jpex {DRIP} | R Documentation |
jpex() takes in any square matrix (noisy blurry image) and deblurs it.
jpex(image, bandwidth, alpha, sigma)
image |
An input blurry image to deblurred. The input image is represented a square matrix. |
bandwidth |
A positive integer that specifies the size of the neighborhood for local smoothing. |
alpha |
A numberic between 0 and 1. This is the signifance level for the Chi-square hypothesis test that a given pixel is in a continuity region and not affected by the blur. |
sigma |
A positive numeric for the noise level in the blurred image. It is used in the Chi-square test. |
deblurred |
The deblurred image |
edge |
The square matrix, the element of which is the value of the Chi-square test statistic at the pixel location. One can classify a given pixel as a blurry pixcel if edge[i,j]>qchisq(1-alpha, 2). |
Yicheng Kang
Kang, Y. (2020) “Consistent Blind Image Deblurring Using Jump-Preserving Extrapolation”, Journal of Computational and Graphical Statistics, 29(2), 372 – 382.
library(DRIP)
data(stopsign)
out <- jpex(image = stopsign, bandwidth = as.integer(2), sigma =
0.00623, alpha = 0.001)