Automatic focusing method

Abstract

The invention discloses an automatic focusing method, comprising the steps as follows: collecting an image; an image processing unit using a definition evaluation function value to evaluate defocusing amount, and controlling a focusing unit to focus according to the defocusing amount, wherein the image definition evaluation function value is a specific value of a high-frequency component amplitude and a low frequency component amplitude, and the frequency of a low frequency component is less than that of a high-frequency component, and the proportion that the low frequency component occupies the number of all frequency components obtained by calculating is 3 to 6 percents. The method of the invention improves sensitivity and accuracy of the image definition evaluation function value at a server defocusing place so as to be beneficial to realize fast and accurate automatic focusing.

Description

technical field [0001] The present invention relates to the technical field of automatic focusing based on image processing of an imaging system, in particular to an automatic focusing method. Background technique [0002] A clear image can correctly reflect the observed things, so that people can get accurate information. The imaging system is an effective tool for image acquisition. However, if the imaging system is out of focus, the image will be blurred. Therefore, it is necessary to adjust the focus so that the system is in focus to obtain a clear image. Due to its strong subjectivity and low efficiency, manual focusing has given birth to the vigorous development of automatic focusing. Nowadays, most imaging devices such as mobile phone cameras, digital cameras, video cameras, microscopes, and space telescopes require and use automatic focusing systems. [0003] Autofocus is the process of first evaluating the defocus amount of the system, and then controlling the foc...

AI Technical Summary

Problems solved by technology

However, in digital images, we can only obtain corresponding discrete point values, so the number of points obtained at certain frequencies is large, and the number of points obtained at certain frequencies is small. If the above method is used, it cannot truly reflect the changes of high-frequency components.

Moreover, when the above method is seriously defocused, the high-frequency components of the image are very small, and the change of the high-frequency components will not be obvious. Therefore, although the defocus amount changes greatly when the image is seriously defocused, the value of the image definition evaluation function does not change much. , the value of the image sharpness evaluation function cannot reflect the change of the defocus amount well

Therefore, the sensitivity and accuracy of the image sharpness evaluation function value of the classical frequency domain method at severe defocus needs to be improved urgently

Images