DEEP LEARNING FOR APPLE AND PEAR RECOGNITION
DOI:
https://doi.org/10.17770/het2021.25.6791Keywords:
AlexNet, apple, CNN, Fruits360, Food2030, neural network, pearAbstract
The aim of this work is to develop a neural network, which can recognize apples and pears. To achieve the goal, the authors applied AlexNet architecture and the open dataset “Fruits360”. The trained model showed a good result testing it on validation images - total accuracy 0.97 and latency 35ms/step. In the future research, authors consider training the neural network model using the MobileNet architecture and verify it using the Cohen`s Kappa coefficient.Downloads
References
Konvolūcijas neironu tīkli attēlu atpazīšanai [tiešsaiste], [atsauce uz 01.03.2021]. Pieejams: https://habr.com/ru/post/456186/
Transfer learning [tiešsaiste], [atsauce uz 01.03.2021]. Pieejams: https://machinelearningmastery.com/transfer-learning-for-deep-learning/
Fruits360 datu kopa [tiešsaiste], [atsauce uz 01.03.2021]. Pieejams: https://www.kaggle.com/moltean/fruits
TensorFlow [tiešsaiste], [atsauce uz 01.03.2021]. Pieejams: https://www.tensorflow.org/about
Anaconda rīks [tiešsaiste], [atsauce uz 01.03.2021]. Pieejams: https://www.anaconda.com/why-anaconda/
CIFAR datu kopa [tiešsaiste], [atsauce uz 01.03.2021]. Pieejams: https://www.cs.toronto.edu/~kriz/cifar.html
Datu augmentācija [tiešsaiste], [atsauce uz 03.03.2021]. Pieejams: https://algorithmia.com/blog/introduction-to-dataset-augmentation-and-expansion
Datu augmentācija [tiešsaiste], [atsauce uz 07.03.2021]. Pieejams:https://medium.com/analytics-vidhya/data-augmentation-is-it-really-necessary-b3cb12ab3c3f
