Skip to content
AI Journal of BUsiness AI Journal of BUsiness AI Journal of BUsiness AI Journal of BUsiness
Home

Article

Study on the Effect of Data Augmentation on Image Learning

Tae-hwan Choi Primary Contact
PDF
Abstract

This paper aims to demonstrate the effective use of data in AI research by examining the impact of data augmentation on AI learning. This study utilized a subset of the AI Hub’s Ocean Deposition Garbage Image dataset. The accuracy of image classification was measured using CNN’s VGG16 and VGG19. After increasing the data volume through augmentation, the accuracy was re-measured and re-compared. The results of the study indicated that accuracy improved in most cases, leading to the conclusion that data augmentation positively affects image learning.

This paper is revision of the author’s master degree thesis. and This research Ocean Deposition Garbage Image used datasets from 'The Open AI Dataset Project (AI-Hub, S. Korea)'. All data information can be accessed through 'AI-Hub (www.aihub.or.kr)'.

References
  1. In-Kyeong Hwang. (2017), Study on Hangul font characteristics using CNN, The Graduate School Seoul National University.
  2. Ian Goodfellow., Yoshua Bengio., Aaron Courville. (2016), “Deep Learning”, MIT Press, https://www.deeplearningbook.org/.
  3. Su-Bin Park. (2019), Implementation of Multiple Object Detection and Recognition System using H-SSD based on VGG Network, Graduate School Don-A University.
  4. Hyeon-Ho Lee. (2020), CNN Generalization Error Evaluation Method, The Graduate School Pusan National University.
  5. Kang-Hyun Park. (2021), Displavement prediction of structures using Signal Image data Based CNN(SIBCNN), The Graduate School Yonsei University.
  6. Gyeong-Won Jang. (2023), Exploring Optimization Techniques for CNN in Binary Image Classification, The Graduate Seoul National University of Science and Technology.
  7. Kwang-Bok Hwang. (2021), A Study on the Design of Steering Controller for Autonomous Vehicle with Variable Speed Using CNN, Graduate School Gyeongnam National University of Science and Technology.
  8. Ae-Ri Min. (2020), A Study on Transfer Learning for Image Classification of Convolutional Neural Network – Based on VGG16 Deep Convolutional Neural Network, The Graduate School Hansei University.
  9. Jong-Bum Kim, Sung-Kwun Oh. (2015), Design of Convolutional RBFNNs Pattern Classifier for Two dimensional Face Recognition, The Korean Institute of Electrical Engineers, 1355-1356.
  10. Hyun-Su Lee. (2018), A Structure of Convolutional Neural Networks for Image Contents Search, Graduate School of Chung-Ang University.
  11. Han-Nah Kim. (2020), Sentiment Analysis and Classification using Convolutional Neural Networks, Graduate School of Soonchunhyang University.
  12. Karen Simonyan., Andrew Zisserman. (2014), “Very Deep Convolutional Networs for Large-Scale Image Recognition”, arXiv:1409.1556.
  13. Alex Krizhevsky., Ilya Sutskever., Geoffrey E. Hinton. (2012), “ImageNet Classification with Deep Convolutional Neural Networks”, NIPS, 1106-1114.
  14. Liang Han Sheng. (2021), “VGGNet – Convolutional Network for clssification and Detection”, https://medium.com/analytics-vidhya/vggnet-convolutional-network-for-classification-and-detection-3543aaf61699.
  15. Jun-Yeol Ryu, Tae-Wan Kim, Ja-Hoon Jeong. (2022), An AI Model for classifying The State of Armored and Mechanized Weapon Systems Based on VGG16, Journal of the Korea Academia-Industrial cooperation Society, 23. 12. 741-748.
  16. Hye-won Kim, Xuefeng Zhang, Yong-Soo Kim, Il-Hong Jung. (2022), Comparison of the Performance of CNN Models for Retinal Diseases Diagnosis, Journal of Korean Institute of Intelligent Systems, 32. 1. 51-60.
  17. Jeong-Hwan Lee. (2021), Detection of Number and Character Area of License Plate Using Deep Learning and Semantic Image Segmentation, Journal of the Korea Convergence Society, 12. 1. 29-35.
  18. Ji-Hoon Jo. (2013), An Implementation of a smart camera system using CUDA Based parallel CAM shift Algorithm, Hannam University.
Keywords
AI Data Augmentation Image CNN VGG
Details
Section Articles
Issue Vol. 1 No. 1 (2024): Volume 1 Issue 1 (Inaugural Issue)
Published 2024-11-15
Pages 26-37
Authors
Tae-hwan Choi Primary Contact
A Seoul School of Integrated Sciences & Technologies(aSSIST)

  • A Seoul School of Integrated Sciences & Technologies(aSSIST) M.S
  • Areas of interest : AI, Data Analysis, Cloud, Computer Vision, Natural Language Processing, Data Mining etc.
License
How to Cite
Study on the Effect of Data Augmentation on Image Learning. (2024). AI Journal of BUsiness, 1(1), 26-37. https://jnl.ampla.page/aijb/article/view/111