Camera focusing method and focusing method based on recurrent neural network

Abstract

The embodiment of the invention provides a camera focusing method and a focusing method based on a recurrent neural network. The method comprises the steps of collecting a current image collected by a camera and the current position of a lens of the camera, wherein the current position is the position of the current lens relative to an image sensor of the camera; dividing the current image into a plurality of image areas as target image areas, and calculating a target definition evaluation value of each target image area; inputting the plurality of target definition evaluation values and the current position into a pre-trained focusing model to obtain target prediction values of a plurality of preset clear point states corresponding to the current image; and adjusting the position of the lens based on the target prediction values of the plurality of preset clear point states so as to enable the lens to reach the clear point. By applying the technical scheme provided by the embodiment of the invention, the image picture oscillation in the focusing process can be reduced, the focusing speed is improved, and the focusing accuracy is improved.

Description

technical field [0001] The present application relates to the technical field of video surveillance, in particular to a camera focusing method and a focusing method based on a cyclic neural network. Background technique [0002] When the camera collects images of different application scenarios, it is necessary to focus the camera and adjust the distance between the camera lens and the image sensor to ensure the clarity of the image. [0003] At present, the contrast-type autofocus method is mainly used to focus the camera, which specifically includes: collecting the current frame image, calculating the sharpness evaluation value of the current frame image, comparing the sharpness evaluation value of the current frame image with the sharpness evaluation value of the previous frame image Value, to determine whether the sharpness evaluation value of the image captured by the camera reaches the maximum value; if it does not reach the maximum value, then drive the motor to adjus...

AI Technical Summary

Problems solved by technology

[0004] In the above camera focusing method, only the sharpness evaluation value of the current frame image and the sharpness evaluation value of the previous frame image are considered

However, in practical application scenarios, the signal-to-noise ratio of the image is low due to interference from external factors.

This makes it easy to cause image shocks and slow focusing speeds during the process of focusing the camera using the above-mentioned camera focusing method.

[0005] In addition, in the above-mentioned camera focusing method, the method of calculating the sharpness evaluation value of the entire image is relatively simple, and it is easy to mix normal features and noise features together, resulting in the inability to accurately predict the clear point, or wrongly determine the clear point

Images