Extraction and recognition of shoe logos with a wide variety of appearance using two-stage classifiers

Kazunori Aoki, Wataru Ohyama, Tetsushi Wakabayashi

Research output: Contribution to journalArticle

Abstract

A logo is a symbolic presentation that is designed not only to identify a product manufacturer but also to attract the attention of shoppers. Shoe logos are a challenging subject for automatic extraction and recognition using image analysis techniques because they have characteristics that distinguish them from those of other products; that is, there is much within-class variation in the appearance of shoe logos. In this paper, we propose an automatic extraction and recognition method for shoe logos with a wide variety of appearance using a limited number of training samples. The proposed method employs maximally stable extremal regions for the initial region extraction, an iterative algorithm for region grouping, and gradient features and a support vector machine for logo recognition. The results of performance evaluation experiments using a logo dataset that consists of a wide variety of appearances show that the proposed method achieves promising performance for both logo extraction and recognition.

Original languageEnglish
Pages (from-to)1325-1332
Number of pages8
JournalIEICE Transactions on Information and Systems
VolumeE101D
Issue number5
DOIs
Publication statusPublished - May 2018

Fingerprint

Classifiers
Image analysis
Support vector machines
Experiments

All Science Journal Classification (ASJC) codes

  • Software
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering
  • Artificial Intelligence

Cite this

Extraction and recognition of shoe logos with a wide variety of appearance using two-stage classifiers. / Aoki, Kazunori; Ohyama, Wataru; Wakabayashi, Tetsushi.

In: IEICE Transactions on Information and Systems, Vol. E101D, No. 5, 05.2018, p. 1325-1332.

Research output: Contribution to journalArticle

Aoki, Kazunori ; Ohyama, Wataru ; Wakabayashi, Tetsushi. / Extraction and recognition of shoe logos with a wide variety of appearance using two-stage classifiers. In: IEICE Transactions on Information and Systems. 2018 ; Vol. E101D, No. 5. pp. 1325-1332.
@article{ec0eef2ee7b242569293072826d44526,
title = "Extraction and recognition of shoe logos with a wide variety of appearance using two-stage classifiers",
abstract = "A logo is a symbolic presentation that is designed not only to identify a product manufacturer but also to attract the attention of shoppers. Shoe logos are a challenging subject for automatic extraction and recognition using image analysis techniques because they have characteristics that distinguish them from those of other products; that is, there is much within-class variation in the appearance of shoe logos. In this paper, we propose an automatic extraction and recognition method for shoe logos with a wide variety of appearance using a limited number of training samples. The proposed method employs maximally stable extremal regions for the initial region extraction, an iterative algorithm for region grouping, and gradient features and a support vector machine for logo recognition. The results of performance evaluation experiments using a logo dataset that consists of a wide variety of appearances show that the proposed method achieves promising performance for both logo extraction and recognition.",
author = "Kazunori Aoki and Wataru Ohyama and Tetsushi Wakabayashi",
year = "2018",
month = "5",
doi = "10.1587/transinf.2017MVP0026",
language = "English",
volume = "E101D",
pages = "1325--1332",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "一般社団法人電子情報通信学会",
number = "5",

}

TY - JOUR

T1 - Extraction and recognition of shoe logos with a wide variety of appearance using two-stage classifiers

AU - Aoki, Kazunori

AU - Ohyama, Wataru

AU - Wakabayashi, Tetsushi

PY - 2018/5

Y1 - 2018/5

N2 - A logo is a symbolic presentation that is designed not only to identify a product manufacturer but also to attract the attention of shoppers. Shoe logos are a challenging subject for automatic extraction and recognition using image analysis techniques because they have characteristics that distinguish them from those of other products; that is, there is much within-class variation in the appearance of shoe logos. In this paper, we propose an automatic extraction and recognition method for shoe logos with a wide variety of appearance using a limited number of training samples. The proposed method employs maximally stable extremal regions for the initial region extraction, an iterative algorithm for region grouping, and gradient features and a support vector machine for logo recognition. The results of performance evaluation experiments using a logo dataset that consists of a wide variety of appearances show that the proposed method achieves promising performance for both logo extraction and recognition.

AB - A logo is a symbolic presentation that is designed not only to identify a product manufacturer but also to attract the attention of shoppers. Shoe logos are a challenging subject for automatic extraction and recognition using image analysis techniques because they have characteristics that distinguish them from those of other products; that is, there is much within-class variation in the appearance of shoe logos. In this paper, we propose an automatic extraction and recognition method for shoe logos with a wide variety of appearance using a limited number of training samples. The proposed method employs maximally stable extremal regions for the initial region extraction, an iterative algorithm for region grouping, and gradient features and a support vector machine for logo recognition. The results of performance evaluation experiments using a logo dataset that consists of a wide variety of appearances show that the proposed method achieves promising performance for both logo extraction and recognition.

UR - http://www.scopus.com/inward/record.url?scp=85046264740&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046264740&partnerID=8YFLogxK

U2 - 10.1587/transinf.2017MVP0026

DO - 10.1587/transinf.2017MVP0026

M3 - Article

AN - SCOPUS:85046264740

VL - E101D

SP - 1325

EP - 1332

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 5

ER -