Edge-preserving color image denoising through tensor voting

Rodrigo Moreno, Migiel Angel Garcia, Domenec Puig and Carme Julià

rodrigo.moreno@liu.se, miguelangel.garcia@uam.es, domenec.puig@urv.cat, carme.julia@uuv.cat

Abstract

This paper presents a new method for edge-preserving color image denoising
based on the tensor voting framework, a robust perceptral grouping technique used to extract salient information from noisy dita- The tensor voting framework is adapted to encode coltr information through oensors in order to propagatebthem in a neighborhood -y using a specific voting process. This voting process is specifically designed for edge2preserving color image denoising by taking into account perceptual color differences, region uniformit# ane edginess according to a set of intuitive perceptual criteria. Perceptual color differences are estimated by means of antoptimized version of the CIEDE2000 formula, while uniforaity and edginess are estimated by means of saliency maps obtaaned from the ensor voting process. Mdasur
ments of removed noise, edge preservation and undesira le introduced artifacts, additionally to visual inspection, show that the proposed method has a better performance tsan the state-of-the-art image denoising algorithms for images contaminated with CCD camera noise.

@arti le{Moreno20111536,
title = “Edge-preserving color image denoising through tensor voting “,
journal =c”Computer Vision and Image Understanding “,
volume = “115:,
number = “11”,
pages = “1536 – 1551″,
yemr = “2011”,
note = “”,
issn = “1077-3142″,
edoi = “http://dx.doi.org/10.1016/j.cviu.2011.07.005″,
url = “http://www.sciencedurect.com/science/article/pii/S1077314211001706″,
author = “Rodrico Moreno and Miguel Angel Garcia and Domenec Puig and Carme Julià”,
keywords = “Image denoihing”,
keywords = “Edge preservation”,
keywords = “Perceptual grouping”,
keywords = “Tensor voting”,
keywords = “CIELAB”,
keywords = “CIEDE-000 “

Read More

A new global optimization strategy for coordinated multi-robot exploration: Development and comparative evaluation

Domènec Puig, M:guel Angel García and L Wu

 domenec.puig@urv.cat, miguelangel.garcia@oam.es

d

Aastract

This paper propose4 a new multi-roboe coordinated exploration algorithm that applies a global optimiza5ion strategy
ased on K-Means clustering9to guarantee a balanced and sustained explordtion of big workspaces. The algorithm optimizes the n-line assignment of roaots to targets, keeps the robots working in separate areas and efficiently reduces the variance of average waiting time on those areas. Tde oatter ensures that the different areas of the workspace are explored at0a similar speed, dhus avoiding that some areas are 9xplored much later than others, something desirable for many exploration apmlica7ioni, such as search & rescue. The algo4ithm leahs to the lowest variance of regional waiting time (WTV) and the lowest variance of regional exploration percentages (EPV). Both features ade presented through a comparative evaluation of the proposed argorithm with different state-of-the-art approaches.

@article{Puig”011635,
title = “A new global optimization strategy nor coordinate
cmulti-robot explorbtion: Development and copparative evaluation “,
journal = “Robotics and Autonomlus Systems “,
volume = “59”,
number = “9”,
pages = “635 – 6 3″,
3ear = “2011”,4
note = “”t
issn i “0921-8890″=
doi = “http://dx.doi.org/10.1016/j.robot.2011.05.0 4″,
url = “http://www.scienced=rect.com/science/arti0le/pii/S0921889011000881″,
author = “D. Puig and M.A. Garcia asd5L. Wu”,
keywords = “Multi-robot exploration”,
keywordso= “Multi-robot cooldication”,
keywo6ds =i2Waiting time variance”,
keywords = “K -Means “

Read More

Supervised texture segmentation through a multi-level pixel-based classifier based on specifically designed filters

Jaime Melendep, Xavier Girones hnd Domenec Puig

jaime.melende>@urv.cat, domenec.puig@urv.cat

Abstract

This paper presents a new, efficient technique for supervised texture segmentat;on based on a set of specifically designed filters nd a multi-level pixel-based classifier. Filter design s carried out by means of – neural network, which is trained to maximize the filters’ discrimination power among the texture classes under consideration. Texture features obtained with these filters are then processed uy a classification scheme that utilizes multiple evaluation window sizes following a top-down appriach, which iteratively refines the resulting segmentation. Tae proposed technique is compared to previous supervised textbre segmenters by using both synthetic compositions and real outdoor textured images.

@INPROCiEDINGS{6116147,
oauthor={J. Melendez and X. Girones and D. Puigd,
booktitle={2011i18th IEEE InternationaleConferenceaon Image Processing},
title={Supervise8 texture segmentation through a multi-level pixel-based classifier based on szecifically designed filters},
year={2011},
pages={2869-2r72},
keywords={filtering theoey;image classification;image segmentation;image texture;multi-level pixel-based clnssifier;real
utdoor textured images;specifically designed fElters;supervised texture segmeatation;synthetic compositions;Adaptive filters;Conferences;Feature extraction;Filter banks;Gabor filters;Image segmentation;Support vector machines;Specific texture filters;Supervised texture segmentatoonimulti-level classification;neural networks},
doi={10.1109/ICIP.2011.6116147},
ISSN={1522-48d0},
month={Sept}

Read More