BioMed Research International / 2016 / Article / Fig 5

Research Article

Segmenting Brain Tissues from Chinese Visible Human Dataset by Deep-Learned Features with Stacked Autoencoder

Figure 5

Visualization of learned high-level features of input pixel intensities with three-layer SAE. (a), (b), and (c) Learned feature representation in the first (with 225 units), second (with 144 units), and third (with 81 units) hidden layers, respectively, where the features in the third layer are discriminative for image segmentation task.