Deep-plant: Plant identification with convolutional neural networks

By Sue Han Lee, Chee Seng Chan, Paul Wilkin, Paolo Remagnino. International Conference on Information Photonics, 2015

442 citations recorded by the source corpus.

This paper studies convolutional neural networks (CNN) to learn unsupervised feature representations for 44 different plant species, collected at the Royal Botanic Gardens, Kew, England. To gain intuition on the chosen features from the CNN model (opposed to a `black box' solution), a visualisation technique based on the deconvolutional networks (DN) is utilized. It is found that venations of different order have been chosen to uniquely represent each of the plant species. Experimental results using these CNN features with different classifiers show consistency and superiority compared to the state-of-the art solutions which rely on hand-crafted features.

Subfields: Ecological Modeling, Ecology, Plant Science

View Paper: https://www.semanticscholar.org/paper/0a5d6a04deae65337e8d16cd2806868e965feba0

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Ecological Modeling

1. Francesc X. Prenafeta-Boldu
2. Andreas Kamilaris
3. Marco Seeland
4. Michael Rzanny
5. Patrick Mäder
6. Jana Wäldchen
7. Maximilian Pichler
8. Florian Hartig

In Ecology

1. Francesc X. Prenafeta-Boldu
2. Andreas Kamilaris
3. Patrick Mäder
4. Jana Wäldchen
5. Marco Seeland
6. Michael Rzanny
7. Chee Seng Chan
8. Johan Potgieter

In Plant Science

1. Patrick Mäder
2. Jana Wäldchen
3. Marco Seeland
4. Michael Rzanny
5. Francesc X. Prenafeta-Boldu
6. Andreas Kamilaris
7. Johan Potgieter
8. Muhammad Hammad Saleem

Citations in the corpus

• How deep learning extracts and learns leaf features for plant classification (2017, methodology): “A preliminary version of this work was presented earlier [31].”
• Deep learning in agriculture: A survey (2018, background): “CNN had also superior performance than Penalized Discriminant Analysis (Grinblat, Uzal, Larese, & Granitto, 2016), SVM Regression (Kuwata & Shibasaki, 2015), area-based techniques (Rahnemoonfar & Sheppard, 2017), texturebased regression models (Chen, et al., 2017), LMC classifiers (Xinshao & Cheng, ”
• Automated plant species identification—Trends and future directions (2018, methodology): “’s [53, 54] leaf classifier that uses the AlexNet architecture pretrained on the ILSVRC2012 dataset and reached an average accuracy of 99.”
• SupportNet: solving catastrophic forgetting in class incremental learning with support data (2018, methodology): “…data of new classes, the previously well-trained deep learning model for enzyme function prediction (Li et al. , 2018a), plant species recognition (Lee et al. , 2015), and disease prediction (Mohanty et al. , 2016; Chen et al. , 2017) would face the serious problem of catastrophic forgetting.”
• A review of the use of convolutional neural networks in agriculture (2018, methodology): “The current survey examines the problems that employ a particular class of DL named convolutional neural networks (CNN), defined as deep, feed-forward ANN.”
• Plant Disease Detection and Classification by Deep Learning (2019, methodology): “[48,50] employed user-modified CNN while [49] used AlexNet architecture.”
• Deep Learning-Based Change Detection in Remote Sensing Images: A Review (2022, methodology): “To properly train systems, DL researchers employ a variety of strategies, including transfer learning [186–192], data augmentation, GAN [148,155], AE, and SSAE [193].”