A novel blind super‐resolution technique based on the improved Poisson maximuma posteriorialgorithm

Abstract

Image restoration has received considerable attention. In many practical situations, unfortunately, the blur is often unknown, and little information is available about the true image. Therefore, the true image is identified directly from the corrupted image by using partial or no information about the blurring process and the true image. In addition, noise will be amplified to induce severely ringing artifacts in the process of restoration. This article proposes a novel technique for the blind super-resolution, whose mechanism alternates between de-convolution of the image and the point spread function based on the improved Poisson maximum a posteriori superresolution algorithm. This improved Poisson MAP super-resolution algorithm incorporates the functional form of a Wiener filter into the Poisson MAP algorithm operating on the edge image further to reduce noise effects and speed restoration. Compared with that based on the Poisson MAP, the novel blind super-resolution technique presents experimental results from 1-D signals and 2-D images corrupted by Gaussian point spread functions and additive noise with significant improvements in quality.

Improving Particle Segmentation from Process Images with Wiener Filtering

“…The noise-to-signal ratio (NSR) required for the filtering is rarely known, but it can be approximated by a constant [9]. Replacing NSR with constant γ, the frequency representation off minimizing the error function in Eq.…”

Improving a Single Down-Sampled Image Using Probability-Filtering-Based Interpolation and Improved Poisson Maximum A Posteriori Super-Resolution

“…The whole process of this improved Poisson MAP includes reblurring, differentiation, restoration, integration, and then correction for a DC offset. More details about this algorithm can be found in the author's previous work [23].…”

“…This paper first addresses an approach to simultaneous image interpolation and smoothing by exploiting the probability filter [22] coupled with a pyramidal decomposition, thereby extending the conventional applications of the probability filter originally designed for noise removal. Then, the improved Poisson maximum a posteriori (MAP) super-resolution [23] is performed to reconstruct the high spatial-frequency spectrum of the interpolated image. Thus, the hybrid scheme shown in Figure 1(b) is proposed for enhancing the spatial and the spatial-frequency resolutions of down-sampled images.…”

Edge-image-based approach for stable super-resolution reconstruction

“…10,11 Similar research using a Poisson MAP algorithm was studied by Pan. 12 Many SR authors first fuse all LR images together, followed by a deblurring step. They aim to increase the perceptual qualities of the degraded ͑e.g., blurred, noisy͒ images by the Lorentzian distribution in the deblurring step.…”