Soil Erosion Effects in Microwave Backscattering from Bare Fields

Zerdev, N. G.; Kulemin, Gennady P.

doi:10.1109/euma.1994.337247

Cited by 5 publications

(1 citation statement)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Three types of classifiers have been investigated: a standard Bayesian classifier22 supposing gaussian distribution of sign distribution, the three layer perceptron and RBFN, both with corresponding optimal number of hidden layer neurons. Without any filtering 0.9 10 0.985-0.989 [5][6][7][8] 0.986 [10][11][12][13] The analysis of data presented in Table permits to make the following conclusions: The probability of correct classification Pc was evaluated.…”

Section: Numerical Simulation Resultsmentioning

confidence: 99%

Soil-erosion-state interpretation using pre- and postprocessing of multichannel radar images

et al. 1998

View full text Add to dashboard Cite

The multistage procedure of multichannel image pre-and post-processing is proposed. It includes nonlinear image-toimage and image-to-topology map superimposing with optimization of transform parameters, image separate and vector filtering for remained error reduction and image enhancement. Then the image preliminary recognition is to be performed for determined homogeneous regions, two possible ways of data interpretation based on the use of radiophysical models or supervised learning neural networks are discussed. It is shown that the proposed procedure provides a satisfactory determination of field lots with different erosion and other useful results.

show abstract

Section: Numerical Simulation Resultsmentioning

confidence: 99%