JPH0377487B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention photoelectrically converts an optical image projected by an imaging lens onto a light-receiving element row in which a plurality of photoelectric conversion elements are arranged, and calculates the output. In the method of detecting the focus of the imaging lens by deriving the evaluation function value of the sharpness of the projected image through processing, if the evaluation value is small, the calculation processing method is changed. Accordingly, the present invention relates to a focus detection method that can always detect focus with high precision.

As a conventional focus detection method of this type, there is, for example, a method described in Japanese Patent Application Laid-open No. 57809/1983, filed by the applicant.

This method converts photoelectric signals generated from a pair of light-receiving element rows arranged at a certain optical distance from the imaging lens into digital signals for each element, and then Among the absolute values of the differences between the output signals of adjacent elements, the sum up to the Nth largest value from the maximum value is calculated for each light-receiving element column, and the sum is calculated as the sum F-1a of the first light-receiving element column. In this method, a position where the sum F-2a of the second light receiving element array is equal is detected as a focused point.

This method uses a pair of light-receiving element arrays because it is necessary to automatically control the focusing of the imaging lens, but if the focus is simply detected, one light-receiving element array is sufficient, so the present invention In order to clarify the problem in comparison with the above, the case of a single row of light receiving elements will be explained.

To simplify the explanation, it is assumed that the evaluation function takes only the maximum absolute value of the difference between the output signals of adjacent light receiving elements.

Figure 1 shows the relationship between the image space of the subject image projected onto the light-receiving element array by the imaging lens and the light intensity distribution of the image, with the horizontal axis representing the light-receiving element arrays P1 to P10, and the vertical axis representing each light-receiving element array. This figure shows the output of the element. A in the same figure shows a case where the reflection density change of the subject changes stepwise, and the outputs of the light receiving elements P5 and P6 adjacent to the changing part have a maximum (Imax) and a minimum (Imin) light intensity distribution. It is becoming.

In this case, the evaluation function F-1a of the light-receiving element array, in which the evaluation function value is the maximum value of the absolute value of the difference between the outputs of adjacent light-receiving elements, is such that the subject image is It shows a sharp peak in the vicinity of the focused plane, and the peak value is [F-
1a (Peak)=|Imax−Imin|], which shows a large value.

However, if the reflection density of the subject changes gradually (this is called a "gradient pattern").
has an image light intensity distribution as shown in FIG. 1B. Evaluation function F- of the light receiving element array in this case
1b has a peak near the focus as shown in Figure 2, and its value (F-1b=|I _(P6) −I _(P7) |)
is a much smaller value than F-1a mentioned above.

On the other hand, as shown in Fig. 2, at a place further away from the focal point, the evaluation function value remains at a certain value, and that value is due to variations in each light receiving element (P1 to P10), and their outputs are compared to analog Digital (A/
D) It is disturbed by quantization noise etc. during conversion. For this reason, it is difficult to detect focus on a greyed-out subject, and therefore, as in the conventional method described above, when a pair of light-receiving element arrays are used, the evaluation function values obtained from the output of each light-receiving element array are equal. If the focal position is set as the focal position, the evaluation function must be determined at a certain value or higher.

Figure 2 shows a step-shaped evaluation function obtained by the above evaluation method for each light-receiving element row by arranging a pair of light-receiving element rows at a fixed distance apart along the optical axis of the imaging lens. The case of light intensity distribution image is F-
1a and F-2a, and the graded light intensity distribution images are shown as F-1b and F-2b, respectively. Then, a position where the evaluation function values obtained for each light receiving element array are equal is detected as a focal position of the imaging lens.

For general objects, the boundary between two objects, the boundary between hair and the face, etc., is a step-like object, while parts of the face that produce gradual shadows, such as the nose and cheeks, are graded objects. It can be said to be the subject.

In the conventional example described above, only the maximum absolute value of the difference between the output signals of adjacent elements is taken as the evaluation function value. In order to optimize the change in the amount of defocus, the finer the pitch of the light receiving elements, the smaller the difference between the outputs of adjacent light receiving elements. In the case of a graded subject image, it is difficult to detect focus.

SUMMARY OF THE INVENTION An object of the present invention is to eliminate the above-mentioned drawbacks and provide a focus detection method that can always accurately detect the focus state even if the light intensity distribution of the subject image projected onto the light receiving element array is graded. be.

That is, the present invention projects an object image using an imaging lens onto a light-receiving element array consisting of a plurality of photoelectric conversion elements, and calculates the evaluation function value of the sharpness of the object image based on the photoelectric conversion signal. In order to detect the focal state of the image lens, the same instantaneous photoelectric conversion signals of a plurality of photoelectric conversion elements in the light receiving element row are simultaneously sampled and held and converted into digital signals in parallel, and based on the digital signals, the basic An evaluation function value is calculated by the evaluation method and compared with a predetermined threshold value, and if the evaluation function value is larger than the predetermined threshold value, the focal state of the imaging lens is determined using the evaluation function value. is detected, and if the evaluation function value obtained by the basic evaluation method is smaller than the predetermined threshold value, the evaluation function value is calculated by an evaluation method different from the basic evaluation method based on the digital signal. The present invention is characterized in that the focal state of the imaging lens is detected.

The method of the present invention will be described in detail below.The basic evaluation method of the present invention may be any method, but for example, a subject image is projected by an imaging lens onto a light receiving element array consisting of a plurality of photoelectric conversion elements. , converts the output obtained by photoelectric conversion into a digital signal, and calculates the Nth output from the difference in output between adjacent photoelectric conversion elements or the largest value of the difference.
This is an evaluation method in which the evaluation function value is the sum of the values up to the largest value.

When the reflection density distribution of the subject is graded, the light intensity distribution on the photodetector array becomes as shown in Figure 1B, and the evaluation function according to the basic evaluation method is F- in Figure 2. As explained earlier, it exhibits flat characteristics with inconspicuous peaks like 1b or F-2b. Such conditions are newly shown in F-1b and F-2b of FIG. 3B.

The evaluation function value calculated in this way is compared with a threshold value K1 that is set in advance at a relatively higher level than the conventional threshold value K, and when the evaluation value is smaller than the threshold value K1, the When the evaluation value is large, the focus state is detected using the evaluation function value based on this basic evaluation method, and when it is small, the evaluation function value is determined by switching to an evaluation method different from the basic evaluation method, and the focus state is Detect the state.

For example, if an evaluation method is adopted as the basic evaluation method in which the evaluation function is the absolute value of the difference in output signals (digital signals) between adjacent photoelectric conversion elements in a photodetector array, As shown in FIG. 3A, the difference in output between photoelectric conversion elements P6 and P7 is i1, and that between P7 and P8 is i2. If the value is lower than a predetermined threshold value K1, the evaluation method is switched to a different evaluation method. If we adopt an evaluation method that uses the value as the evaluation function value, we will obtain an evaluation function with a value of i3 equal to the sum of i1 and i2, as shown in the figure, and the peak indicated by that function is It shows a larger value than before switching the evaluation method.

As another example, the basic evaluation method includes the sum of the absolute values of the differences between the output signals (digital signals) between adjacent photoelectric conversion elements in the photodetector array to the second largest value from the maximum value. If,
For example, in a subject image with a light intensity distribution as shown in FIG. Since i2 is large, the evaluation function value is the sum i3 of the absolute values of these two differences.
As is clear from the figure, in this example, the evaluation function value obtained using the basic evaluation method has already shown the same value as in the previous example. The method of the present invention compares the evaluation function value obtained by the basic evaluation method obtained in this way with a predetermined threshold value K1, as in the case of the previous example,
When it is larger than K1, the evaluation function value based on this basic evaluation method is used to detect the focus state, and
When the value is below the threshold value K1, such as F-1b or F-2b, an evaluation method different from the basic evaluation method is used, for example, as shown in Figure A, every second photodetector array is used. The evaluation method is changed to one in which the absolute value i4 of the difference between the output signals (digital signals) between the photoelectric conversion elements P5 and P8 is used as the evaluation function value. As is clear from the figure, the evaluation function value obtained in this way is much larger than that obtained by the basic evaluation method, and the evaluation function value obtained in this way is much larger than that obtained by the basic evaluation method. Since it has a sharp peak in the vicinity of the focus, it becomes possible to accurately detect the in-focus state even for gray-detailed subject images that are difficult to detect using conventional methods. In addition, in that example, as a different evaluation method, the absolute value of the difference in output signal between adjacent photoelectric conversion elements in every second photodetector array was determined, but photoelectric conversion elements located every third or more apart If the difference in output signals between the two is used, even more remarkable detection effects can be expected.

In fact, as in the examples described later, it is possible to exhibit a synergistic effect of the effects of both the basic evaluation method and the different evaluation method, and therefore it is possible to achieve an effect that exceeds expectations.

FIG. 4 is a block diagram showing the configuration of an example of a focus detection device employing the method of the present invention.

In the figure, a subject 1 is projected as an optical image onto a light receiving element array 3 by an imaging lens 2. The output of each light-receiving element in the light-receiving element row is converted into a digital signal in parallel by an analog-digital (A/D) conversion circuit 4, and then sequentially taken into the central processing unit 5, where the output of the adjacent light-receiving element is converted into a digital signal. Calculates the absolute value of the difference between the outputs of the light receiving elements, calculates the specified basic evaluation function, and compares it with a predetermined threshold.
When the calculation result is less than or equal to the threshold value, the memory selection circuit 30 is controlled to select every N digital memories for storing the digital output of each photoelectric conversion element provided in the A/D conversion circuit 4. The central processing unit 5
The basic evaluation method is switched to a different evaluation method by inputting the above-mentioned basic evaluation method. The central processing unit 5 also includes a control signal generation circuit for an optical system drive control circuit 29 for controlling an optical system drive device 8 for displacing the imaging lens 2 in the direction of the arrow along the optical axis; In Figure 3
It includes a constant memory for determining the threshold value indicated by K1 and a comparison circuit for comparing the calculated value with a predetermined level read from the memory. Note that the in-focus state is displayed on the display device 7. Note that in a simple device, the optical system is driven by the photographer using the display device 6.
The optical system drive device 8 and the optical system drive control circuit 29 may be omitted so that the optical system drive device 8 and the optical system drive control circuit 29 are manually operated according to an instruction.

5 to 8 are configuration diagrams of embodiments of the present invention. FIG. 5 shows the arrangement rows of light receiving element rows,
This is an example of an eye reflex camera, in which a pair of light receiving element arrays 9 and 10 are arranged in parallel on a substrate 11, and a part of the subject image is captured by the imaging lens 2. Through the return half mirror 12, a mirror 13, a half mirror 14, and a mirror 15 are used to generate a constant value before and after a surface that is optically conjugate with the film surface (not shown), which is the predetermined focal plane of the imaging lens 2. The light is projected onto each of the light receiving element arrays 9 and 10 arranged at an optical distance apart. Note that 16 is a focusing plate, 17 is a pentaprism for viewfinder, and 18 is a shutter curtain.

For the subject image projected onto each light-receiving element row, the evaluation function value is calculated for each light-receiving element row according to the intensity distribution of the light, as explained earlier, and the evaluation function value of each light-receiving element row is made to match. What is necessary is to move the imaging lens 2 along the optical axis.

That is, F-1b and F-2B in FIG. 3 are evaluation functions obtained by the method of the present invention from the output of each light-receiving element array, and therefore, the position where the values of both are equal is the summation of the imaging lens. It becomes the focal position. If the performance of each light-receiving element array in a pair is equal, the position is as follows:
This is an intermediate position between the half mirror 14 and the mirror 15 in FIG.

FIG. 6 is a block diagram mainly showing the A/D conversion circuit 4 as an example of a circuit configuration for implementing the method of the present invention. In the figure, the outputs of a pair of photodetector arrays 9 and 10 are alternately switched by a selection circuit 21 operated by a control signal from a central processing unit 5, and the outputs of each photoelectric conversion element forming the photodetector array 9 or 10 are alternately switched. The sample and hold circuits 22-1, 22-2, .

Each sample hold circuit 22-1, 22-2...
By applying a sampling signal and a readout signal from the central processing unit 5 to 22-K, the output signal of each of the photoelectric conversion elements is simultaneously sampled, held, and output according to the timing of the sampling signal. It is configured.

Each sample hold circuit 22-1, 22-2...
The output signal of 22-K is sent to each of the following comparators 23-1,
23-2...23-K, and compared with the output signal whose level changes stepwise from the D/A conversion circuit 24 in each of the comparators 23-1, 23-2...23-K, Take the comparison output and
The signal is guided as a storage control signal to each of the digital memories 25-1, 25-2, . . . 25-K provided corresponding to each of the comparators.

Each digital memory 25-1, 25-2...25
-K and the D/A conversion circuit 24 are led to a digital signal from a counter 26 which outputs the input pulse signal from the central processing unit 5 in the form of, for example, a 4-bit parallel binary code signal. There is.

Therefore, each comparator 23-1, 23-2...23-
The output level of the D/A conversion circuit 24 input to K is determined by the counter 36 input to each digital memory.
Since this corresponds to the same level as the level represented by the digital output signal, each comparator 23-
1, 23-2...23-K output timing,
Each digital memory 25-1, 25-2...25-
If the digital signal from the counter 26 is stored in K, the output signal from each photoelectric conversion element of the light receiving element array 9 or 10 is stored in the corresponding digital memory 25-1, 25-2...25.
-K can be converted into a 4-bit binary code digital signal and stored in parallel.

Moreover, the timing of the sample hold is
The output of each comparator 23-1, 23-2,...23-K is led to the OR circuit 27, the logical sum is taken, and the output is led to the central processing unit. A sample pulse with timing is generated from the central processing unit, and this pulse is sent to the sample and hold circuits 22-1, 22-2...
The photoelectric conversion signals of the respective photoelectric conversion elements at the same instant are simultaneously sampled and held using a sampling pulse added to 22-K. On the other hand, each comparator 23-1,
The outputs of the circuits 23-2, .

The digital memories 25-1, 25-2, . . .
25-K, the central processing unit compares the evaluation value determined by any specified basic evaluation method with a predetermined threshold, and if it is less than the threshold, a different evaluation method is used. In calculating the evaluation value by switching to , this embodiment is characterized in that the address of the digital memory to be read can be specified by the memory selection circuit 30 according to the calculation result.

FIG. 7 is a block diagram showing an example of the memory selection circuit 30, in which the counter 31 counts from zero to the number K of digital memories according to a control signal from the central processing unit 5, and then counts back to zero. The digital memory 25-
1, 25-2, . . . 25-K are sequentially called. However, between the counter 31 and each digital memory 25-1, 25-2...25-K, there is a first decoder 32 and an AND gate group 3.
3-1, 33-2,...33-K are inserted, and digital memories 25-1, 25-2...25-K are inserted.
I am trying to limit the number of calls.

That is, each AND gate 33-1, 33-2
One input of ...33-K is led to the output of the first decoder 32, and the other input is connected to the output of the down counter 34 that counts the comparison output from the central processing unit 5, the details of which will be described later. Decipher and
An output from a second decoder 35 outputting a parallel digital signal consisting of C, B, and A is directed to enable AND gates 33-1 to 33-K to be selectively operated.

For example, when the read signal from the first decoder 32 for the digital memories 25-1 to 25-K is sequentially applied to each AND gate 33-1 to 33-K, the outputs C and B of the second decoder 35・When A is 1, 1, 1, all AND gates 33-1 to 33-K output, and when C, B, and A are 0, 1, 1, every other AND gate outputs.
AND gates 33-1, 33-3, 33-5... output, and when the above C, B, and A become 0, 0, 1, every third AND gate 33-1, 3
3-5, 33-9, . . . are configured to output respectively.

In this way, the signals for reading from the first decoder 32 are sequentially selected according to the designations of the second decoder 35, and the digital memory 2 of FIG.
5-1 to 25-K sequentially or selectively, and sequentially read out to the central processing unit 5.

Digital memory 25 read out in that case
-1 to 25-K are only those specified by the AND gate.

Therefore, the outputs of the light receiving element array read into the central processing unit 5 are only every L photoelectric conversion elements among the adjacent photoelectric conversion elements according to the command from the memory selection circuit 30, and the central processing unit 5
In this case, the difference in output signals between every L photoelectric conversion elements is calculated, and the result is that the evaluation function value is switched to a different evaluation method.

In this example, as a basic evaluation method, the Nth
Since a method is adopted in which the sum up to the largest value is used as the evaluation function value, the central processing unit 5 includes a maximum value detection circuit including a plurality of memory blocks and an adder.

Let us now consider the evaluation function when the subject image of the light receiving element array 9 or 10 has a graded light intensity distribution as shown in FIG. 3A. For the sake of simplicity, the evaluation function is based on the digital memories 25-1 and 25-1, which are sequentially input to the central processing unit 5.
Only the maximum absolute value of the difference in output between the output signals of 5-2, . . . 25-K is taken.

That is, digital memories 25-1, 25-
2...The signal read out from 25-K is processed by the shift register 41 and the subtraction circuit 42 in the central processing unit 5 of FIG. is calculated,
The maximum detection circuit 43 includes a plurality of memory blocks. The maximum value detection circuit 43 is configured to detect N by a control signal from the control circuit 44, and detects the absolute value up to the Nth largest value from the maximum value from among the absolute values of the differences in output signals between the light receiving elements. The structure is configured so that a value can be selected and detected, and when N is 1, the basic evaluation method is to evaluate the absolute value of the difference between simple output signals between adjacent photoelectric conversion elements in the light receiving element array. This corresponds to the case of using a function value, and only the absolute value of the maximum value in the output of the subtraction circuit 42 is selected by the maximum value detection circuit 43.

Now, if N is set to 2, the maximum value selection circuit 43 selects the values from the maximum value to the second value among the outputs of the subtraction circuit 42, and sequentially leads them to the adder 45. will be added.

The addition output obtained by the adder 45 is
input to the comparator circuit 46, predetermined threshold value K1 (third
(see figure) is compared with a reference value from a constant memory 47 in which the constant value is stored. Since the comparator circuit 46 is set to produce an output when the output of the adder 45 is lower than the threshold, if the output of the adder 45 is lower than the threshold;
Since the down counter 34 of the memory selection circuit 30 is counted down by one step and the outputs C, B, and A are changed from 1.1.1 to 0.1.1, the AND of the memory selection circuit 30 is The gates output every other gate, and therefore the digital memory is also addressed to be read every other gate, and its contents are sequentially read into the central processing unit 5. Since the central processing unit 5 performs arithmetic processing on the difference between adjacent signals read in this way, the evaluation function value is obtained by an evaluation method different from the basic evaluation method described above, and the evaluation The function value corresponds to the difference between the outputs of the digital memory corresponding to the outputs of the L addressed light receiving elements, that is, every other light receiving element.

Furthermore, even if every other digital memory is read, the evaluation function value still falls below the predetermined threshold value.
If it is less than K1, the output of the comparison circuit 46 is generated again, so the down counter 34 is further counted down and every second AND gate is operated. In this embodiment, at most every third AND gate is operated, but the number may be increased and the same operation may be repeated.

On the other hand, if the evaluation function value obtained by the basic evaluation method is equal to or higher than the threshold value K1, the comparison circuit 4
Since there is no output from 6, the output of the down counter 34 will maintain the state of 1, 1, 1, and all the digital memories 25-1 to 25-K will be sequentially read out and added to the central processing unit, and the maximum The in-focus state is detected based on the basic evaluation method set by the value detection circuit 43 and the adder 45.

In this embodiment, it is assumed that, for example, the maximum value detection circuit 43 is set to select the Nth largest value from the maximum value of the input signal, and L pieces of digital memories 25-1 to 25-K are selected. If the data are addressed and sequentially imported into the central processing unit 5, the synergistic effect of the former basic evaluation method and the latter different evaluation method,
Compared to the conventional case where the absolute value of the difference in output signals between adjacent photoelectric conversion elements is used as the evaluation function value, almost NL
Since the evaluation function value has twice the value, Figure 3B
As shown in FIG.
1B and F-2B, it shows a significantly large value near the in-focus point.

In this embodiment, a signal corresponding to the detected focusing state is led to the display determination circuit 48, and the quality of the imaging is displayed on the display device 7.

The above-mentioned arithmetic processing is performed by the central processing unit 5 for the pair of light-receiving element arrays shown at 9 and 10 in FIG.
If the evaluation functions are performed alternately according to the commands of 1 and 2, it is possible to compare the evaluation function values in the so-called front focus and rear focus states of the imaging lens 2 shown in FIG. By moving along the axis, the in-focus position can be detected accurately.

Although the connections between the control circuit 44 and each functional block are not shown in FIG. 8, each functional block is configured to be controlled by a control signal from the control circuit 44.

According to the above embodiment, when the evaluation value obtained by the basic evaluation method is less than or equal to a predetermined threshold value, the central processing unit 5 of the digital memory reads every L pieces.
Since the evaluation function is the maximum absolute value of the difference between adjacent signals, if L is set to 1, the result is one of the light intensity distributions of the image received by the light receiving element array 9 or 10. This means that the maximum absolute value of the difference between the signals at intervals is set as the evaluation function, and the evaluation function value is calculated using an evaluation method different from the basic evaluation method.

As a result, as shown by F-1B or F-2B in Fig. 3, the evaluation value F-1B was determined using the basic evaluation method, that is, simply calculating the maximum absolute value of the difference in output signals between adjacent light receiving elements. This value is much larger than that of F-2b, and the in-focus position can be detected even in cases where detection would not have been possible using conventional methods.

In addition, as explained in the previous example, if you calculate the difference between every other value of the light receiving element output and use this as the evaluation value, if detection is still difficult, use the down counter (see Figure 7). By advancing step 34) one by one, the outputs of every second or third light receiving element are sequentially input to the central processing unit. The focal position of the image lens can be easily detected. In addition, using the conventional method, the sum of the absolute values of the output differences between adjacent light receiving elements up to the Nth largest value from the maximum value is calculated for each light receiving element column, and Japanese F-
Due to the magnitude relationship between 1a and the sum F−2a of the second column, F
In the method of detecting the in-focus point at the position where -1a and F-2a are equal, if the focus detection is optimal for subjects with step-like reflection density, Even when in-focus detection becomes impossible, the method of the present invention combines the basic evaluation method and a different evaluation method, and detects the in-focus state based on the evaluation method that has a synergistic effect between the two. Since it is a method of focusing on focus detection, it can achieve a much superior focus detection effect compared to conventional methods.

In addition, since the same instantaneous photoelectric conversion signals of multiple photoelectric conversion elements in the light receiving element row are simultaneously sampled and held and converted into digital signals in parallel to obtain the evaluation value, A/D conversion is shortened. It is possible to use signals with no time lag between photoelectric conversion elements even when the subject moves, so accurate evaluation values can always be obtained in a short time, and the focus The state can always be detected accurately.

It goes without saying that the basic evaluation method and evaluation methods different from the basic evaluation method in the method of the present invention are not limited to the evaluation methods described in the above-mentioned Examples.

[Brief explanation of drawings]

1 and 2 are a subject image light intensity distribution diagram and an evaluation function curve diagram for explaining the conventional method, FIG. 3 is a diagram for explaining the action of the method of the present invention, and FIG. FIG. 5 is a block diagram showing the configuration of an example of a focus detection device employing the method of the present invention. FIG. FIG. 7 is a block diagram showing the circuit configuration of the embodiment with a focus on the A/D conversion circuit. FIG. 7 is a block diagram of an example of the configuration of the memory selection circuit in the embodiment of the present invention. FIG. FIG. 2 is a block diagram of a configuration example of a central processing unit in an example. 1... Subject, 2... Imaging lens, 11... Board, 12... Quick return half mirror,
13, 14...half mirror, 15...mirror,
16... Focusing board, 17... Pentaprism, 1
8...Shutter curtain, 33-1 to 33-K...
AND gate.

Claims (1)

[Scope of Claims] 1. An object image is projected by an imaging lens onto a light receiving element array consisting of a plurality of photoelectric conversion elements, and an evaluation function value of the sharpness of the object image is calculated based on the photoelectric conversion signal. In detecting the focal state of the imaging lens, the same instantaneous photoelectric conversion signals of the plurality of photoelectric conversion elements in the light receiving element row are read out and converted into digital signals, and based on the digital signals, the basic evaluation method is carried out. A first evaluation function value is calculated by the evaluation method of 1 and compared with a predetermined threshold, and if the first evaluation function value is larger than the predetermined threshold, the first evaluation detecting the focal state of the imaging lens using the function value, and when the first evaluation function value is smaller than the predetermined threshold value, the first evaluation method, which is the basic evaluation method, is performed based on the digital signal; A focus detection method that calculates a second evaluation function value by a second evaluation method different from the evaluation method and detects the focus state of the imaging lens using the second evaluation function value, The first evaluation method includes determining, as the first evaluation function value for the light receiving element array, a value that is up to the Nth largest value from the absolute value of the difference between digital signal outputs of adjacent photoelectric conversion elements. The second evaluation method is a method of calculating the sum of digital signal outputs of every L adjacent photoelectric conversion elements as the second evaluation function value for the light receiving element array. A focus detection method, characterized in that it is a method of calculating the sum of values up to the Nth largest value among the values.