Presentation 2019-03-05
Detecting the Number of Compartments Using Convolutional Neural Network
Takayuki Koga, Yuta Takahashi, Shuntaro Okazaki, Rieko Osu,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) n this study, we examined the compartment counting ability of Convolutional Neural Network (CNN) using images that are controlled for their features (i.e. brightness, shape, and the number of lines that divided compartments). We trained the CNN by images with limited features and tested its generalizability by images that are different from training data in the shape combination and shape itself. Consequently, the CNN achieved approximately 56% accuracy in the generalization test. This result indicated that the CNN learned to count the division of compartments without identifying the shape of the compartments. Our findings suggested that although the acquired counting ability in the CNN is still imperfect it may not be so far from the concept of number in human.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Neural Network / Concept Acquisition / Counting
Paper # NC2018-74
Date of Issue 2019-02-25 (NC)

Conference Information
Committee NC / MBE
Conference Date 2019/3/4(3days)
Place (in Japanese) (See Japanese page)
Place (in English) University of Electro Communications
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Yutaka Hirata(Chubu Univ.) / Masaki Kyoso(TCU)
Vice Chair Hayaru Shouno(UEC) / Taishin Nomura(Osaka Univ.)
Secretary Hayaru Shouno(Nagoya Univ.) / Taishin Nomura(NAIST)
Assistant Keiichiro Inagaki(Chubu Univ.) / Takashi Shinozaki(NICT) / Takumi Kobayashi(YNU) / Yasuyuki Suzuki(Osaka Univ.)

Paper Information
Registration To Technical Committee on Neurocomputing / Technical Committee on ME and Bio Cybernetics
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Detecting the Number of Compartments Using Convolutional Neural Network
Sub Title (in English)
Keyword(1) Neural Network
Keyword(2) Concept Acquisition
Keyword(3) Counting
1st Author's Name Takayuki Koga
1st Author's Affiliation Waseda University(Waseda Univ.)
2nd Author's Name Yuta Takahashi
2nd Author's Affiliation Waseda University(Waseda Univ.)
3rd Author's Name Shuntaro Okazaki
3rd Author's Affiliation Waseda University(Waseda Univ.)
4th Author's Name Rieko Osu
4th Author's Affiliation Waseda University(Waseda Univ.)
Date 2019-03-05
Paper # NC2018-74
Volume (vol) vol.118
Number (no) NC-470
Page pp.pp.163-168(NC),
#Pages 6
Date of Issue 2019-02-25 (NC)