Automatic focusing method for integrated movement

Abstract

The invention discloses an automatic focusing method for an integrated movement. The method comprises: calculating the position of a clear point according to a set prediction curve; deleting and selecting details by using evaluation values of a high-pass filter according to small window WIN-A, deleting small windows with small evaluation values, and selecting key windows key-win, wherein the window with rich scene generally has a large evaluation value; judging the current scene according to the key-windos, and selecting a search algorithm; determining the direction with little step walk, and then quickly climbing a slope by using an L curve; judging a deceleration point threshold according to an H curve, and beginning fine adjustment; after focusing, starting a scene monitoring module; starting an intelligent virtual focus detection module, and judging whether an image is in a virtual focus state or not by edge detection, wherein the module assists the monitoring module in achieving complete no virtual focus. By adopting the method, the automatic focusing technology is more quick, accurate and intelligent, and the focusing success probability is improved.

Description

technical field [0001] The invention relates to the field of security monitoring, in particular to an automatic focusing method of an integrated movement. Background technique [0002] At present, digital cameras and video cameras based on CMOS and CCD have replaced traditional tape cameras. Cameras with zoom lenses are widely used because they can switch the focal length at will, obtain scenes from different perspectives, and can observe details very clearly. Traditional manual focusing is not only slow, but also the entire focusing process is cumbersome and inoperable. The zoom camera based on auto-focus technology can automatically and quickly focus. [0003] At present, domestic advanced auto-focus technology still lags behind foreign countries, and few have independent intellectual property rights. The existing auto-focus algorithm mainly uses the high-pass filter of the digital image to calculate an evaluation value, and then uses this evaluation value to drive the m...

AI Technical Summary

Problems solved by technology

[0005] 1. The motor will move back and forth near the clear point, resulting in slow focusing speed;

[0006] 2. The focusing process is not smooth, the overshoot range will be relatively large, and the sensory effect of the human eye is relatively poor;

[0007] 3. Point light source environments, environments with less details, and low-illumination scenes will cause focusing failures because the evaluation function is not monotonous;

[0008] 4. For the foreground and foreground scenes, it will focus on the non-clearest point

Images