Image processing apparatus, learning apparatus, image processing method, learning method, and storage medium for estimating printing result
摘要
說明書
Note that the ranges of u and v are 0≤u≤127 and 0≤v≤127, respectively. By the above operation, the output image data IG3(u, v, p) from the pooling layer 403 is 128×128×2 image data.
Then, a deconvolution layer 404 expands the output image data IG3(u, v, p). In this process, the output image data IG3(u, v, p) from the pooling layer 403 is partially expanded and then convolution operation is performed across the planes. First, the expansion of the output image data IG3(u, v, p) from the pooling layer 403 is performed through the procedure below. IT(2u,2v,p)=IG3(u,v,p) IT(2u,2v+1,p)=0 IT(2u+1,2v,p)=0 IT(2u+1,2v+1,p)=0??[Math. 5]
Then, using the equation below, the deconvolution layer 404 performs convolution operation with IT(x, y, p) obtained by the above expansion process as an input.
