LULC-NEAT: Land Use Land Cover Classification Using Neuroevolutionary of Augmenting Topologies

Sumayyea Salahuddin; Nasru Minallah; Muhammad Athar Sethi; Muhammad Ajmal; Maryam Mahsal Khan

Authors

Sumayyea Salahuddin Department of Computer Systems Engineering, University of Engineering and Technology Peshawar, 25000, Pakistan.
Nasru Minallah Department of Computer Systems Engineering, University of Engineering and Technology Peshawar, 25000, Pakistan.
Muhammad Athar Sethi Department of Computer Systems Engineering, University of Engineering and Technology Peshawar, 25000, Pakistan.
Muhammad Ajmal Department of Agricultural Engineering, University of Engineering and Technology Peshawar, 25000, Pakistan.
Maryam Mahsal Khan Department of Computer Science, CECOS University of IT and Emerging Sciences Peshawar, 25000, Pakistan.

Keywords:

Satellite Image Classification, Neuroevolutionary of Augmenting Topologies (NEAT), Deep Learning, Convolutional Neural Network (CNN), EuroSAT

Abstract

Introduction/Importance of Study: In this paper, a novel application of NeuroEvolution of Augmenting Topologies for Land Use Land Cover Classification, which remains a perennial activity in environmental monitoring and management, is considered.

Novelty statement: We introduce NEAT for evolving feed-forward neural networks (FFNNs) tailored for LULC classification, offering a unique solution that addresses the challenge of optimal neural network architecture design.

Material and Method: The EuroSAT RGB benchmark satellite dataset was preprocessed using Numpy, Keras, and TensorFlow, and then evaluated using the NEAT algorithm to create diverse FFNNs with varying hidden layers.

Result and Discussion: The NEAT-evolved FFNN architecture with two hidden layers showed excellent and high accuracy percentages during the training and testing, respectively. Although high training accuracy implies successful feature learning, it also indicates probable overfitting. However, the high accuracy obtained in testing, 99.83%, shows the excellent generalization ability of the model toward unseen data and thus does not overfit. The results were cross-validated with the state-of-the-art CNN models, and the experiments prove that NEAT can be effectively used for LULC classification.

Concluding Remarks: The study confirms that NEAT can effectively evolve neural networks for high-accuracy LULC classification, offering a robust alternative to traditional CNN models.

References

S. Talukdar et al., “Land-Use Land-Cover Classification by Machine Learning Classifiers for Satellite Observations—A Review,” Remote Sens. 2020, Vol. 12, Page 1135, vol. 12, no. 7, p. 1135, Apr. 2020, doi: 10.3390/RS12071135.

A. M. Abdi, “Land cover and land use classification performance of machine learning algorithms in a boreal landscape using Sentinel-2 data,” GIScience Remote Sens., vol. 57, no. 1, pp. 1–20, Jan. 2020, doi: 10.1080/15481603.2019.1650447.

K. Kaku, “Satellite remote sensing for disaster management support: A holistic and staged approach based on case studies in Sentinel Asia,” Int. J. Disaster Risk Reduct., vol. 33, pp. 417–432, Feb. 2019, doi: 10.1016/J.IJDRR.2018.09.015.

“5 crimes solved using Google Earth | The Week.” Accessed: Jun. 30, 2024. [Online]. Available: https://theweek.com/articles/491975/5-crimes-solved-using-google-earth

M. Majeed et al., “Monitoring of Land Use–Land Cover Change and Potential Causal Factors of Climate Change in Jhelum District, Punjab, Pakistan, through GIS and Multi-Temporal Satellite Data,” L. 2021, Vol. 10, Page 1026, vol. 10, no. 10, p. 1026, Sep. 2021, doi: 10.3390/LAND10101026.

Y. Lecun, Y. Bengio, and G. Hinton, “Deep learning,” Nat. 2015 5217553, vol. 521, no. 7553, pp. 436–444, May 2015, doi: 10.1038/nature14539.

J. Schmidhuber, “Deep learning in neural networks: An overview,” Neural Networks, vol. 61, pp. 85–117, Jan. 2015, doi: 10.1016/J.NEUNET.2014.09.003.

Y. Bengio, “Learning Deep Architectures for AI,” Found. Trends® Mach. Learn., vol. 2, no. 1, pp. 1–127, Nov. 2009, doi: 10.1561/2200000006.

D. Floreano, P. Dürr, and C. Mattiussi, “Neuroevolution: From architectures to learning,” Evol. Intell., vol. 1, no. 1, pp. 47–62, Jan. 2008, doi: 10.1007/S12065-007-0002-4/METRICS.

K. O. Stanley and R. Miikkulainen, “Evolving Neural Networks Through Augmenting Topologies.” pp. 99–127. Accessed: Jun. 23, 2024. [Online]. Available: http://mitpress.mit.edu/e-mail

“GitHub - phelber/EuroSAT: EuroSAT: Land Use and Land Cover Classification with Sentinel-2.” Accessed: Jun. 23, 2024. [Online]. Available: https://github.com/phelber/EuroSAT

P. Helber, B. Bischke, A. Dengel, and D. Borth, “EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification,” IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., vol. 12, no. 7, pp. 2217–2226, Aug. 2017, doi: 10.1109/JSTARS.2019.2918242.

P. Helber, B. Bischke, A. Dengel, and D. Borth, “Introducing eurosat: A novel dataset and deep learning benchmark for land use and land cover classification,” Int. Geosci. Remote Sens. Symp., vol. 2018-July, pp. 204–207, Oct. 2018, doi: 10.1109/IGARSS.2018.8519248.

R. Naushad, T. Kaur, and E. Ghaderpour, “Deep Transfer Learning for Land Use and Land Cover Classification: A Comparative Study,” Sensors 2021, Vol. 21, Page 8083, vol. 21, no. 23, p. 8083, Dec. 2021, doi: 10.3390/S21238083.

A. Loganathan, S. Koushmitha, and Y. N. K. Arun, “Land Use/Land Cover Classification Using Machine Learning and Deep Learning Algorithms for EuroSAT Dataset – A Review,” Lect. Notes Networks Syst., vol. 418 LNNS, pp. 1363–1374, 2022, doi: 10.1007/978-3-030-96308-8_126.

“Welcome to NEAT-Python’s documentation! — NEAT-Python 0.92 documentation.” Accessed: Jun. 30, 2024. [Online]. Available: https://neat-python.readthedocs.io/en/latest/

J. Li et al., “Deep Discriminative Representation Learning with Attention Map for Scene Classification,” Remote Sens. 2020, Vol. 12, Page 1366, vol. 12, no. 9, p. 1366, Apr. 2020, doi: 10.3390/RS12091366.

G. Chen et al., “Training Small Networks for Scene Classification of Remote Sensing Images via Knowledge Distillation,” Remote Sens. 2018, Vol. 10, Page 719, vol. 10, no. 5, p. 719, May 2018, doi: 10.3390/RS10050719.

“Land Cover Classification with EuroSAT Dataset.” Accessed: Jun. 30, 2024. [Online]. Available: https://www.kaggle.com/code/nilesh789/land-cover-classification-with-eurosat-dataset

P. Jain, B. Schoen-Phelan, and R. Ross, “Self-Supervised Learning for Invariant Representations from Multi-Spectral and SAR Images,” IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., vol. 15, pp. 7797–7808, May 2022, doi: 10.1109/JSTARS.2022.3204888.

A. Stateczny, S. M. Bolugallu, P. B. Divakarachari, K. Ganesan, and J. R. Muthu, “Multiplicative Long Short-Term Memory with Improved Mayfly Optimization for LULC Classification,” Remote Sens. 2022, Vol. 14, Page 4837, vol. 14, no. 19, p. 4837, Sep. 2022, doi: 10.3390/RS14194837.

M. Ç. Aksoy, B. Sirmacek, and C. Ünsalan, “Land classification in satellite images by injecting traditional features to CNN models,” Remote Sens. Lett., vol. 14, no. 2, pp. 157–167, Feb. 2023, doi: 10.1080/2150704X.2023.2167057.

A. Rangel, J. Terven, D. M. Cordova-Esparza, and E. A. Chavez-Urbiola, “Land Cover Image Classification,” Jan. 2024, Accessed: Jun. 23, 2024. [Online]. Available: http://arxiv.org/abs/2401.09607

D. Yadav, K. Kapoor, A. K. Yadav, M. Kumar, A. Jain, and J. Morato, “Satellite image classification using deep learning approach,” Earth Sci. Informatics, vol. 17, no. 3, pp. 2495–2508, Jun. 2024, doi: 10.1007/S12145-024-01301-X/METRICS.

“Land Cover Classification with EuroSAT Dataset.” Accessed: Jun. 23, 2024. [Online]. Available: https://www.kaggle.com/code/nilesh789/land-cover-classification-with-eurosat-dataset

D. Phiri, M. Simwanda, S. Salekin, V. R. Nyirenda, Y. Murayama, and M. Ranagalage, “Sentinel-2 Data for Land Cover/Use Mapping: A Review,” Remote Sens. 2020, Vol. 12, Page 2291, vol. 12, no. 14, p. 2291, Jul. 2020, doi: 10.3390/RS12142291.

“S2 Mission.” Accessed: Jun. 30, 2024. [Online]. Available: https://sentiwiki.copernicus.eu/web/s2-mission

“Artificial Neural Network (ANN) with Practical Implementation | by Amir Ali | The Art of Data Scicne | Medium.” Accessed: Jun. 23, 2024. [Online]. Available: https://medium.com/machine-learning-researcher/artificial-neural-network-ann-4481fa33d85a

O. I. Abiodun, A. Jantan, A. E. Omolara, K. V. Dada, N. A. E. Mohamed, and H. Arshad, “State-of-the-art in artificial neural network applications: A survey,” Heliyon, vol. 4, no. 11, p. e00938, Nov. 2018, doi: 10.1016/J.HELIYON.2018.E00938.

“Neural networks and back-propagation explained in a simple way | by Assaad MOAWAD | DataThings | Medium.” Accessed: Jun. 23, 2024. [Online]. Available: https://medium.com/datathings/neural-networks-and-backpropagation-explained-in-a-simple-way-f540a3611f5e

S. Ruder, “An overview of gradient descent optimization algorithms,” Sep. 2016, Accessed: Jun. 23, 2024. [Online]. Available: http://arxiv.org/abs/1609.04747

M. G. M. Abdolrasol et al., “Artificial Neural Networks Based Optimization Techniques: A Review,” Electron. 2021, Vol. 10, Page 2689, vol. 10, no. 21, p. 2689, Nov. 2021, doi: 10.3390/ELECTRONICS10212689.

J. N. D. Gupta and R. S. Sexton, “Comparing backpropagation with a genetic algorithm for neural network training,” Omega, vol. 27, no. 6, pp. 679–684, 1999, doi: 10.1016/S0305-0483(99)00027-4.

“Genetic algorithms”, [Online]. Available: https://www.jstor.org/stable/24939139

“Introduction to Genetic Algorithms — Including Example Code | by Vijini Mallawaarachchi | Towards Data Science.” Accessed: Jun. 30, 2024. [Online]. Available: https://towardsdatascience.com/introduction-to-genetic-algorithms-including-example-code-e396e98d8bf3

D. Floreano, P. Dürr, and C. Mattiussi, “Neuroevolution: From architectures to learning,” Evol. Intell., vol. 1, no. 1, pp. 47–62, Jan. 2008, doi: 10.1007/S12065-007-0002-4/METRICS.

K. O. Stanley and R. Miikkulainen, “Efficient Evolution of Neural Network Topologies”.

K. O. Stanley, “Efficient Evolution of Neural Networks Through Complexification.” 2004. Accessed: Jun. 30, 2024. [Online]. Available: http://www.cs.utexas.edu/users/kstanley/

“How I Built an Intelligent Agent to Play Flappy Bird | by Danny Zhu | Analytics Vidhya | Medium.” Accessed: Jun. 23, 2024. [Online]. Available: https://medium.com/analytics-vidhya/how-i-built-an-ai-to-play-flappy-bird-81b672b66521

K. O. Stanley and R. Miikkulainen, “Evolving Neural Networks through Augmenting Topologies,” Evol. Comput., vol. 10, no. 2, pp. 99–127, Jun. 2002, doi: 10.1162/106365602320169811.

K. O. Stanley and R. Miikkulainen, “Efficient evolution of neural network topologies,” Proc. 2002 Congr. Evol. Comput. CEC 2002, vol. 2, pp. 1757–1762, 2002, doi: 10.1109/CEC.2002.1004508.

Y. Bengio, “Practical Recommendations for Gradient-Based Training of Deep Architectures,” Lect. Notes Comput. Sci. (including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics), vol. 7700 LECTURE NO, pp. 437–478, 2012, doi: 10.1007/978-3-642-35289-8_26.

“Simple Guide to Hyperparameter Tuning in Neural Networks | by Matthew Stewart, PhD | Towards Data Science.” Accessed: Jun. 23, 2024. [Online]. Available: https://towardsdatascience.com/simple-guide-to-hyperparameter-tuning-in-neural-networks-3fe03dad8594

L. Alzubaidi et al., “Review of deep learning: concepts, CNN architectures, challenges, applications, future directions,” J. Big Data 2021 81, vol. 8, no. 1, pp. 1–74, Mar. 2021, doi: 10.1186/S40537-021-00444-8.

A. Vali, S. Comai, and M. Matteucci, “Deep Learning for Land Use and Land Cover Classification Based on Hyperspectral and Multispectral Earth Observation Data: A Review,” Remote Sens. 2020, Vol. 12, Page 2495, vol. 12, no. 15, p. 2495, Aug. 2020, doi: 10.3390/RS12152495.

“StratifiedShuffleSplit — scikit-learn 1.5.0 documentation.” Accessed: Jun. 30, 2024. [Online]. Available: https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.StratifiedShuffleSplit.html

“Built-in Functions — Python 3.12.4 documentation.” Accessed: Jun. 30, 2024. [Online]. Available: https://docs.python.org/3/library/functions.html#next

“Image Data Generators in Keras. How to effectively and efficiently use… | by Manpreet Singh Minhas | Towards Data Science.” Accessed: Jun. 30, 2024. [Online]. Available: https://towardsdatascience.com/image-data-generators-in-keras-7c5fc6928400

A. M. Fred Agarap, “Deep Learning using Rectified Linear Units (ReLU),” Mar. 2018, Accessed: Jun. 23, 2024. [Online]. Available: https://arxiv.org/abs/1803.08375v2

P. Guilherme B. A, D. Fernanda B. J. R, A. Fazenda, and F. A. Faria, “Neuroevolution-based Classifiers for Deforestation Detection in Tropical Forests,” Proc. - 2022 35th Conf. Graph. Patterns, Images, SIBGRAPI 2022, pp. 13–18, 2022, doi: 10.1109/SIBGRAPI55357.2022.9991798.