Abstract

Precise land use land cover (LULC) classification plays a crucial role in urban and regional planning. Policymakers ought to have an in-depth understanding of the LULC of the landscape. Classical remote sensing methods of LULC classification are time-consuming and expensive. In recent years, deep learning (DL) has shown remarkable performances in remote sensing applications especially like LULC classification. In spite of its promising result, generalization ability of DL models, optimization of hyperparameters, and accuracy with limited datasets are major concerns of DL algorithms. In this study, an OptTransEnsembleNet DL framework which integrates deep transfer learning, deep ensemble learning, and optimization is proposed. Initially transfer learning approach is employed to train the 4 base learners on EuroSAT spectral indices derived dataset which is generated from benchmark EuroSAT dataset. The pretrained models InceptionV3, MobileNetV2, Xception, and InceptionResNetV2 are used as feature extractors and further custom layers are added. The hyperparameters of these layers of base learners are optimized with the grid search optimization algorithm. Next, the aforementioned 4 models are used as base learners in deep ensemble learning and a weighted average ensemble is created for LULC. The optimal weights for the weighted average ensemble are found using the grid search algorithm. Overall, OptTransEnsembleNet-ensemble of 4 base learners produce an overall accuracy (OA) of 96.40% which is higher in comparison to individual models such as MobileNetV2 (92.15%), Xception (93.52%), InceptionV3 (91.50%) and InceptionResNetV2 (93.80%). Likewise, the model is applied on Jagatsinghpur district in Odisha state of India for classification followed by generation of its LULC map. The proposed model helps to improve the generalization performance and robustness of a DL model. This study could be beneficial for urban planners and administrators for resource management.

Keyword

Deep learning, Remote sensing, Land use land cover classification, Ensemble learning, Optimization.

Cite this article

Bathe KD, Patil NS.OptTransEnsembleNet: deep ensemble learning framework for land use land cover classification. International Journal of Advanced Technology and Engineering Exploration. 2024;11(120):1546-1561. DOI:10.19101/IJATEE.2024.111100036

Refference

[1]Wang SW, Gebru BM, Lamchin M, Kayastha RB, Lee WK. Land use and land cover change detection and prediction in the Kathmandu district of Nepal using remote sensing and GIS. Sustainability. 2020; 12(9):1-18.

[2]Langat PK, Kumar L, Koech R, Ghosh MK. Monitoring of land use/land-cover dynamics using remote sensing: a case of Tana River basin, Kenya. Geocarto International. 2021; 36(13):1470-88.

[3]Chen Z, Wang J. Land use and land cover change detection using satellite remote sensing techniques in the mountainous three gorges area, China. International Journal of Remote Sensing. 2010; 31(6):1519-42.

[4]Li W, Dong R, Fu H, Wang J, Yu L, Gong P. Integrating google earth imagery with landsat data to improve 30-m resolution land cover mapping. Remote Sensing of Environment. 2020; 237:111563.

[5]Bathe KD, Patil NS. Assessment of land use-land cover dynamics and its future projection through Google earth engine, machine learning and QGIS-MOLUSCE: a case study in Jagatsinghpur district, Odisha, India. Journal of Earth System Science. 2024; 133(2):111.

[6]Amalisana B, Hernina R. Land cover analysis by using pixel-based and object-based image classification method in Bogor. In conference series: earth and environmental science 2017 (pp. 1-7). IOP Publishing.

[7]Šandera J, Štych P. Selecting relevant biological variables derived from Sentinel-2 data for mapping changes from grassland to arable land using random forest classifier. Land. 2020; 9(11):1-20.

[8]Alshari EA, Abdulkareem MB, Gawali BW. Classification of land use/land cover using artificial intelligence (ANN-RF). Frontiers in Artificial Intelligence. 2023; 5:964279.

[9]Thanh NP, Kappas M. Comparison of random forest, k-nearest neighbor, and support vector machine classifiers for land cover classification using Sentinel-2 imagery. Sensors. 2017; 18(1):1-20.

[10]Zhang S, Yang P, Xia J, Wang W, Cai W, Chen N, et al. Remote sensing inversion and prediction of land use land cover in the middle reaches of the Yangtze River basin, China. Environmental Science and Pollution Research. 2023; 30(16):46306-20.

[11]Loukika KN, Keesara VR, Sridhar V. Analysis of land use and land cover using machine learning algorithms on Google earth engine for Munneru River basin, India. Sustainability. 2021; 13(24):1-15.

[12]Sharma A, Gulati T. Change detection in remotely sensed images based on modified log ratio and fuzzy clustering. In information and communication technology for intelligent systems 2018 (pp. 412-9). Springer International Publishing.

[13]Hu Y, Raza A, Syed NR, Acharki S, Ray RL, Hussain S, et al. Land use/land cover change detection and NDVI estimation in Pakistan’s Southern Punjab province. Sustainability. 2023; 15(4):1-21.

[14]Campos-taberner M, García-haro FJ, Martínez B, Izquierdo-verdiguier E, Atzberger C, Camps-valls G, et al. Understanding deep learning in land use classification based on Sentinel-2 time series. Scientific Reports. 2020; 10(1):17188.

[15]Bhakthan SM, Loganathan A. A hyperspectral unmixing model using convolutional vision transformer. Earth Science Informatics. 2024; 17(3):2255-73.

[16]Devanand BK, Patil NS. Leveraging potential of deep learning for remote sensing data: a review. Intelligent systems and human machine collaboration: select proceedings of ICISHMC 2022 (pp. 129-45). Springer.

[17]Chen H, Wu C, Du B, Zhang L, Wang L. Change detection in multisource VHR images via deep siamese convolutional multiple-layers recurrent neural network. IEEE Transactions on Geoscience and Remote Sensing. 2019; 58(4):2848-64.

[18]Naik N, Chandrasekaran K, Sundaram VM, Panneer P. Dual attention guided deep encoder-decoder network for change analysis in land use/land cover for Dakshina Kannada district, Karnataka, India. Environmental Earth Sciences. 2023; 82(1):33.

[19]Ansith S, Bini AA. Land use classification of high resolution remote sensing images using an encoder based modified GAN architecture. Displays. 2022; 74:102229.

[20]Clark A, Phinn S, Scarth P. Optimised U-Net for land use–land cover classification using aerial photography. PFG–Journal of Photogrammetry, Remote Sensing and Geoinformation Science. 2023; 91(2):125-47.

[21]Aryal J, Sitaula C, Frery AC. Land use and land cover (LULC) performance modeling using machine learning algorithms: a case study of the city of Melbourne, Australia. Scientific Reports. 2023; 13(1):13510.

[22]Amini S, Saber M, Rabiei-dastjerdi H, Homayouni S. Urban land use and land cover change analysis using random forest classification of landsat time series. Remote Sensing. 2022; 14(11):1-23.

[23]Avcı C, Budak M, Yağmur N, Balçık F. Comparison between random forest and support vector machine algorithms for LULC classification. International Journal of Engineering and Geosciences. 2023; 8(1):1-10.

[24]Yuh YG, Tracz W, Matthews HD, Turner SE. Application of machine learning approaches for land cover monitoring in northern Cameroon. Ecological Informatics. 2023; 74:101955.

[25]Zhao Z, Islam F, Waseem LA, Tariq A, Nawaz M, Islam IU et al. Comparison of three machine learning algorithms using google earth engine for land use land cover classification. Rangeland Ecology & Management. 2024; 92:129-37.

[26]Alshari EA, Gawali BW. Development of classification system for LULC using remote sensing and GIS. Global Transitions Proceedings. 2021; 2(1):8-17.

[27]MohanRajan SN, Loganathan A, Manoharan P, Alenizi FA. Fuzzy swin transformer for land use/land cover change detection using LISS-III satellite data. Earth Science Informatics. 2024; 17(2):1745-64.

[28]Qi X, Peng J, Wang Y, Qi X, Peng Y. High-resolution land-cover mapping based on a cross-resolution deep learning framework and available low-resolution labels. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2023; 17:1839-56.

[29]Aljebreen M, Mengash HA, Alamgeer M, Alotaibi SS, Salama AS, Hamza MA. Land use and land cover classification using river formation dynamics algorithm with deep learning on remote sensing images. IEEE Access. 2024; 12:11147-56.

[30]Sawant S, Garg RD, Meshram V, Mistry S. Sen-2 LULC: land use land cover dataset for deep learning approaches. Data in Brief. 2023; 51:109724.

[31]Naushad R, Kaur T, Ghaderpour E. Deep transfer learning for land use and land cover classification: a comparative study. Sensors. 2021; 21(23):1-13.

[32]Helber P, Bischke B, Dengel A, Borth D. Eurosat: a novel dataset and deep learning benchmark for land use and land cover classification. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2019; 12(7):2217-26.

[33]Ebenezer PA, Manohar S. Land use/land cover change classification and prediction using deep learning approaches. Signal, Image and Video Processing. 2024; 18(1):223-32.

[34]Albarakati HM, Khan MA, Hamza A, Khan F, Kraiem N, Jamel L, et al. A novel deep learning architecture for agriculture land cover and land use classification from remote sensing images based on network-level fusion of self-attention architecture. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2024; 17:6338-6353.

[35]Fayaz M, Nam J, Dang LM, Song HK, Moon H. Land-cover classification using deep learning with high-resolution remote-sensing imagery. Applied Sciences. 2024; 14(5):1-15.

[36]Dastour H, Hassan QK. A comparison of deep transfer learning methods for land use and land cover classification. Sustainability. 2023; 15(10):1-18.

[37]Arrechea-castillo DA, Solano-correa YT, Muñoz-ordóñez JF, Pencue-fierro EL, figueroa-casas A. Multiclass land use and land cover classification of Andean sub-basins in Colombia with sentinel-2 and deep learning. Remote Sensing. 2023; 15(10):1-20.

[38]Zhang W, Tang P, Corpetti T, Zhao L. WTS: a weakly towards strongly supervised learning framework for remote sensing land cover classification using segmentation models. Remote Sensing. 2021; 13(3):1-18.

[39]Dewangkoro HI, Arymurthy AM. Land use and land cover classification using CNN, SVM, and channel squeeze & spatial excitation block. In conference series: earth and environmental science 2021 (pp. 1-8). IOP Publishing.

[40]Yang Z, Mu XD, Zhao FA. Scene classification of remote sensing image based on deep network and multi-scale features fusion. Optik. 2018; 171:287-93.

[41]Sonune N. Land cover classification with EuroSAT dataset [Internet]. 2020.

[42]Yassine H, Tout K, Jaber M. Improving LULC classification from satellite imagery using deep learning–eurosat dataset. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. 2021; 43:369-76.

[43]Vinaykumar VN, Babu JA, Frnda J. Optimal guidance whale optimization algorithm and hybrid deep learning networks for land use land cover classification. EURASIP Journal on Advances in Signal Processing. 2023; 2023(1):13.

[44]Tejasree G, Agilandeeswari L. Land use/land cover (LULC) classification using deep-LSTM for hyperspectral images. The Egyptian Journal of Remote Sensing and Space Sciences. 2024; 27(1):52-68.

[45]Hosseiny B, Mahdianpari M, Brisco B, Mohammadimanesh F, Salehi B. WetNet: a spatial–temporal ensemble deep learning model for wetland classification using Sentinel-1 and Sentinel-2. IEEE Transactions on Geoscience and Remote Sensing. 2021; 60:1-4.

[46]Yao J, Zhang B, Li C, Hong D, Chanussot J. Extended vision transformer (ExViT) for land use and land cover classification: a multimodal deep learning framework. IEEE Transactions on Geoscience and Remote Sensing. 2023; 61:1-5.

[47]Mcfeeters SK. The use of the normalized difference water index (NDWI) in the delineation of open water features. International Journal of Remote Sensing. 1996; 17(7):1425-32.

[48]Zha Y, Gao J, Ni S. Use of normalized difference built-up index in automatically mapping urban areas from TM imagery. International Journal of Remote Sensing. 2003; 24(3):583-94.

[49]Rikimaru A, Roy PS, Miyatake S. Tropical forest cover density mapping. Tropical Ecology. 2002; 43(1):39-47.

[50]Chong E. EuroSAT land use and land cover classification using deep learning. 2020 [Internet].

[51]Yamashkin SA, Yamashkin AA, Zanozin VV, Radovanovic MM, Barmin AN. Improving the efficiency of deep learning methods in remote sensing data analysis: geosystem approach. IEEE Access. 2020; 8:179516-29.

[52]Temenos A, Temenos N, Kaselimi M, Doulamis A, Doulamis N. Interpretable deep learning framework for land use and land cover classification in remote sensing using SHAP. IEEE Geoscience and Remote Sensing Letters. 2023; 20:1-5.