Automated recognition by multiple convolutional neural networks of modern, fossil, intact and damaged pollen grains.
Benjamin BourelRoss MarchantThibault de Garidel-ThoronMartin TetardDoris BarboniYves GallyLuc BeaufortPublished in: Comput. Geosci. (2020)
Keyphrases
- convolutional neural networks
- recognition rate
- convolutional network
- semi automated
- action recognition
- objects in cluttered scenes
- automatic recognition
- image recognition
- recognition accuracy
- object recognition
- fully automated
- combining multiple
- feature extraction
- case study
- unsupervised learning
- human computer interaction
- expert systems
- pattern recognition
- multiple objects
- learning algorithm
- neural network
- database