Assessing The Use Of Artificial Neural Network For Feature Detection In Sentinel 2 And Landsat 8

• Dr. (Mrs) E.O Makinde

  University of Lagos

  Author

  https://orcid.org/0000-0001-5857-1349

• Oyewole Ayomide. P

  University of Lagos

  Author

• Abdul Hikmat. T

  University of Lagos

  Author

With advancements in remote sensing technology, higher-resolution satellite data became increasingly common, leading to varying outcomes influenced by the classifier and software capabilities due to distinct operational processes of each classifier. This thesis aims to evaluate the effectiveness of Artificial Neural Network (ANN) for feature detection in satellite imagery obtained from Sentinel-2 and Landsat 8 sensors. The study focuses on land use and land cover (LULC) mapping and compares the performance of ANN between the two satellite datasets. The research methodology involves using identical training and validation datasets to ensure a fair comparison. The accuracy of the classified imageries is assessed using the confusion matrix, while change detection analysis is conducted to identify temporal variations in land cover. The results show that ANN demonstrates satisfactory results in both Sentinel-2 and Landsat 8 imagery, with higher accuracy achieved in Sentinel-2 data. The change detection analysis highlights the dynamics of land cover changes and provides valuable insights into the spatio-temporal patterns. This research contributes to the field of remote sensing by enhancing our understanding of the effectiveness of ANN for feature detection and its potential in LULC mapping. The findings have implications for various applications such as urban planning, environmental management, and natural resource assessment. 

1. Dr. (Mrs) E.O Makinde, University of Lagos

   Dr. (Mrs) E.O Makinde is an associate professor in the Department of Surveying and Geoinformatics at the University of Lagos, Nigeria. (Email: estherdanisi@gmail.com)

• pdf

Copyright (c) 2026 The Publication of International Journal of Data Analytics and Smart Education (IJDASE)

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.