International Journal of Advances in Applied Computational Intelligence
Journal DOI
2833-5600ISSN (Online)
Volume 6 , Issue 1 , PP: 13-29, 2024 | Cite this article as | XML | Html | PDF | Full Length Article
Bahromjon Urmanov 1 , Maha Ibrahim 2
Doi: https://doi.org/10.54216/IJAACI.060102
Accurate remote sensing (RS) monitoring of wetland ground objects is an enormous importance for ecological preservation. Wetland classification on multi-source remote sensing images (MS-RSI) includes leveraging data from different sensors for accurately describing and categorizing wetland regions. This method normally incorporates data from infrared, radar, and optical sensors to take a wide-ranging view of wetland features. Advanced image processing methodologies, comprising machine learning (ML) approaches are often implemented for analyzing these multi-source images as well as recognizing spectral and spatial patterns indicative of wetland characteristics. The interaction of various RS data increases the accuracy and robustness of wetland classification models, allowing a more complex analysis of wetland ecosystems and aiding environmental observation, conservation, and control measures. To accomplish effective training for wetland mapping through the RS, it is essential for a significant training data that comprises a numerous array of class variants. In this article, we propose an Enhanced Wetland Classification using a Deep Learning based Fusion Approach (EWC-DLFA) on MS-RSI. The proposed EWC-DLFA technique examines the MS-RSI for wetland classification using the DL model which can be used for other land cover classification types. To accomplish this, the EWC-DLFA technique utilizes the data from multiple sources such as Sentinel-1 (SAR), Landsat-8, Sentinel2 (multi-spectral), and digital elevation model (DEM). In the presented EWC-DLFA technique, a deep convolutional neural network-based EfficientNetB-5 model can be applied for the extraction of features from the multi-source images. For increasing the performance of the EfficientNet-B5 model, the marine predators algorithm (MPA) based hyper parameter tuning process can be applied. Finally, an ensemble of three ML classifiers such as extreme learning machine (ELM), multilayer perceptron (MLP), and gradient boosting machine (GBM) are used to classify the wetland into different types such as fen, bog, marsh, swamps, and upland. The performance of the EWC-DLFA technique can be validated using a large set of simulations. The resultant values pointed out that the EWC-DLFA technique reaches better performance over other models on wetland classification.
Wetland , Remote Sensing , Deep Learning , Marine Predators Algorithm , Fusion Model
[1] Ahmed, K.R.; Akter, S.; Marandi, A.; Schüth, C. A Simple and Robust Wetland Classification Approach by Using Optical Indices, Unsupervised and Supervised Machine Learning Algorithms. Remote Sens. Appl. Soc. Environ. 2021, 23, 100569.
[2] Miyamoto, M.; Kushida, K.; Yoshino, K.; Nagano, T.; Sato, Y. Evaluation of multispatial scale measurements for monitoring wetland vegetation, Kushiro wetland, JAPAN: Application of SPOT images, CASI data, airborne CNIR video images and balloon aerial photography. In Proceedings of the IGARSS 2003: IEEE International Geoscience and Remote Sensing Symposium, Learning from Earth’s Shapes and Sizes, Toulouse, France, 21–25 July 2003; Volume I–VII, pp. 3275–3277.
[3] van Asselen, S.; Verburg, P.H.; Vermaat, J.E.; Janse, J.H. Drivers of Wetland Conversion: A Global Meta-Analysis. PLoS ONE 2013, 8, e81292.
[4] Mahdianpari, M.; Brisco, B.; Granger, J.; Mohammadimanesh, F.; Salehi, B.; Homayouni, S.; Bourgeau-Chavez, L. The Third Generation of Pan-Canadian Wetland Map at 10 m Resolution Using Multisource Earth Observation Data on Cloud Computing Platform. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2021, 14, 8789–8803.
[5] Amani, M.; Mahdavi, S.; Afshar, M.; Brisco, B.; Huang, W.; Mirzadeh, S.M.J.; White, L.; Banks, S.; Montgomery, J.; Hopkinson, C. Canadian Wetland Inventory using Google Earth Engine: The First Map and Preliminary Results. Remote Sens. 2019, 11, 842.
[6] Bwangoy, J.-R.B.; Hansen, M.C.; Roy, D.P.; De Grandi, G.; Justice, C.O. Wetland mapping in the Congo Basin using optical and radar remotely sensed data and derived topographical indices. Remote Sens. Environ. 2010, 114, 73–86.
[7] Ceron, C.N.; Melesse, A.M.; Price, R.; Dessu, S.B.; Kandel, H.P. Operational actual wetland evapotranspiration estimation for South Florida using MODIS imagery. Remote Sens. 2015, 7, 3613–3632.
[8] Davranche, A.; Lefebvre, G.; Poulin, B. Wetland monitoring using classification trees and SPOT-5 seasonal time series. Remote Sens. Environ. 2010, 114, 552–562.
[9] Mahdianpari, M.; Salehi, B.; Mohammadimanesh, F.; Motagh, M. Random forest wetland classification using ALOS-2 L-band, RADARSAT-2 C-band, and TerraSAR-X imagery. ISPRS J. Photogramm. Remote Sens. 2017, 130, 13–31.
[10] Shafizadeh-Moghadam, H.; Khazaei, M.; Alavipanah, S.K.; Weng, Q. Google Earth Engine for Large-Scale Land Use and Land Cover Mapping: An Object-Based Classification Approach Using Spectral, Textural and Topographical Factors. GIScience Remote Sens. 2021, 58, 914–928.
[11] Jamali, A., Mahdianpari, M., Brisco, B., Mao, D., Salehi, B. and Mohammadimanesh, F., 2022. 3DUNetGSFormer: A deep learning pipeline for complex wetland mapping using generative adversarial networks and Swin transformer. Ecological Informatics, 72, p.101904.
[12] Hosseiny, B., Mahdianpari, M., Brisco, B., Mohammadimanesh, F. and Salehi, B., 2021. WetNet: A spatial–temporal ensemble deep learning model for wetland classification using Sentinel-1 and Sentinel-2. IEEE Transactions on Geoscience and Remote Sensing, 60, pp.1-14.
[13] Guo, F., Li, Z., Meng, Q., Ren, G., Wang, L., Wang, J., Qin, H. and Zhang, J., 2023. Semi-supervised cross-domain feature fusion classification network for coastal wetland classification with hyperspectral and LiDAR data. International Journal of Applied Earth Observation and Geoinformation, 120, p.103354.
[14] Zhao, J., Wang, L., Yang, H., Wu, P., Wang, B., Pan, C. and Wu, Y., 2022. A land cover classification method for high-resolution remote sensing images based on NDVI deep learning fusion network. Remote Sensing, 14(21), p.5455.
[15] Bosco, M.J., Wang, G. and Hategekimana, Y., 2021. Learning multi-granularity neural network encoding image classification using DCNNs for Easter Africa Community Countries. IEEE Access, 9, pp.146703-146718.
[16] Hu, X., Zhang, P., Zhang, Q. and Wang, J., 2021. Improving wetland cover classification using artificial neural networks with ensemble techniques. GIScience & Remote Sensing, 58(4), pp.603-623.
[17] Yan, J., Liu, J., Liang, D., Wang, Y., Li, J. and Wang, L., 2023. Semantic segmentation of land cover in urban areas by fusing multi-source satellite image time series. IEEE Transactions on Geoscience and Remote Sensing.
[18] Jafarzadeh, H., Mahdianpari, M. and Gill, E.W., 2022. Wet-GC: A Novel Multimodel Graph Convolutional Approach for Wetland Classification Using Sentinel-1 and 2 Imagery with Limited Training Samples. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 15, pp.5303-5316.
[19] Pour, A.M., Seyedarabi, H., Jahromi, S.H.A. and Javadzadeh, A., 2020. Automatic detection and monitoring of diabetic retinopathy using efficient convolutional neural networks and contrast limited adaptive histogram equalization. IEEE Access, 8, pp.136668-136673.
[20] Sekhar, J.C., Rajyalakshmi, C., Nagaraj, S., Sankar, S., Saturi, R. and Harshavardhan, A., 2023. Deep Generative Adversarial Networks with Marine Predators Algorithm for Classification of Alzheimer’s Disease Using Electroencephalogram. Journal of King Saud University-Computer and Information Sciences, p.101848.
[21] Xiao, D., Li, B., Shan, J., Yan, Z. and Huang, J., 2023. SOC Estimation of Vanadium Redox Flow Batteries Based on the ISCSO-ELM Algorithm. ACS Omega.
[22] Altay, O. and Varol Altay, E., 2023. A novel hybrid multilayer perceptron neural network with improved grey wolf optimizer. Neural Computing and Applications, 35(1), pp.529-556.
[23] AlShourbaji, I., Helian, N., Sun, Y., Hussien, A.G., Abualigah, L. and Elnaim, B., 2023. An efficient churn prediction model using gradient boosting machine and metaheuristic optimization. Scientific Reports, 13(1), p.14441.
[24] Judah, A. and Hu, B., 2022. An advanced data fusion method to improve wetland classification using multi-source remotely sensed data. Sensors, 22(22), p.8942.
