Category Ranking
98%
Total Visits
921
Avg Visit Duration
2 minutes
Citations
20
Article Abstract
Optical imaging systems are significantly affected by aerodynamic thermal effects under varying flight conditions, resulting in complex image blurring. To address this challenge, this study proposes a novel wavefront-coded image restoration method based on a multi-scale deep autoencoder neural network (MS-DAE). By modulating blur levels and incorporating a multi-scale loss function with residual attention mechanisms, the proposed method achieves a remarkable improvement in peak signal-to-noise ratio (PSNR) by 16.03 dB and structural similarity index measure (SSIM) by 0.3834 compared to Wiener filtering; compared to the BaseNet model referenced in this paper, the proposed model achieves improvements of 5.69 dB and 0.026 in PSNR and SSIM, respectively, and demonstrates superior performance in detail restoration in the reconstructed images. The approach effectively restores image details, suppresses artifacts, and adapts to diverse flight conditions, demonstrating significant potential for practical applications.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1364/OE.560884 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
We present the design of a single lens infrared imaging system based on the wavefront coding approach. Spatially invariant PSF with consistent performance over the full field of view ±(4×3) and athermalization (-20 to +60) is achieved in a compact single lens design by a combination of both lens bending and introduction of a generalized cubic phase on the back surface of the lens. The aberration compensation characteristics of the cubic phase and the generalized cubic phase are analyzed, and it is observed that the generalized cubic phase performs better against quadratic-order aberrations.
View Article and Find Full Text PDFOptical imaging systems are significantly affected by aerodynamic thermal effects under varying flight conditions, resulting in complex image blurring. To address this challenge, this study proposes a novel wavefront-coded image restoration method based on a multi-scale deep autoencoder neural network (MS-DAE). By modulating blur levels and incorporating a multi-scale loss function with residual attention mechanisms, the proposed method achieves a remarkable improvement in peak signal-to-noise ratio (PSNR) by 16.
View Article and Find Full Text PDFWe propose a computational imaging technique for expanding the field of view of infrared thermometers. The contradiction between the field of view and the focal length has always been a chief problem for researchers, especially in infrared optical systems. Large-area infrared detectors are expensive and technically arduous to be manufactured, which enormously limits the performance of the infrared optical system.
View Article and Find Full Text PDFWavefront-coded phase measuring deflectometry for the all-focused measurement.
Opt Lett
September 2022
Article Synopsis
- Phase measuring deflectometry is a way to measure complicated shapes by using reflected light patterns, but it has trouble focusing on both the image and the surface at the same time.
- To fix this problem, scientists used a new technique called wavefront coding, which helps the system work better even when things aren't perfectly in focus.
- Their new method was tested on a really curved surface and showed that it could make measurements four times more accurate, helping to see and measure complex shapes more easily.
The Richardson-Lucy (RL) algorithm is a well-known nonlinear restoration method and has been widely applied in the fields of astronomical image restoration, microscopic image restoration, and so on because of its capability of generating high-quality restoration results and potential in realizing super-resolution. However, when being applied to restore the wavefront coded blurry images, the classical RL algorithm converges very slowly and has to be iterated many times before obtaining a satisfactory result, which severely prohibits its real-time application. Vector-extrapolation-based RL algorithm was invented to solve this problem, but the noise amplification increases fast, and additional post-processing is needed to further improve the signal-to-noise ratio.
View Article and Find Full Text PDF