Using the Hopfield neuron network enhances the accuracy of identifying objects smaller than 1 pixel in the spectral image
PDF (Tiếng Việt) | Download: 42

Working Languages

How to Cite

Dang, T. T., Do, N. H., Pham, D. T., Do, M. H., Nguyen, T. L., Luong, T. K. D., & Ly, T. H. (2019). Using the Hopfield neuron network enhances the accuracy of identifying objects smaller than 1 pixel in the spectral image. Journal of Geodesy and Cartography, (40), 27–33. https://doi.org/10.54491/jgac.2019.40.309

Abstract

In the field of remote sensing, soft classification is considered to have higher accuracy than hard classification. For the super resolution, the Hopfield neuron network was used to perform an opti-mization problem in determining the position of the layer in each pixel. This study evaluates the effectiveness of the new algorithm with the SPOT4 image and the new method can increase the accuracy of all layers, especially those of smaller 1 pixel.
https://doi.org/10.54491/jgac.2019.40.309
PDF (Tiếng Việt) | Download: 42

Downloads

Download data is not yet available.