JP3158772B2 - Focus information detection device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an in-focus information detecting apparatus, and more particularly to an in-focus information detecting apparatus in which image information relating to a subject is formed on an imaging means surface by an imaging optical system, and an autocorrelation of an output signal from the imaging means is obtained. The present invention relates to a focus information detecting device suitable for, for example, a video camera or the like, which detects focus information and performs a focusing operation of the photographing optical system.

[0002]

2. Description of the Related Art Conventionally, various types of light-receiving type focus information detecting devices which detect focus information using image information based on a light beam having passed through a photographing optical system (photographing lens) have been proposed.

[0003] Among them, in a focusing information detecting device which forms image information on an image pickup means (image pickup element) surface by a photographing optical system and detects a focus shift from an output signal (image information) from the image pickup means, A ranging frame is set near the center of the photographing area, and frequency components having relatively high frequencies are separated and extracted from the data in the ranging frame by a band-pass filter.

While moving the imaging position of the taking lens in the direction of the optical axis, the change of the separated high frequency component is measured, and the position where the value becomes the maximum value is detected as the focus position. .

[0005]

In a conventional apparatus for detecting high frequency components of an output signal from an imaging means and detecting focusing information of the imaging lens, the relative positional relationship between the imaging lens and the imaging means is determined. It is necessary to make various changes and use a plurality of pieces of image information obtained at this time.

For this reason, there is a problem that it takes much time to obtain the focusing information, and it is difficult to obtain instantaneous focusing information (ranging information).

According to the present invention, image information is formed on an image pickup means surface by a light beam passing through a predetermined opening on a pupil plane of a photographing lens, and an autocorrelation of the image information is appropriately obtained, whereby the photographing lens and the photographing lens are formed. An object of the present invention is to provide a focus information detecting device suitable for a video camera or the like, which can detect focus information with high accuracy and high speed from one piece of image information without performing relative movement with respect to an image pickup means. And

[0008]

According to a first aspect of the present invention, there is provided a focusing information detecting apparatus in which a light shielding member having three or more openings is disposed in an optical path of a photographing optical system so as to be insertable and removable. Image information is formed on the surface of the imaging means by a light beam passing through the opening of the light-shielding member, and a third or higher order autocorrelation of an output signal from the imaging means is calculated by the calculation means. The feature is that the direction and the amount of blur are obtained.

According to a second aspect of the present invention, in the first aspect of the present invention, the calculating means has a shape similar to the opening on the imaging surface of the imaging means according to the shape and arrangement of the three or more openings of the light shielding member. It is characterized in that the autocorrelation between pixels having the following arrangement is obtained.

[0010]

FIG. 1 is a block diagram of a main part of a first embodiment of the present invention.

In the drawing, reference numeral 11 denotes a photographing optical system, and a light-shielding plate having a plurality of openings is provided in the optical path so as to be insertable and removable. Reference numeral 12 denotes an image pickup unit (image pickup element), which converts image information formed by the photographing optical system 11 into an electric signal.

13 is a sample and hold circuit, 14 is a color separation and gamma conversion circuit, 15 is a matrix circuit for obtaining luminance and color difference outputs, 16 is an A / D conversion circuit for converting analog signals to digital signals, 17 Is arithmetic means,
It consists of a correlation operation processing block for performing an autocorrelation operation.

Reference numeral 18 denotes a focus information detection block for detecting focus information based on the result of the correlation operation, reference numeral 19 denotes a lens driving circuit, and reference numeral 20 denotes a clock generator for supplying a clock to each block.

In FIG. 1, image information from a subject that has passed through a photographing optical system 11 forms an image on an image sensor 12.

Here, the image information converted into an electric signal is sampled and held by a sample and hold circuit 13, and then converted into an RGB signal by a color separation and γ conversion circuit 14. The RGB output is converted into a luminance signal and a color difference signal by the matrix circuit 15 and output.

On the other hand, the green signal (G signal) output from the color separation / γ conversion circuit 14 is converted into digital data by an A / D converter 16, and a correlation operation circuit 17 calculates an autocorrelation function. The focus information detecting unit 18 obtains focus information, that is, distance information of the subject based on the information.

Based on the distance information of the subject obtained by the focus information detecting section 18, the lens driving circuit 19
Is moved to the in-focus position.

In this embodiment, the focusing operation of the photographing optical system 11 is performed by the above operation.

Next, the photographing optical system 11 in the present embodiment,
The operation and processing of the correlation operation circuit 17 and the focus information detection unit 18 will be described.

FIG. 2 is a schematic view showing the relationship between the projection optical system 11 and the image pickup device (sensor) 12 according to the present invention. In the figure, 1 is a taking lens, 2 is a plurality of openings 2a, 2
Reference numeral 3 denotes a position of the sensor in the back focus state, 4 denotes a position of the sensor in the focused state, and 5 denotes a position of the sensor in the front focus state.

The light shielding plate 2 has three openings 2a, 2b, 2c.
The light passing through each opening forms an image on the sensor. When the sensor is at the position 3, light passing through the opening 2a is in the region 3a, and light passing through the opening 2b is in the region 3b.
Then, the light passing through the opening 2c forms an image on the area 3c.

When the sensor is at the in-focus position, the three light beams form an image at one point. When the sensor is at the position 5, light passing through the opening 2a forms an image on the area 5a, light passing through the opening 2b forms an image on the area 5b, and light passing through the opening 2c forms an image on the area 5c.

FIG. 3 shows the state of image information on each image plane of FIG.

In FIG. 3, reference numerals 3, 4, and 5 represent the images of the sensors provided at positions 3, 4, and 5, respectively, in FIG. As is clear from FIG. 3, the image of the light passing through the openings 2a, 2b, and 2c has a point-symmetrical relationship between the positions 3 and 5 exactly.

Therefore, in the present embodiment, if the orientation of the pixel on the sensor is the same as that of the light shielding plate 2, it is determined that the pixel is in the back focus state, and if the pixel is rotated 180 degrees, it is determined that the pixel is in the front focus state.

Next, a correlation operation section 1 as an operation means in FIG.
7 will be described with reference to FIG.

FIG. 4A is a view showing the shape of the light shielding plate 30.
FIG. 4B is a diagram showing a positional relationship between the openings a, b, and c of the light-shielding plate 30, and FIG. 4C is a diagram showing a positional relationship between pixels on the image sensor 12 for performing an operation.

As shown in FIG. 4A, the light shielding plate 30 has three holes (openings) a, b, and c. The positional relationship among the three openings a, b, and c is, for example, as shown in FIG.

As shown in FIG. 4B, when the grids are written in the horizontal and vertical directions, the three openings are arranged in the horizontal direction at equal intervals in the order of b, a, and c from the left. In the vertical direction, they are also arranged at equal intervals in the order of c, b, a from the top. Here, it is assumed that the horizontal and vertical grid pitches are equal.

When the openings a, b, and c are in such a positional relationship, the image for the point light source is similar to the image when the back focus is set, and the image is in the same direction as viewed from the subject side. When the front focus is set, the image is similar to this and rotated 180 degrees when viewed from the subject side.

Since a general image can be regarded as an infinitely superimposed point light source having different brightness, the image on the sensor is also superimposed with an image similar in shape to the openings a, b, and c of the light shielding plate 30. It will be.

Therefore, when the camera is not in the focused state, the correlation between the pixels having this positional relationship becomes stronger. Also, the size of the blurred image changes in proportion to the magnitude of the focus shift while maintaining a similar relationship to the openings a, b, and c.

Therefore, when calculating the correlation between the image of the opening b and the image of the opening c based on the image of the opening a, the following calculation may be performed.

As shown in FIG. 4C, when the coordinates of the pixels in the horizontal direction are represented by i and the coordinates of the pixels in the vertical direction are represented by j, the reference point x (i, j) is located at the opening b. The corresponding image exists on the line of the arrow 31, and the image corresponding to the opening C exists on the line of the arrow 32. (In the case of a rear focus, the position is before the arrow and in the case of the front focus, after the arrow.) Therefore, the calculation of the autocorrelation is performed with respect to the reference pixel as a pixel on the extension of the arrow 31 and a pixel on the extension of the arrow 32 with respect to the reference pixel. It is sufficient to calculate only for pixels.

[0035]

(Equation 1) Equation (1) shows the equation for calculating the autocorrelation. When the autocorrelation C (n, m) is obtained based on the equation (1), the result is as shown in FIG.

FIG. 5 shows the correlation of the blurred image with respect to the point light source. FIG. 5A shows the blurred image at the time of back focus.
(B) is a calculation result of the correlation with respect to the image of FIG.
FIG. 5C is a blurred image at the time of the front focus, and FIG.
It is the calculation result of the correlation about the image of (C). FIG.
The peak 51 appearing at the center in (B) and (D) is m =
It is the origin at 0 and n = 0, and appears regardless of the deviation amount. Apart from this central peak, a peak 52 appears in the negative region for both n and m in FIG. 5B, and a peak 53 appears in the positive region for both n and m in FIG. 5D.

As described above, the region where the peak appears differs between the front focus and the rear focus.

The distance between the peaks 52 and 53 from the origin is
It corresponds to the amount of defocus. If the shape of the blur, that is, the shape of the opening is known in advance, the relationship between n and m is determined in one way, so that it is possible to express the correlation with either one of the variables.

In this case, since it is known that a peak occurs on the line of m = n, it is sufficient to calculate only on the line of m = n.

FIG. 6 shows values on the line of m = n.

In FIG. 6, FIG. 6 (A) is replaced with FIG. 5 (B).
FIG. 6B corresponds to FIG. 5D, respectively. Peak 51,
52, 53 correspond to the same numbers.

FIGS. 5 and 6 have described the case of a point light source. However, cases other than the point light source will be described with reference to FIG.

FIG. 7A shows a state in which focus is achieved, and FIG.
FIG. 7B is an image of the state of the rear focus, and FIG. 7D is an image of the state of the front focus. 7 (C) and 7 (E) show the results of calculating the correlation for the images of FIGS. 7 (B) and 7 (D) in the same manner as in FIG.

In FIG. 7C, two peaks 72 and 73 are generated in addition to the peak 71 at the origin. FIG.
In (E), two peaks 74 and 75 are generated in addition to the peak 71 at the origin.

In the case of FIG. 7 (C), n is the negative peak 7
2 is higher than the peak 73.

Conversely, in the case of FIG. 7E, the peak 75 on the positive side of n is higher than the peak 74.

As described above, even in the case of a point light source other than the point light source, by obtaining the maximum point other than the origin, the amount and direction of the out-of-focus can be known.

The focus information detector 18 shown in FIG. 1 obtains the peak position excluding the origin of the autocorrelation function thus obtained, and obtains the value of the variable n at the time of the maximum peak. Since the value of n is proportional to the amount of focus shift, the amount of lens movement necessary for focusing is determined from this value, and the data is sent to the lens drive circuit 19.

Then, the focusing lens is driven by the lens driving circuit 19, thereby performing the focusing operation.

Next, a second embodiment of the present invention will be described.
The main components of the configuration requirements are substantially the same as those shown in FIG.

This embodiment is different from the first embodiment shown in FIG. 1 in the operation processing method performed by the correlation operation circuit 18.

In the correlation operation circuit 18 of this embodiment, the autocorrelation is obtained after the image signal has passed through a prefilter (digital filter).

Next, the features of this embodiment will be described with reference to FIGS.

FIG. 8 shows the coefficients of the digital filter Q. When the light shielding member 30 shown in FIG. 4A is used, the output signal from the image sensor 12 is subjected to the digital filter Q shown in FIG. 8, and then the same correlation operation as in the first embodiment is performed.

Here, the output signal of the image sensor 12 is x
(T), the signal y (t) that has been subjected to the digital filter Q
Since this digital filter Q is a high-pass filter, y (t) has a positive value and a negative value. Since the calculation of the correlation only needs to be performed for signals of the same polarity, the calculation is performed only for the signal having the same polarity of y (t), and when y (t) is negative, the calculation result of y (t) is negative. If the absolute value is used, the peak of the correlation becomes clearer.

FIG. 9 shows the result of the operation thus obtained. The image data used for the calculation is the same as in FIG. 9 (A) is obtained from the data of FIG. 7 (B), that is, the state of the rear focus, and FIG. 9 (B) is obtained from the data of FIG. 7 (D), ie, the state of the front focus.

FIGS. 9A and 9B and FIGS. 7C and 7E.
As can be seen from the comparison, the peak of the correlation calculation result becomes sharper by using the digital filter shown in FIG. 8, and the peak can be easily identified.

In the case of FIG. 7C, there are three peaks, a first peak 71 having the largest correlation, a second peak 72 having the next largest correlation, and a third peak 73 having the third largest correlation. Although the front or rear focus is determined based on which of the second and third peaks is larger, in the case of FIG. 9A, the first peak 91 and the second peak 92 are determined.
, But the third peak does not come out clearly. Therefore, it is easy to determine whether the pin is the front pin or the rear pin.

7 (E), the second and third peaks among the three peaks 71, 74 and 75 were compared, whereas the case of FIG. 9 (B) was used. The first peak 91 and the second peak 95 appear remarkably, and the third peak does not come out clearly. Therefore, since the peak 95 is on the plus side, it can be easily determined that the front focus is on. Further, if a filter as in the present embodiment is used, the peak appears sharply, so that even if there is a correlation peak near the origin, the correlation peak can be detected.

Next, a third embodiment of the present invention will be described.

The main parts of the components are substantially the same as those shown in FIG. The present embodiment is different from the first embodiment in FIG. 1 in the arithmetic processing method performed by the correlation operation unit 18. As in the second embodiment, the correlation calculator 18 of the present embodiment obtains the autocorrelation of the image signal using a digital filter.

Next, the features of this embodiment will be described with reference to FIGS.

This embodiment improves the digital filter used in the second embodiment. In this embodiment, the coefficient matrix of the digital filter is set to “4” at the center, the coefficient at positions shifted by n pixels in the horizontal and vertical directions from the center is set to “−1”, and the other coefficients are set to “0”.

FIG. 10 shows coefficients when n = 1, n = 2, and n = 3. T is a matrix when n = 1, U is n = 2, and V is n = 3. Similarly, when n is 4 or more, the coefficient at the position shifted by n pixels is “−1”. In the case of n = 0, the coefficients are all “0” because the five coefficients overlap and 4 −1 −1 −1 −1 = 0. The value of n is the same as the variable n for the correlation operation used in the first embodiment, and the digital filter changes according to the variable of the correlation operation.

When the correlation operation is performed after the digital filter of this embodiment is applied, the results shown in FIGS. 11 and 12 are obtained.

FIGS. 11A and 11B show FIGS. 5A and 5B, respectively.
This is the result of processing for the data of (C). Calculation results when no filter is used FIG. 5 (B),
Compared with (D), in the case of FIGS.
The peak 51 at = 0 is eliminated. The maximum value is obtained at the position 52 in FIG. 5B and the position 5 in FIG.
Positions 111 and 112 corresponding to No. 3 are obtained. Since there is no peak near the origin as described above, it is easy to detect a correlation peak generated due to a focus shift.

FIGS. 12A and 12B show FIGS. 7B and 7B, respectively.
This is the result of processing the data of (D) with n = m. 7 (C), (E), 9 (A) and 9 (B) show correlation data obtained from the data of FIGS. 7 (B) and (D). = 0, since no peak occurs at the position of each maximum value
And the position 122 can be detected as a correlation peak.

As described above, according to the present embodiment, it is possible to remove the influence of the correlation strongly appearing near the origin at the time of the correlation calculation, and thus it is possible to easily detect the defocus.

[0069]

According to the present invention, as described above, image information is formed on the imaging means surface by the light beam passing through the predetermined aperture on the pupil plane of the photographing lens, and the autocorrelation of the image information is determined. By appropriately determining, it is possible to detect the blur direction and the amount of focus from one image information with high accuracy and at high speed without relative movement between the photographing lens and the imaging means. Thus, it is possible to achieve an in-focus information detecting device suitable for, for example, the following.

In particular, according to the present invention, the in-focus position can be detected only from information of one screen at a certain moment.
It is not necessary to move the photographing lens or the image sensor for focus information detection, and it is possible to achieve a focus information detection device having such an effect that speed of detection of a focus blur direction and a blur amount can be increased. it can.

[Brief description of the drawings]

FIG. 1 is a block diagram of a main part of a first embodiment of the present invention.

FIG. 2 is a schematic diagram of a main part of the photographing optical system of FIG. 1;

FIG. 3 is an explanatory diagram of image information of FIG. 1;

FIG. 4 is a schematic diagram showing a correlation calculation method according to the present invention.

FIG. 5 is an explanatory diagram of a calculation result by a correlation calculation according to the present invention.

FIG. 6 is an explanatory diagram of a calculation result by a correlation calculation according to the present invention.

FIG. 7 is an explanatory diagram of a calculation result by a correlation calculation according to the present invention.

FIG. 8 is an explanatory diagram of a digital filter according to a second embodiment of the present invention.

FIG. 9 is an explanatory diagram of a result of a correlation operation according to the second embodiment of the present invention.

FIG. 10 is an explanatory diagram of a digital filter according to a third embodiment of the present invention.

FIG. 11 is an explanatory diagram of a result of a correlation operation according to the third embodiment of the present invention.

FIG. 12 is an explanatory diagram of a result of a correlation operation according to the third embodiment of the present invention.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 photographing lens 2 light shielding plate 2 a, 2 b, 2 c opening 11 photographing optical system 12 imaging means 13 sample hold circuit 14 color separation and γ conversion circuit 15 matrix circuit 16 A / D converter 17 correlation operation circuit 18 focus information detection section 19 Lens drive circuit 30 Light shield

Claims (2)

(57) [Claims] A light shielding member having three or more openings is removably disposed in an optical path of a photographing optical system, and image information is captured by a light beam passing through the photographing optical system and the openings of the light shielding member. formed on the surface, determine the third or higher order autocorrelation of the output signal from the image pickup means by the operation means, from the calculation result, the pin
A focus information detecting device , wherein a blur direction and a blur amount are obtained. 2. The arithmetic means calculates an autocorrelation between pixels arranged in a similar shape to the openings on the imaging surface of the imaging means in accordance with the shape and arrangement of the three or more openings of the light shielding member. 2. The focus information detecting apparatus according to claim 1, wherein the focus information is obtained.