Comparison of VGG and MobileNet Models for Grass Seed Dataset

ERYİĞİT R., TUĞRUL B.

5th International Conference on Informatics and Computational Sciences (ICICoS), Aachen, Almanya, 24 - 25 Ekim 2021 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/icicos53627.2021.9651865
  • Basıldığı Şehir: Aachen
  • Basıldığı Ülke: Almanya
  • Anahtar Kelimeler: Deep Learning, Seed Classification, MobileNet, VGG, WEED SEEDS, IDENTIFICATION
  • Ankara Üniversitesi Adresli: Evet

Özet

Convolutional neural networks (CNN) have transformed the computer vision research area with tremendous success over the traditional machine learning approaches in the last decade. Here, we report the results of an investigation of seed classification problem by using two widely used CNNs, mobile centric MobileNet and a parameter rich VGG19. We have found that the classification accuracy for both nets strongly depends on the resolution of the images used in the training.