Calligraphy and Painting Identification 3D-CNN Model Based on Hyperspectral Image MNF Dimensionality Reduction

Figure 9

Output comparison of the convolution layer of 2D-CNN models with different functions under the same painting data as input. (a) Input layer: first nine channels of MNF images (the upper shows the 2D-CNN author classification result/the following shows the 2D-CNN authenticity identification result). (b) Output feature images of the fourth convolution layer (the upper shows the 2D-CNN author classification result/the following shows the 2D-CNN authenticity identification result).

(a)

(b)