<

This Article Statistics
Viewed : 1855 Downloaded : 1165


 

Analysis of Pooling Effect on CNN using Leaf Database

Yunus Camgözlü*, Yakup Kutlu

Abstract

Usage of artificial intelligence and machine learning is widespread in many areas such as information technology, driverless vehicles, health technology and marketing. The remarkable upward trend in studies on data science, machine learning, data visualization, artificial intelligence and deep learning is progressing even faster today. Between the studies of leaf classification in literature, it is seen that it uses feature extraction techniques such as various distance calculations according to the marked points, curvature-based shape feature and the use of different components of the image. The variety of feature extraction techniques is effects performance and models selection. Therefore, nowadays the models are used without features such as deep learning. In this study, leaf images have been classified using CNN model. Feature training, which is one of the many advantages of deep learning, enables to achieve results without using the above mentioned approach.CNN model was created in 2 stages. First step is feature learning contain process such as convolutional layer, nonlinearity layer, pooling, convolution and relu, pooling. Classification process used structures such as flatten, fully connected and softmax steps. Each species is regarded as a label and the classified by CNN model.

Keywords

Deep Learning, Leaf Classification

Volume 4, No 3, SUPPLEMENT I of SYMPOSIUM ARTICLES, pp 118-123, 2019



Download full text   |   How to Cite   |   Download XML Files

References
  • Altan, G., & Kutlu, Y. (2018). Superiorities of deep extreme learning machines against convolutional neural networks. Natural and Engineering Sciences, 3(3), 103-109.
  • Banerjee, B., & Murino, V. (2017). Efficient pooling of image based cnn features for action recognition in videos. 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).
  • Deng, L., Hamid, O., & Yu, D. (2013). A deep convolutional neural network using heterogeneous pooling for trading acoustic invariance with phonetic confusion. 2013 IEEE International Conference on Acoustics, Speech and Signal Processing. ISSN: 978-1-4799-0356-6
  • Fernando, B., Anderson P., Hutter, M., & Gould, S. (2016). Discriminative hierarchical rank pooling for activity recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
  • Kadir, A., Nugroho, L., Susanto, A., & Santosa, P. (2011). Leaf classification using shape, color and texture features. International Journal of Computer Trends and Technology, 1(3), 225-230.
  • Kumar, N., Belhumeur, P., Biswas, A., & Jacobs, D. (2012). Leafsnap: a computer vision system for automatic plant species identification. ECCV 2012: Computer Vision, 502-516.
  • Lawrence, S., Giles, L., Tsoi, A., & Back, A. (1997). Face recognition: a convolutional neural-network approach. IEEE Transactıons on Neural Networks, 8(1).
  • Lee, C., & Chen, S. (2006). Classification of leaf images. International Journal Of Imaging Systems and Technology, 16(1), 15-23.
  • Lee, C., Gallagher, P., & Tu, Z. (2015). Generalizing pooling functions in convolutional neural networks: mixed, gated and tree. arXiv: 1509.08985.
  • Liu, T., Fang, S., Zhao, Y., Wang, P., & Zhang, J., (2015). Implementation of training convolutional neural networks. arXiv: 1506.01195v2.
  • Wang, S., Lv, Y., Sui, Y., Liu, S., & Wang, S. (2018). Alcoholism detection by data augmentation and convolutional neural network with stochastic pooling. Journal of Medical Systems. ISSN: 1573-6689X.
  • Zhang, Y., & Wallace, B. (2016). A sensitivity analysis of (and practitioners’ guide to) convolutional neural networks for sentence classification. arXiv: 1510.03820v4