| bibtype |
J -
Journal Article
|
| ARLID |
0041079 |
| utime |
20240103182728.1 |
| mtime |
20060912235959.9 |
| title
(primary) (eng) |
Building Road-Sign Classifiers Using a Trainable Similarity Measure |
| specification |
|
| serial |
| ARLID |
cav_un_epca*0258150 |
| ISSN |
1524-9050 |
| title
|
IEEE Transactions on Intelligent Transportation Systems |
| volume_id |
7 |
| volume |
3 (2006) |
| page_num |
309-321 |
|
| title
(cze) |
Klasifikace dopravních značek založená na míře podobnosti zkoumaného objektu k třídě reprezentované typickou značkou |
| keyword |
classifier system design |
| keyword |
road-sign classification |
| keyword |
similarity data representation |
| author
(primary) |
| ARLID |
cav_un_auth*0212668 |
| name1 |
Paclík |
| name2 |
P. |
| country |
NL |
|
| author
|
| ARLID |
cav_un_auth*0101171 |
| name1 |
Novovičová |
| name2 |
Jana |
| institution |
UTIA-B |
| fullinstit |
Ústav teorie informace a automatizace AV ČR, v. v. i. |
|
| author
|
| ARLID |
cav_un_auth*0216548 |
| name1 |
Duin |
| name2 |
R.P.W. |
| country |
NL |
|
| source |
|
| COSATI |
09K |
| COSATI |
12B |
| COSATI |
09J |
| cas_special |
| project |
| project_id |
IAA2075302 |
| agency |
GA AV ČR |
| ARLID |
cav_un_auth*0001801 |
|
| project |
| project_id |
507752 |
| country |
XE |
| agency |
EC |
| ARLID |
cav_un_auth*0200689 |
|
| research |
CEZ:AV0Z10750506 |
| abstract
(eng) |
A frequently used strategy for road sign classification is based on the normalized cross-correlation similarity to class prototypes followed by the nearest neighbor classifier. Because of the global nature of the cross-correlation similarity, this method suffers from presence of uninformative pixels (caused e.g. by occlusions), and is computationally demanding. In this paper, a novel concept of a trainable similarity measure is introduced which alleviates these shortcomings. The similarity is based on individual matches in a set of local image regions. The set of regions, relevant for a particular similarity assessment, is refined by the training process. It is illustrated on a set of experiments with road sign classification problems that the trainable similarity yields high-performance data representations and classifiers. Apart from a multi-class classification accuracy, also non-sign rejection capability, and computational demands in execution are discussed. It appears that the trainable similarity representation alleviates some difficulties of other algorithms, currently used in road sign classification. |
| abstract
(cze) |
Návrh klasifikátoru dopravních značek založeného na podobnosti zkoumaného objektu ke třídě značek reprezentované vždy typickou značkou (prototype-based rule). Navržen algoritmus založený na trénování podle množiny prototypů (trainable similarity). Experimenty na několika datových souborech ilustrují vyšší účinnost klasifikátoru v porovnání s klasifikátory až dosud používanými pro klasifikaci značek. Byly testovány i další vlastnosti navrženého klasifikátoru jako robustnost a časová náročnost. |
| reportyear |
2007 |
| RIV |
BB |
| permalink |
http://hdl.handle.net/11104/0134664 |
| arlyear |
2006 |
| mrcbU63 |
cav_un_epca*0258150 IEEE Transactions on Intelligent Transportation Systems 1524-9050 1558-0016 Roč. 7 č. 3 2006 309 321 |
|