Corner detection-based simple lens PSF iterative optimization initial value method

Abstract

The present invention discloses a corner detection-based simple lens PSF iterative optimization initial value method, and relates to the image restoration field. The method comprises the steps of using a simple lens to shoot a clear image and a checkerboard image to obtain a fuzzy image and the checkerboard image; based on the checkerboard image, adopting a corner detection method to obtain the accurately matched clear image and fuzzy image pairs; according to the prior knowledge of a simple lens PSF, generating a series of disc-shaped PSF of different fuzzy degrees; convoluting the matched clear images and the generated series of PSF separately to obtain the generated fuzzy images, comparing the similar degrees between the generated fuzzy images and the matched fuzzy images, taking the PFS corresponding to the most similar generated fuzzy image as the initial value of a simple lens PSF estimation iterative optimization process; putting the selected PSF initial value in a PSF estimation optimization algorithm, and estimating the PSF corresponding to the simple lens rapidly. The method of the present invention can avoid the influence of the local optimal solution of an optimization algorithm on the PSF precision, and enables the final image restoration quality to be improved.

Description

technical field [0001] The invention relates to the field of digital image intelligent processing, in particular to the field of image restoration, in particular to a method for iteratively optimizing the initial value of a simple lens PSF based on corner detection. Background technique [0002] At present, SLR cameras play an increasingly important role in people's daily life. However, in order to compensate for the geometric distortion and aberration of the lens in the SLR lens and further improve the image quality, the design of the SLR lens is increasingly complex, and even contains dozens of independent optical devices. While improving image quality, complex lenses also increase the size and weight of the lens, resulting in a significant increase in the cost of the lens. In recent years, with the development of computational photography technology, simple lenses combined with post-image restoration algorithms have gradually become a new research direction in the field ...

AI Technical Summary

Problems solved by technology

Although this method can obtain an initial value closer to the real PSF to a certain extent, the main problem of this method is that in order to estimate the PSF of a simple lens, it is necessary to make a batch of the same simple lenses. It is not easy to realize, and even a simple lens of the same type has errors in the production process

Images