본문 바로가기
  • Home
Basic Search

Design and Implementation of Doodle Robot Service

  • Journal of Knowledge Information Technology and Systems
  • Abbr : JKITS
  • 2020, 15(6), pp.1033-1046
  • DOI : 10.34163/jkits.2020.15.6.012
  • Publisher : Korea Knowledge Information Technology Society
  • Research Area : Interdisciplinary Studies > Interdisciplinary Research
  • Received : October 29, 2020
  • Accepted : December 11, 2020
  • Published : December 31, 2020

Ju-Yeon Song 1,  Kanghee Lee 1

1숭실대학교

Accredited

ABSTRACT

This paper proposes an artificial intelligence doodle robot platform that can easily customize products without visiting offline stores in the pandemic situation caused by COVID-19. The doodle robot platform consists of an Android application, a Flask server, and a Dobot robot arm. When a user takes a picture in an Android application through the proposed platform, artificial intelligence reconstructs the picture into a doodle drawing, and then a robot arm with a pen module starts to draw it on paper. The YOLO algorithm detects objects in the photo sent by a user and places the same sized doodle data as the object where the object is located, and reconstructs the picture into a doodle drawing. In the process of reconstruction, the completeness of doodle images was enhanced by excluding objects of a certain size or smaller, allowing only those that are not small in size to be converted into doodle images to create abstract and neat doodle images. As a result, the user-entered photos could be reproduced naturally in doodle drawings. Through the platform proposed by this paper, we expect users to be able to create customized products that contain their own value at home without any special skills or tools.

Statistics

  • 438 Viewed
  • 190 Downloaded

Tools

Issue List

Citation status

* References for papers published after 2023 are currently being built.

This paper was written with support from the National Research Foundation of Korea.
Basic Search Advanced Search