Phase difference detection methd, phase difference detector, distance finder and imaging appts.

Abstract

An attempt is made to simplify compensation calculations in a phase difference detecting device for detecting a phase difference between images formed on a pair of optical sensor arrays. When the compensation effective row inspection unit 5d checks out that the compensation to the paired image data rows (IL, IR) is effective, the left and right difference compensation unit 5e calculates: the difference Max and Min of the maximum value of each pair of image data rows After summing the difference Min, divide it by 2 as the compensation amount, and use the calculated compensation amount to compensate the difference between IL and IR in the memory 5b. According to the compensated IL and IR, the correlation calculation unit 5f performs correlation calculation, and the maximum correlation detection unit 5g detects the maximum correlation level, and performs interpolation calculation according to the maximum correlation level interpolation calculation unit 5h, and the phase detection unit 5i detects the phase Difference.

Description

technical field [0001] The invention designs a phase difference detection method, a phase difference detection device, a distance measuring device and an imaging device. Background technique [0002] In the past, in devices such as auto-focus cameras, when using a so-called passive system to focus on an object to be photographed, if it is a non-TTL camera, the distance to the object is detected using the object image that does not pass through the shooting lens, and then based on The detected distance to the subject controls the shooting lens; and if it is a TTL camera, the object image obtained through the shooting lens is used to detect the offset relative to the focus state, and then the shooting lens is controlled according to the detected offset. The following will refer to Figure 7A The principle of the above operation is explained. [0003] In this figure, a pair of lenses 1a and 1b are separated from each other by a predetermined baseline length b, and a pair of op...

AI Technical Summary

Problems solved by technology

[0008] However, in the technology described in the above-mentioned No. 3230759 patent (JP-A-5-264892), in order to obtain the compensation value, it is necessary to perform complex calculation of the difference between the average values ​​of the individual image signals

[0009] Moreover, in the above-mentioned patent, since the compensation value is only calculated from the difference of the average value of each image signal, it may not be possible to perform high-precision compensation.

like Figure 8 As shown, when the image signal ILX exists in the image signal IL but not in the image signal IR, and the image signal IRX exists in the image signal IR but not in the image signal IL, if only a pair containing If the difference between the image signal average values ​​of these image signals ILX and IRX is used to calculate the compensation, since the compensation value is affected by different image signals ILX and IRX, high-precision compensation cannot be performed.

Images