¹Tokushima University, Tokushima, Japan.
²Aomori University, Aomori, Japan.

---

Abstract— In recent years, a lot of non-verbal expressions have been used on social media.Ascii art (AA) is an expression using characters withvisual technique. In this paper, we set up an experiment to classify AA pictures by using character features and image features.We try to clarifywhich feature is more effective for amethod to classify AA pictures.We proposed fivemethods:1) a method based on character frequency, 2) a method based on character importance value and 3) a method based on image features, 4) a method based on image features using pre-trained neural networksand5) a method based on image features of characters.We trained neural networks by using these fivefeatures.Inthe experimental result, the best classification accuracy was obtained inthe feed forward neural networks that usedimage features of characters.

Index Terms— ASCII art, deep neural networks, classification, image feature, character feature

[PDF]

Cite: Kazuyuki Matsumoto, Akira Fujisawa, Minoru Yoshida, Kenji Kita, "ASCII Art Classification based on Deep Neural Networks Using Image Feature of Characters," Journal of Software vol. 13, no. 10, pp. 559-572, 2018.