On Combining Sequence Alignment and Feature-quantization for
Sub-image Searching

J 2012

On Combining Sequence Alignment and Feature-quantization for Sub-image Searching

HOMOLA, Tomáš, Vlastislav DOHNAL a Pavel ZEZULA

Základní údaje

Originální název

On Combining Sequence Alignment and Feature-quantization for Sub-image Searching

Autoři

HOMOLA, Tomáš (203 Česká republika, domácí), Vlastislav DOHNAL (203 Česká republika, garant, domácí) a Pavel ZEZULA (203 Česká republika, domácí)

Vydání

International Journal of Multimedia Data Engineering and Management (IJMDEM), Hershey PA 17033-1240, USA, IGI Global, 2012, 1947-8534

Další údaje

Jazyk

angličtina

Typ výsledku

Článek v odborném periodiku

Obor

10201 Computer sciences, information science, bioinformatics

Stát vydavatele

Spojené státy

Utajení

není předmětem státního či obchodního tajemství

Kód RIV

RIV/00216224:14330/12:00073206

Organizační jednotka

Fakulta informatiky

DOI

http://dx.doi.org/10.4018/jmdem.2012070102

UT WoS

000218974600002

Klíčová slova anglicky

image matching; sub-image retrieval; local image features; sequence alignment; performance evaluation

Štítky

DISA, RIV - zkontrolováno

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 1. 4. 2015 09:01, RNDr. Pavel Šmerk, Ph.D.

Anotace

V originále

The availability of various photo archives and photo sharing systems made similarity searching much more important because the photos are not usually conveniently tagged. So the photos (images) need to be searched by their content. Moreover, it is important not only to compare images with a query holistically but also to locate images that contain the query as their part. The query can be a picture of a person, building, or an abstract object and the task is to retrieve images of the query object but from a different perspective or images capturing a global scene containing the query object. This retrieval is called the sub-image searching. In this paper, we propose an algorithm, called SASISA, for retrieving database images by their similarity to and containment of a query. The novelty of it lies in application of a sequence alignment algorithm, which is commonly used in text retrieval. This forms an orthogonal solution to currently used approaches based on inverted files. We improve efficiency of SASISA by applying vector-quantization of local image feature descriptors. The proposed algorithm and its optimization are evaluated on a real-life data set containing photographs where images of logos are searched. It is compared to a state-of-the-art method (Joly & Buisson, 2009) and the improvement of 16% in mean average precision (mAP) is obtained.

Návaznosti

GBP103/12/G084, projekt VaV

Název: Centrum pro multi-modální interpretaci dat velkého rozsahu

Investor: Grantová agentura ČR, Centrum pro multi-modální interpretaci dat velkého rozsahu

VF20102014004, projekt VaV

Název: Multimediální analýza (Akronym: Multimediální analýza)

Investor: Ministerstvo vnitra ČR, Multimediální analýza

Přiložené soubory

IJMDEM2012.pdf

Požádat o autorskou verzi souboru

Citovat

HOMOLA, Tomáš, Vlastislav DOHNAL a Pavel ZEZULA. On Combining Sequence Alignment and Feature-quantization for Sub-image Searching. International Journal of Multimedia Data Engineering and Management (IJMDEM). Hershey PA 17033-1240, USA: IGI Global, 2012, roč. 3, č. 3, s. 20-44. ISSN 1947-8534. Dostupné z: https://dx.doi.org/10.4018/jmdem.2012070102.

@article{991044,
   author = {Homola, Tomáš and Dohnal, Vlastislav and Zezula, Pavel},
   article_location = {Hershey PA 17033-1240, USA},
   article_number = {3},
   doi = {http://dx.doi.org/10.4018/jmdem.2012070102},
   keywords = {image matching; sub-image retrieval; local image features; sequence alignment; performance evaluation},
   language = {eng},
   issn = {1947-8534},
   journal = {International Journal of Multimedia Data Engineering and Management (IJMDEM)},
   title = {On Combining Sequence Alignment and Feature-quantization for Sub-image Searching},
   volume = {3},
   year = {2012}
}

TY  - JOUR
ID  - 991044
AU  - Homola, Tomáš - Dohnal, Vlastislav - Zezula, Pavel
PY  - 2012
TI  - On Combining Sequence Alignment and Feature-quantization for Sub-image Searching
JF  - International Journal of Multimedia Data Engineering and Management (IJMDEM)
VL  - 3
IS  - 3
SP  - 20-44
EP  - 20-44
PB  - IGI Global
SN  - 19478534
KW  - image matching
KW  - sub-image retrieval
KW  - local image features
KW  - sequence alignment
KW  - performance evaluation
N2  - The availability of various photo archives and photo sharing systems made similarity searching much more important because the photos are not usually conveniently tagged. So the photos (images) need to be searched by their content. Moreover, it is important not only to compare images with a query holistically but also to locate images that contain the query as their part. The query can be a picture of a person, building, or an abstract object and the task is to retrieve images of the query object but from a different perspective or images capturing a global scene containing the query object. This retrieval is called the sub-image searching. In this paper, we propose an algorithm, called SASISA, for retrieving database images by their similarity to and containment of a query. The novelty of it lies in application of a sequence alignment algorithm, which is commonly used in text retrieval. This forms an orthogonal solution to currently used approaches based on inverted files. We improve efficiency of SASISA by applying vector-quantization of local image feature descriptors. The proposed algorithm and its optimization are evaluated on a real-life data set containing photographs where images of logos are searched. It is compared to a state-of-the-art method (Joly & Buisson, 2009) and the improvement of 16% in mean average precision (mAP) is obtained.
ER  -

HOMOLA, Tomáš, Vlastislav DOHNAL a Pavel ZEZULA. On Combining Sequence Alignment and Feature-quantization for Sub-image Searching. \textit{International Journal of Multimedia Data Engineering and Management (IJMDEM)}. Hershey PA 17033-1240, USA: IGI Global, 2012, roč.~3, č.~3, s.~20-44. ISSN~1947-8534. Dostupné z: https://dx.doi.org/10.4018/jmdem.2012070102.

Podrobný výpis o publikaci