17 July 2015 Research on super-resolution image reconstruction based on an improved POCS algorithm
Author Affiliations +
Proceedings Volume 9524, International Conference on Optical and Photonic Engineering (icOPEN 2015); 95242V (2015) https://doi.org/10.1117/12.2195255
Event: International Conference on Optical and Photonic Engineering (icOPEN2015), 2015, Singapore, Singapore
Abstract
Super-resolution image reconstruction (SRIR) can improve the fuzzy image’s resolution; solve the shortage of the spatial resolution, excessive noise, and low-quality problem of the image. Firstly, we introduce the image degradation model to reveal the essence of super-resolution reconstruction process is an ill-posed inverse problem in mathematics. Secondly, analysis the blurring reason of optical imaging process - light diffraction and small angle scattering is the main reason for the fuzzy; propose an image point spread function estimation method and an improved projection onto convex sets (POCS) algorithm which indicate effectiveness by analyzing the changes between the time domain and frequency domain algorithm in the reconstruction process, pointed out that the improved POCS algorithms based on prior knowledge have the effect to restore and approach the high frequency of original image scene. Finally, we apply the algorithm to reconstruct synchrotron radiation computer tomography (SRCT) image, and then use these images to reconstruct the three-dimensional slice images. Comparing the differences between the original method and super-resolution algorithm, it is obvious that the improved POCS algorithm can restrain the noise and enhance the image resolution, so it is indicated that the algorithm is effective. This study and exploration to super-resolution image reconstruction by improved POCS algorithm is proved to be an effective method. It has important significance and broad application prospects – for example, CT medical image processing and SRCT ceramic sintering analyze of microstructure evolution mechanism.
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Haiming Xu, Haiming Xu, Hong Miao, Hong Miao, Chong Yang, Chong Yang, Cheng Xiong, Cheng Xiong, } "Research on super-resolution image reconstruction based on an improved POCS algorithm", Proc. SPIE 9524, International Conference on Optical and Photonic Engineering (icOPEN 2015), 95242V (17 July 2015); doi: 10.1117/12.2195255; https://doi.org/10.1117/12.2195255
PROCEEDINGS
10 PAGES


SHARE
Back to Top