The difference images between the input image and the reconstructed image are presented in (c), (e), (g), and (i).
The amount of I 1 indicates the similarity between the input image and the stored image (#1).
Such processors are designed based on cross correlation in the frequency domain between the input image and the input weights of neural networks.
This observation intuitively corresponds to the fact that learning shared weights improves the inner relation between the input image and the baseline.
(1 - 3), the output intensity can be gotten by convolution between the input image and the Gaussian function, the process can be represented as: {I}_cleft x,yright)={I}_{blur}^0{left x,yright)}^{ast }gleft x,yright) (16).
The average absolute discrete entropy difference DEave between the input image X i and the output image Y i is defined as DE ave = 1 n ∑ i = 1 n DE X i - DE Y i. (19).
In order to compare the performance of the JEC process with various window sizes, the difference images between the input image and the reconstruction images are presented in Figure12c,e,g,i, and the difference image presents the jagged edge artifacts corrected by the JEC process.
Image noise estimation methods can be categorized into either smoothing-based approaches, where noise is computed using the difference between the input image and a smoothed version of it, or block-based approaches, where block variances of the most homogenous block in a set of image blocks is taken as noise variance[63].
In order to handle such sudden background changes, we calculated the image intensity difference between the input image and the RBI and determined that sudden background changes occurred if the difference was larger than the given threshold: if 1 N x ⋅ N y ∑ i = 0, j = 0 i = N x - 1, j = N y - 1 I k I i, j - RBI I i, j > 30, a sudden background change occurs at the k ‒ th sequence.
The top-left image represents the input images.
