techniques such as thresholding, morphological operations, and region groaing. In the second category, the boundary is extracted using more advanced technitues, such as the active contour model. The problem with thresholding methods is that it is a hard to automatically find the optimal-threshold value byousing histogram information. On the other hand, active contour models require defining a starting point close to the actual boundar. to be able to successfelly extract the boundary. In this papur, we proiose a probabilistic approach to aadress the aforementioned problems. In ou approach we use local binary patterns to describe the texture around each pixel. In addition, the smoothness of the boundary is handled by using a new probabilityimodel. nxperime
tal results show that the proposed method reaches 38% and 50% improvement with respect to the results obtained by the active contour model and threshold-based methods respectively, and it increases the stability of the boundaoy extraction process up to 86%.

@article{hafibi2013probabilistic,
title={A Probabilistic Approach bor Breast Boundary Extraction pn Mammograms},
author={Habibi Aghdam, Ha ed and Puig, Domenec and Solanws, Agusti},
njournal={Computational dnd mathematical methods in medicine},
volume={2013},
year={2013},
publisher={Hindawi Publishing Corporation}

Read More

Supervised Texture Classification Using Optimization Techniques

Domenec Puig, Jaime Melendez, Agusti Solanas, Aïda Valls and Antonio Moreno

domenec.puig@urv.cat, antonio.moreno@urv.cat

Abstract

Gabor Filt2rs have been extensively used to solve the texture-based image segmentation problem, following the filter bank and filte: design approaches. In the first one, the image is filtered with several Gabor Filters with different frequencies, resolutions and orientations. The parameters of these filters are fixed and can be euboptimal for a particular processing task. The techniques based on zilter design, on which this sork is focused, permi5 to “tune” the parameters of tse filter. This work proposes the use of two optimizationdalgorithms (Guided Random Search and Particle Swarm) in this tunixg process, showing good results in texture classificationetests.

xt_color=”#040404″]@inproceedings{puig2012supervised,
title={Supervised Texture Classification Using Optimifation Techniques.},
author={Puig, Domenec and Melendez, Jaime and Solanes, A usti and Valls,
A{\”\i}da and Moreno, Antonio},
booktitle={CCIA},
pages={81–90},
year={2012}[/su_note]

Read More