An optical super-resolution image reconstruction method

Abstract

The invention discloses an optical super-resolution image reconstruction method. The method comprises steps of estimating a point spread function of a low resolution image according to an imaging system and collected pixel grey value distribution of the low resolution image; performing quality degradation of a read low resolution reference image according to the estimated point spread function, comparing a processing result with the low resolution image, and determining whether or not a comparison result exceeds a threshold range; and if so, correcting the reference image using a specific projection correction operator, and repeating a previous step until the comparison result is within the threshold range, thereby obtaining a reconstructed optical super-resolution image. The invention can effectively improve the resolution of a picture.

Description

technical field [0001] The invention relates to the field of optical imaging and image processing, in particular to an optical super-resolution image reconstruction method. Background technique [0002] Image super-resolution reconstruction technology is an emerging interdisciplinary subject in the fields of information optics and digital image processing. It has a wide range of applications in national defense, medicine, intelligent transportation, public safety, space exploration, geographic information system, optical microscopic measurement, computer vision, pattern recognition and other fields. It is of great significance to study image super-resolution reconstruction technology. [0003] There are various factors that affect image quality. The first reason is the "limitation" of the imaging system, since all optical imaging systems are diffraction-limited systems. Simply put, this system cannot incorporate all the "frequency" information into the image, and many "hi...

AI Technical Summary

Problems solved by technology

The first reason is the "limitation" of the imaging system, this is because all optical imaging systems are diffraction limited systems

To put it simply, this system cannot incorporate all the "frequency" information into the image, and many "high frequency" information are not included in the image information, resulting in blurred images without high frequency structural information

The second reason is that due to the existence of evanescent waves, this determines that many "ultra-high frequencies" cannot enter the imaging system, and these evanescent waves have been "annihilated" and attenuated on the surface of the object

The third reason is that many imaging devices are now based on digital CCD imaging. Due to the limitation of sensor unit size and performance, the sampling rate of imaging is insufficient, which leads to poor imaging quality.

However, there is no relatively mature solution in the existing technology, so it is impossible to effectively reconstruct high-quality "super-resolution" images

Images