|Communications on Applied Electronics|
|Foundation of Computer Science (FCS), NY, USA|
|Volume 7 - Number 33|
|Year of Publication: 2020|
|Authors: Temitope M. Ogungboyega, Kingsley M. Udofia, Chidinma N. Kalu|
Temitope M. Ogungboyega, Kingsley M. Udofia, Chidinma N. Kalu . Adaptive Neuro-fuzzy Inference System based Earth Surface Features Classification System. Communications on Applied Electronics. 7, 33 ( Mar 2020), 13-18. DOI=10.5120/cae2020652856
This paper aimed at developing a model that will aid the process of identifying and extracting earth surface features from satellite images using adaptive neuro-fuzzy inference system. Conventional methods of classifying earth features (Normalized Difference Vegetation Index, NDVI, and Normalized Difference Water Index, NDWI) were first used to generate the data for the training of the ANFIS model using the three bands in Landsat 8 (band 2: Blue, band 4: Red and band 5: NIR). The performance of the developed ANFIS model was validated using four different satellite images and the results compared with the conventional classification methods. An accuracy level of 98.66 – 99.88 % with a RMSE of 0.0218 – 0.0506 were obtained.