Rapid multi-point automatic focusing method and system and application equipment thereof

Abstract

The invention provides a rapid multi-point automatic focusing method, which comprises the following steps of: 1, acquiring a target image based on an equipment acquisition terminal, dividing the target image into a plurality of sub-images to obtain M * N sub-target images, and 2, adjusting the exposure time of each sub-target image to obtain M * N sub-target images, obtaining the brightness of each sub-target image at the optimal exposure time; 3, obtaining a sharpness evaluation score of each sub-target image under the current focusing value; 4, screening the sub-target image with the maximum sharpness score number, and carrying out the fine adjustment of a step value; and 5, obtaining a focusing value of the sub-target image of which the step size value is finely adjusted at the maximum value of the new sharpness evaluation score. A partitioning strategy is adopted for a whole image, and a block interval abandoning threshold value and an edge abandoning threshold value are set, so that the method is different from the prior art and has the characteristics of reducing the operation complexity and improving the efficiency.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a fast multi-point autofocus method, system and application equipment thereof. Background technique [0002] The current existing autofocus algorithms are as follows: [0003] One is to directly calculate the image gradient value, and use the hill-climbing search algorithm to obtain the local optimal focal plane position, which requires the evaluation function to have absolute unimodality and extremely high sensitivity, otherwise, the algorithm will easily regard the local extremum of the evaluation function as the optimal The best focal plane position will lead to focusing failure, or the evaluation function tends to be flat in the out-of-focus area, resulting in the motor not getting the reverse signal for a long time, and finally entering an endless loop; [0004] The other is to use deep learning algorithms to obtain a large amount of training set data. [0005] In...

AI Technical Summary

Problems solved by technology

[0006] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a fast multi-point autofocus method, system and application equipment to solve the problems raised in the above-mentioned background technology. strategy, and set the block interval discarding threshold and edge discarding threshold to achieve a way to reduce computational complexity and improve efficiency, which solves the problems in the prior art

Images