Academic Journal of Computing & Information Science, 2025, 8(5); doi: 10.25236/AJCIS.2025.080506.
Yuan Zhang1, Xinghong He1, Xiyan Yang2
1College of Hydraulic and Architectural Engineering, Tarim University, Alar City, China
2College of Information Engineering, Tarim University, Alar City, China
With the rapid evolution of global land use patterns, accurate and efficient land use identification technologies are crucial for resource management and ecological protection. This study focuses on the application of the Extreme Learning Machine (ELM) in land use identification and systematically conducts a sensitivity analysis of Landsat bands. By constructing an ELM classification model, we quantify the contribution of each band and band combination to the recognition accuracy of different land use types. The results indicate that the ELM model exhibits excellent performance in land use identification, achieving an overall identification accuracy of 80.9% when the number of hidden neurons is set to 90. The sensitivity analysis reveals that bands B3, B4, and B7 demonstrate significant advantages in desert identification, consistently maintaining an accuracy of 100%. Bands B5 and B7 achieve an accuracy of 99% in water identification, while identifying construction land presents the main challenge, with an average accuracy of only 54.7%. The further introduction of vegetation indices (e.g., EVI, NDVI) and band combination analysis shows that EVI enhances the recognition accuracy of forest land by 29.3%, and NDWI improves the recognition accuracy of construction by 10%. This verifies the important value of band sensitivity analysis in optimizing feature selection.This study confirms that band sensitivity analysis based on ELM can effectively reveal the intrinsic correlations between spectral data and land use types, providing a scientific basis for constructing high-precision and robust land use identification models, which is significant for advancing the application of remote sensing technology in land resource monitoring and management.
Extreme Learning Machine; Land Use; Band Sensitivity Analysis
Yuan Zhang, Xinghong He, Xiyan Yang. Application of Extreme Learning Machine for Land Use Identification. Academic Journal of Computing & Information Science (2025), Vol. 8, Issue 5: 55-63. https://doi.org/10.25236/AJCIS.2025.080506.
[1] Sekrecka A ,Karwowska K .Classical vs. Machine Learning-Based Inpainting for Enhanced Classification of Remote Sensing Image[J].Remote Sensing,2025,17(7):1305-1305.
[2] Diab M ,Kolokoussis P ,Brovelli A M .Optimizing zero-shot text-based segmentation of remote sensing imagery using SAM and Grounding DINO[J]. Artificial Intelligence in Geosciences, 2025, 6(1):100105-100105.
[3] G. Cheng, P. Zhou, J. Han, et al., Auto-encoder-based shared mid-level visual dictionary learning for scene classification using very high resolution remote sensing images, IET Comput. Vis. 9 (5) (2015) 639–647.
[4] He M , Zhang J ,He Y , et al. Annotated Dataset for Training Cloud Segmentation Neural Networks Using High-Resolution Satellite Remote Sensing Imagery[J].Remote Sensing,2024,16(19):3682-3682. .
[5] G.B. Huang, Q.Y. Zhu, C.K. Siew, Extreme learning machine: theory and applications, Neurocomputing 70 (1-3) (2006) 489–501.
[6] Han M ,Liu B . Ensemble of extreme learning machine for remote sensing image classification [J]. Neurocomputing, 2015, 149 65-70.
[7] Li Q, Peng Q, Chen J, et al. Improving Image Classification Accuracy with ELM and CSIFT[J]. Computing in Science & Engineering, 2018.
[8] G.B. Huang , An insight into extreme learning machines: Random neurons, random features and kernels, Cognit Comput 6 (3) (2014) 376–390 .
[9] Arai K ,Nakaoka Y ,Okumura H .Method for Landslide Area Detection with RVI Data Which Indicates Base Soil Areas Changed from Vegetated Areas[J].Remote Sensing,2025,17(4):628-628.
[10] Meng Z ,Lu Y ,Wang H .Correlation change analysis and NDVI prediction in the Yellow River Basin of China using complex networks and GRNN-PSRLSTM.[J].Environmental monitoring and assessment, 2024, 196(11) :1092.
[11] Wei H Z ,Chen L H ,Zha N F .The Modification of Difference Vegetation Index (DVI) in middle and late growing period of winter wheat and its application in soil moisture inversion[J].E3S Web of Conferences, 2019,13101098.
[12] Huiliang W ,Linpo H ,Jun Y , et al.Effects of Effective Precipitation and Accumulated Temperature on the Terrestrial EVI (Enhanced Vegetation Index) in the Yellow River Basin, China[J]. Atmosphere, 2022,13(10):1555-1555.
[13] Castro D L A ,Duarte L M ,Ewbank H , et al.Use of synthetic aperture radar data for the determination of normalized difference vegetation index and normalized difference water index[J]. Journal of Applied Remote Sensing,2024,18(1):014516-014516.