Presentation 2019-03-08
Examination of estimation accuracy of deep learning by data augmentation
Masashi Kasamatsu, Yukikazu Murakami, Miyoshi Kengo,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In agriculture, contract cultivation can make price decisions by producers.For contract cultivation, it is necessary to accurately predict harvest date and crop yields.However, to use a neural network, a huge amount of data is required.The farm work diary corresponding to that data is mainly recorded manually by farmers.Therefore, there is a problem that the data of the agricultural industry as a whole is small.In this paper, we implemented a conventional harvest date prediction system using NeuralNetworkConsole (NNC), NeuralNetworkLibraries (NNL) and Python.Then we extended the data to improve the prediction accuracy of the harvest date.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Agriculture / Neural Network / Harvest Prediction System / Data Augmentation
Paper # LOIS2018-75
Date of Issue 2019-02-28 (LOIS)

Conference Information
Committee LOIS
Conference Date 2019/3/7(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Miyakojima-shi Central Community Center
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Tomohiro Yamada(NTT)
Vice Chair Toru Kobayashi(Nagasaki Univ.)
Secretary Toru Kobayashi(NTT)
Assistant Shinichiro Eitoku(NTT)

Paper Information
Registration To Technical Committee on Life Intelligence and Office Information Systems
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Examination of estimation accuracy of deep learning by data augmentation
Sub Title (in English)
Keyword(1) Agriculture
Keyword(2) Neural Network
Keyword(3) Harvest Prediction System
Keyword(4) Data Augmentation
1st Author's Name Masashi Kasamatsu
1st Author's Affiliation National Institute of Technology, Kagawa College(NIT, Kagawa)
2nd Author's Name Yukikazu Murakami
2nd Author's Affiliation National Institute of Technology, Kagawa College(NIT, Kagawa)
3rd Author's Name Miyoshi Kengo
3rd Author's Affiliation National Institute of Technology, Kagawa College(NIT, Kagawa)
Date 2019-03-08
Paper # LOIS2018-75
Volume (vol) vol.118
Number (no) LOIS-485
Page pp.pp.115-119(LOIS),
#Pages 5
Date of Issue 2019-02-28 (LOIS)