%0 Conference Proceedings
%T Word Recognition by Combining Outline Emphasis and Synthesize Background
%+ Chubu University
%+ Rakuten Institute of Technology
%A Achiha, Yukihiro
%A Yamashita, Takayoshi
%A Nakazawa, Mitsuru
%A Masuko, Soh
%A Yamauchi, Yuji
%A Fujiyoshi, Hironobu
%Z Part 9: Workshops and Tutorials
%< avec comité de lecture
%( Lecture Notes in Computer Science
%B 16th International Conference on Entertainment Computing (ICEC)
%C Tsukuba City, Japan
%Y Nagisa Munekata
%Y Itsuki Kunita
%Y Junichi Hoshino
%I Springer International Publishing
%3 Entertainment Computing – ICEC 2017
%V LNCS-10507
%P 492-496
%8 2017-09-18
%D 2017
%R 10.1007/978-3-319-66715-7_70
%K Character recognition
%K Synthesis image
%K CNN
%Z Computer Science [cs]Conference papers
%X Character recognition collects item keywords from images from e-commerce websites; however, it requires a huge amount of training data. In this paper, we propose an efficient method to collect the training data by generating synthesis images and emphasizing outlines to obtain realistic images. The proposed method improves recognition accuracy on both generated images and real images from e-commerce websites.
%G English
%Z TC 14
%2 https://inria.hal.science/hal-01771238/document
%2 https://inria.hal.science/hal-01771238/file/978-3-319-66715-7_70_Chapter.pdf
%L hal-01771238
%U https://inria.hal.science/hal-01771238
%~ IFIP-LNCS
%~ IFIP
%~ IFIP-ICEC
%~ IFIP-TC14
%~ IFIP-LNCS-10507