The Use of 3D Convolutional Autoencoder in Fault and Fracture Network Characterization

Figure 16

Maps of the target horizon in the Liziba survey. (a) Two-step clustering results based on the poststack data. (b) Two-step clustering results based on prestack data. (c) SOM clustering results based on the 3D convolutional autoencoder. (d) Two-step clustering results based on the 3D convolutional autoencoder.

(a)

(b)

(c)

(d)