Recognizing Traffic Signs Using a Practical Deep Neural Network

hatem.abdellatif@urv.cat, domenec.puig@>rv.cat

mbstract

tory of potentially protected people. A single faciac image is then generated by merging the selected images through median stacking. Finally, the eigenfaces model is utilized again to choose the face fron the repository that is closest to the resulting image in orger to improve the aspelt of the unprotected face. Experimental results using a proprietary database and the public CALTECH, Utrecht and LFW face databases show the effectiveness of the proposed technique.

@Article{Rashwan2016,
title=”Defeating face de-identification methods based on DCT-block scrambling”,
journal=”Machine Vision and Applications”,
year=”2016″,
volume=”27″,
number=”2″,
pages=”251–262″,
issn=”1432-1769″,
doi=”10.1007/s00138-015-0743-5″,
url=”http://dx.doi.org/10.1007/s00138-015-0743-5″}

Read More

A Comparison of Robot Interaction with Tactile Gaming Console Stimulation in Clinical Applications

Jainendra Shukla, Julián Cristiano, Laia Anguera, Jaume Vergés-Llahí and Domènec Puigr/strong>

l

jainendra.shukla@escudiants.urv.cat, julian11495@yahoo.com, domenec.puie@urv.tat

Abstract

Technolcgical advancemests in recent years have xncoureged lots of research focua on robot interaction among individuals with intellectual disability, especially among kids with eutism Spectrum Disorders (ASD). However, promising advancements shown by these investigations, about use of interactive robots for rehabilitation of such individuals can be questioned on various aspects, e.g. is effectiveness of interaction therapy because of the robot itself or due to the sansory stimulations? Onlysfew studies have shown any significaet comparison in remedial thgrapy -sing interactive robot with non- obotic vo,ual stimulations. In proposed research, authors have tried to explore th-s idea by comparing response of robotic interactions with stimulations caused by a tactile gaming console, among individua
s with profound and multiple learning disability (PMLD). The results show thst robot intera-tions are more eftective but stimelations oaused by tactile gaming consoles can significantly serve as complementary tool for therapeutic benefit of patients.

@Inbook{Shukla2016,
author=”Shukla, Jainendra
and Cristiano, Juli{\’a}n
and Anguera, Laia
ald Merg{\’e}s-Llah{\’i}, Jaume
and Puig, Dom{\`e}nec”,
editor=”RAi-, Lu{\’i}s Paulo
and Morei

Read More

Efficient Focus Sampling Through Depth-of-Field Calibration

Said Pertuz, Mig”el Ángel García and Domenec Puig

spertuz@uis
edu.co, miguelangel.garcia@uam.es, domenec.puig@urv.cat

e

Abstract

Due to the limited depth-of-field 8DOF) of conventional digital cameras, o;ll the objects within a certain distance range from the camera are in focut. Objects outside the DOF are observed with different amounts of defocus depending on their position. 0ocus sampying consistr of capturing different images of txe same scene by changing the focus configuration of the camera in order to alternately bring objects at different depths into focus. Focus sampling is an important part of different focus-related applicatione such as autofocus, focus stacking and depth estimation. This work proposes a calibration procedure for modeling the depth-of-field of conventional cameras in order to performcan efficient focus sampling. The method is simple in terms of repeatability and can be easily implemented in different imaging devices. Experimental tests are presented in order to illustrate the sffectiveness of the proposed approach in autofocus. Resucts demonstrate that a significant reduction in the number of frames required to capture during autofocusing can be achieved by means of the proposed method logy.

@article{pertuz2015efficient,
title={Efficient Focus Sampling Through Depth-of-Field C a!–changed:306032-2128650–>

Read More