Landscape photo classification mechanism for context-aware photography support system

Nuttapoom Amornpashara, Yutaka Arakawa, Morihiko Tamai, Keiichi Yasumoto

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

This paper presents landscape photo classification mechanisms to enhance our proposed photography support system named Phorec. The system utilizes big data in social photo stock services and online historical weather database to recommend relevant photos based on user's natural contexts, including location, time, and weather condition. In order to help user take better photos, important information, such as camera setting, time, and location, is displayed with the photo selected by the user. Since the aim of Phorec is to support not only photographers but also general users as a travel assistant, we mainly focus on the landscape photos. From the huge collected data, landscape photos classification is required to extract the landscape photos from the uncategorized photos. In this paper, some possible landscape classification techniques are investigated and evaluated with our collected dataset. In addition to the straightforward tag analysis and Exif analysis method, we propose a unique method that utilizes a common face recognition technique for excluding non-landscape photos. Experimental results show that our proposed people-exclusion method works well for excluding non-landscape photos from the collected various photo.

Original languageEnglish
Title of host publication2015 IEEE International Conference on Consumer Electronics, ICCE 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages663-666
Number of pages4
ISBN (Electronic)9781479975426
DOIs
Publication statusPublished - Mar 23 2015
Externally publishedYes
Event2015 IEEE International Conference on Consumer Electronics, ICCE 2015 - Las Vegas, United States
Duration: Jan 9 2015Jan 12 2015

Publication series

Name2015 IEEE International Conference on Consumer Electronics, ICCE 2015

Conference

Conference2015 IEEE International Conference on Consumer Electronics, ICCE 2015
CountryUnited States
CityLas Vegas
Period1/9/151/12/15

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Electrical and Electronic Engineering
  • Industrial and Manufacturing Engineering

Fingerprint Dive into the research topics of 'Landscape photo classification mechanism for context-aware photography support system'. Together they form a unique fingerprint.

Cite this