An auto-focusing method suitable for optical field camera

Abstract

The invention discloses an automatic focusing method suitable for an optical field camera. Firstly, a two-dimensional image is obtained under any initial image distance and decoded into a four-dimensional optical field according to calibration information, and a four-dimensional Fourier transform is carried out on the four-dimensional optical field by adopting a fast Fourier transform algorithm toobtain a light field spectrum. Secondly, the slice function with different slope is used to reduce the dimension of the four-dimensional light field spectrum, to obtain the two-dimensional spectrum of each layer depth, to extract the on-axis spectrum of the two-dimensional spectrum, to synthesize one-dimensional curve, and to calculate the ratio of high frequency energy to low frequency energy after high frequency amplification. Then the local maximum point of the ratio curve is detected, the slope of the local maximum point is obtained, and the arithmetic average of the maximum and minimum slope is calculated. Finally, the distance between the main lens and the microlens array is adjusted according to the deviation between the average value and 0. This method utilizes the spectrum to calculate the depth level of the object quickly, and adjusts the initial focusing depth to make the depth of field of the optical field camera coincide with the actual depth range of the scene effectively.

Description

technical field [0001] The invention relates to an automatic focusing method suitable for a light field camera, which is an optimized focusing method for automatically adjusting the distance between a main mirror and a microlens array to optimize the focus depth of a light field camera so that the depth of field of the light field camera and the actual scene depth range are effectively overlapped method. The invention belongs to the technical field of automatic focusing in an optical imaging system. Background technique [0002] In the field of optical imaging, light-field cameras, as a new computational imaging device, have attracted extensive attention because of their outstanding depth-of-field extension and distance detection capabilities. Because the imaging parameters of the light field camera are difficult to obtain when the distance between the primary mirror and the microlens array changes, and the corresponding calibration data cannot be determined, most light fie...

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is: in the field of light field imaging, autofocus is the future development trend, but the traditional sharpness evaluation based on a single depth image fails to fully consider the depth of field expansion ability of light field imaging, and cannot fully utilize the light field camera The advantages

Images