JP2692050B2 - Camera shooting condition determination device and camera - Google Patents

Description

Description: FIELD OF THE INVENTION The present invention relates to an improvement of a camera photographing condition determination device and a camera that enable more appropriate photographing conditions to be determined.

(Background of the Invention) Conventionally, for determining a shooting condition of a camera, for example, focus adjustment of a lens is performed at each distance measuring point as disclosed in an example of JP-A-58-201015. Which gives priority to short-distance information among the respective ranging information,
Those based on majority logic as shown in JP-A-59-193307, or limit the distance measuring range of distance measuring points provided in the peripheral portion of the screen as shown in JP-A-60-172008 ( Near side and far side).

However, in the proposed device, for example, it is difficult to distinguish between a subject located at a short distance and a short distance obstacle such as the ground, and it is difficult to give accurate lens drive information.

(Object of the Invention) The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a photographing condition determining device for a camera or a camera that can easily determine more appropriate photographing conditions.

(Characteristics of the Invention) In order to achieve the above object, the present invention performs an fuzzy operation by using input means for inputting information about a subject and information about the subject input to the input means as an input value of a predetermined membership function. A first computing means for performing the first computing means;
Second computing means for obtaining the barycentric value of the computation result of the computing means, photographing condition determination for deciding the photographing condition based on the information about the subject obtained from the input means and the computation result of the second computing means And a photographing condition determining device of a camera having a means.

(Embodiment of the Invention) FIG. 1 is a block diagram showing an embodiment of the present invention.
In the embodiment, for example, left, center (including substantially the center, the same applies to the following) and right points in the photographing screen specified by the photographing lenses (not shown) by the three known distance measuring units 1, 2, 3 The distance measurement is performed, and a lower analog voltage is output as the distance measurement result (DV1 to DV3) is near distance information. Then, in the arithmetic circuit 4 which receives the analog voltage from each distance measuring unit, the lens is determined by the information such as the respective distance measuring information and each difference information, the camera posture information ANG, the use / non-use information ST of the strobe, and the aperture value information AV. Drive information (focus adjustment information) is calculated, and the focus adjustment system 5 performs lens drive control based on the lens drive information.

The calculation performed by the calculation circuit 4 and its configuration will be described below.

In this case, in calculating one distance measurement information to be used as the lens drive information from the distance measurement information from the three distance measurement points, the following logic is used.

-In general, the closest distance value is often the distance value of the target subject, so in principle, the closest distance value is output.

・ “But”, the distance measurement value of either left or right indicates a short distance,
In addition, when the distance measurement value at the center indicates a medium distance and the distance measurement value on the side opposite to the side indicating the short distance is a long distance, the distance measurement value at the short distance is such that the distance between the camera and the subject is low. Assuming that the obstacle is a short-distance obstacle, the center distance measurement value is output. This is because, for example, when shooting with the camera tilted, the lower distance measuring point is measuring the ground, and even in normal shooting, obstacles such as vegetation in front of the subject will enter. This is because there is a high probability that

An embodiment in which this is realized on a program when the arithmetic circuit 4 of FIG. 1 is configured by a microcomputer or the like is the flowchart of FIG. 2, and here, it is determined whether or not the condition of “But” mentioned above is satisfied. As a judgment reference value, the short distance is less than 1.5 m, the distance is more than 2 m, the distance is less than 4 m is the medium distance, and the distance is more than 5 m is the long distance.

In FIG. 2, in # 1, the distance measurement value from the distance measurement unit 1 (left distance measurement value) is less than 1.5 m, and the distance measurement value from the distance measurement unit 2 (central distance measurement value) is less than 2 m. It is large and less than 4 m, and it is judged whether or not the distance measurement value (right distance measurement value) from the distance measurement unit 3 is larger than 5 m, and if these conditions are satisfied, the central distance measurement value is measured at this time. It is selected as information, and if not satisfied, the process proceeds to # 2. In # 2, the distance measurement value from the distance measurement unit 3 is less than 1.5 m, and the distance measurement unit 2
It is determined whether the distance measurement value of is greater than 2 m and less than 4 m, and the distance measurement value of the distance measurement unit 1 is greater than 5 m. If these conditions are satisfied, the central distance measurement value is set to the same as the above. If the distance measurement information at this time is selected and not satisfied, # 3
Proceed to. In # 3, whether the distance measurement value from the distance measurement unit 3 is smaller than the distance measurement value from the distance measurement unit 2 and the distance measurement value from the distance measurement unit 3 is smaller than the distance measurement value from the distance measurement unit 1. If the condition is satisfied, the distance measurement value from the distance measurement unit 3 is selected as the distance measurement information at this time, and if it is not satisfied, the process proceeds to # 4. In # 4, ranging unit 1
Is smaller than the distance measurement value from the distance measurement unit 2, and the distance measurement value from the distance measurement unit 1 is smaller than the distance measurement unit 3
It is determined whether or not the distance measurement value from the distance measurement unit 1 is smaller than that, and if this condition is satisfied, the distance measurement value from the distance measurement unit 1 is selected as the distance measurement information at this time. Similar to 2, the distance measurement value from the distance measurement unit 2 is selected as the distance measurement information at this time.

By doing this, compared to conventional AF cameras with a narrow field of view, it is possible to achieve an appropriate AF camera that focuses on the subject even if the framing is freely selected. In addition, compared to conventional wide-field distance measurement, it is possible to eliminate accidental distance measurement of short-distance obstacles such as the ground, so such false distance measurement is prevented and the distance to the subject is accurately determined. Can measure distance.

Although FIG. 2 is a flow chart in the case of being implemented by a program, it is of course possible to implement it as an analog circuit, and FIG. 3 shows this example.

In FIG. 3, 31, 32 and 33 are distance measuring units 1 of FIG.
Output lines from ~ 3. Since the output from each of these outputs has a low analog voltage as it gets closer, the diodes 34, 35 and 36 and the pull-up resistor 37 generate the closest distance information of three on the signal line 38. A window comparator 39 sets the signal line 40 to a high level when the center distance measurement value via the output line 32 is within a predetermined width (that is, a medium distance). Reference numeral 41 is also a window comparator, which sets the signal line 43 to a high level when the left side distance measurement value via the output line 31 is equal to or more than a predetermined width (that is, a long distance), or when the left distance measurement value is less than the predetermined width (that is, a short distance) 42 is high level. 44 is also a window comparator, and the output line
The signal line 46 is set to a high level when the right side distance measurement value through 33 is equal to or larger than a predetermined width (that is, a long distance), and the signal line 45 is set to a high level when the value is a predetermined width or less (that is, a short distance).
47 is an AND gate, and the left side of the above-mentioned “But” part is a short distance, the center is a middle distance, and the right side is a long distance. To do. 49 is an AND gate, which is the same as the above
The signal line 50 is set to a high level because each input is at a high level when the right side of the portion is a short distance, the center is a medium distance, and the left side is a long distance. Reference numeral 51 is an OR gate, which sets the signal line 52 to a high level when the condition of "But" is satisfied. Reference numeral 53 is an analog switch, and when the signal line 52 is at a high level, that is, when the condition of “however” is satisfied, the central distance measurement value via the output line 32 is not the closest distance value via the signal line 38. Output as lens drive information.

 It can be realized with a simple configuration as described above.

Regarding the distance, each distance measurement value is a signal of higher analog level as it is farther away, but it is not a signal proportional to absolute distance (0 at 0 m, infinity at infinity), but the reciprocal form of the absolute distance suitable for AF, That is, it is a signal corresponding to the depth of field.

Next, a case in which the judgment logic of the above "however" part is changed will be described.

・ In principle, output the closest distance value.

"But", the distance measurement value of either left or right is considerably closer than the distance measurement value of the center, and the distance measurement value of the center indicates a medium distance,
In addition, when the distance measurement value on the side opposite to the side showing the fairly short distance is farther than the central distance measurement value, the central distance measurement value is output.

That is, in the above-mentioned logic, when the distance measurement values in the left and right directions are opposite to the distance measurement value in the center, the condition is rewritten as a relative value.

FIG. 4 shows an example in which this logic is digitally implemented in an analog circuit, and the portions having the tone generating function are designated by the same reference numerals as in FIG.

In FIG. 3, the window comparators 44 and 41 are used to determine whether the distance is a long distance or a short distance. 4th
In the figure, in accordance with the logic change, the differential amplifier 61 to the signal line 62
(Left side distance measurement value-center distance measurement value) is output to. as a result,
When the left side shows a distance measurement value which is considerably closer to that in the center, a signal of a considerably low level flows in the signal line 62, and a high level is generated in the signal line 64 via the window comparator 63. When the left side shows a distance measurement value that is far from the center, a signal of a considerably high level flows in the signal line 62, and a high level is generated in the signal line 65 via the window comparator 63. On the other hand, the differential amplifier 66 outputs (right-side distance measurement value-center distance measurement value) to the signal line 67. As a result, when the right side shows a distance measurement value which is considerably closer to the center, a signal of a considerably low level flows in the signal line 67, and a high level is generated in the signal line 69 via the window comparator 68. When the right side shows a distance measurement value that is far from the center, a signal of a considerably high level flows in the signal line 67, and a high level is generated in the signal line 70 via the window comparator 68.

After that, as in FIG. 3, the AND gate 47 is a high level of the signal line 40 when the central distance measurement value is the middle distance and a high level of the signal line 70 when the right distance measurement value is far from the central distance measurement value. , And the high level of the signal line 64 when the left distance measurement value is closer to the center distance measurement value, the signal line 48 is set to the high level. Further, the AND gate 49 sets the high level of the signal line 40 when the central distance measurement value is the middle distance, the high level of the signal line 65 when the left distance measurement value is far from the central distance measurement value, and the right distance measurement value. The signal line 50 is set to the high level when each of the high level of the signal line 69 when the distance is closer to the central distance measurement value is satisfied. As a result, in such a case, the central distance measurement value is selectively output as the lens drive information by the analog switch 53 as in the portion "But" in the logical statement.

By considering the distance difference in the logic as described above, the second
As shown in Fig. 3 and Fig. 3, the judgment can be made without being constrained by the fixed distance measuring zone. That is, the correct determination can be made even when the center distance measurement value is near the middle distance or far from the middle distance (details will be described later).

Also, considering the distance difference leads to consideration of the amount of blur when one side is focused. That is, a large distance difference in the logic means that one side is blurred, which means an important judgmental division in the selection act.

Next, FIG. 5 shows an example in which the condition of "however" of the above logic is judged in an analog manner.

As shown in the figure, 80 is an output circuit that produces an output that has a large output when medium-distance information is input and a small output when short-distance or long-distance information is input. For 40, a large output is generated as the distance is increased. Similarly, reference numeral 81 is an output circuit having a function for generating a larger output as the input is "positive", and a larger output is generated on the signal line 70 as the right side distance measurement value is farther than the central distance measurement value. On the contrary, reference numeral 82 denotes an output circuit having a function for generating a larger output as the input is "negative" and larger, so that a larger output is generated in the signal line 69 as the right distance measurement value is closer to the central distance measurement value. Further, 83 is the output circuit 82.
And 84 are functional circuits similar to the output circuit 81, respectively.

As described above, analog signals having a tendency similar to that of the digital configuration of FIG. 4 can be obtained on the signal lines 40, 70, 69, 64 and 65.

Here, an example for realizing the input / output characteristics of the output circuit 81 will be shown.

FIG. 6 shows an example of the characteristic, and shows the characteristic that the output value shown on the vertical axis 92 is output by the function shown by the straight line 93 for the input shown on the horizontal axis 91 (the right is positive and large). That is, if the input is less than the value shown at point 94, the output is zero, and if there is more input, the output increases accordingly. FIG. 7 shows an example of a circuit configuration for realizing this, in which an input is applied to a signal line 95 and a voltage having a value indicated by the point 94 is applied to the signal line 97. Reference numerals 98 and 99 are diodes, which have a function of transmitting the higher voltage applied to the signal lines 95 and 96 to the signal line 100. A differential amplifier 97 outputs the voltage difference between the signal lines 100 and 96 to the signal line 101. For this reason, the signal line 101 has (the higher level value of the signal lines 95 and 96 −
The difference information of the level value of the signal line 96, that is, the signal line 96
When the level of the signal line 95 is higher, the higher level value appears as an output. In this circuit, the signal line 95
Put an inverting circuit in front of the diode 99 of the signal line
If the voltage at the zero-cross point of the inverting circuit 82 is added to 99,
The circuit configuration of the output circuit 82 is obtained.

Referring back to FIG. 5 again, the operation will be described. 110,111
Is a multiplier, which performs the operation for each "and" part of the condition "however". And the signal lines 48,50
A signal having a level indicating the degree to which the condition "however" is satisfied is generated on the signal line 113 through the adder 112 for the output via the. The comparator 114 determines whether the central distance measurement value is selected or not by determining whether the value is true or false by the comparator 114 and causes the analog switch 53 to select.

This method is not binary for the condition of "however",
It will be calculated in an analog manner. As a result, each condition can be comprehensively determined. That is, even if the center distance value deviates slightly from the middle distance, if the left side distance measurement value is the closest distance and the right side distance measurement value is infinity such as infinity, the center distance measurement value is selected, and conversely If the distance measurement value of is not extreme, the central distance measurement value may not be selected. That is, as described above, it becomes possible to comprehensively determine each element of the condition "however". In other words, even if the determination condition "medium distance" is not strict, it is possible to prevent erroneous determination because the conditions are comprehensively evaluated. For example, even if the determination of "medium distance" is slightly deviated from the ideal, similar results can be obtained if the other conditions work a little strongly as described above, so it can be said that the degree of freedom of logic is high.

FIG. 8 shows an example of so-called probability selection, in which the calculation of FIG. 5 is carried out in the form of “selecting the distance measurement value that is most likely to be correct”.

In this embodiment, first, the logic of "selecting the closest one" is "increased to be closer from infinity to 1 m, and when it is closer than 1 m, the selected ratio is lower than that of 1 m. Change to ". Also, if the left and right distance measurement values are larger in the opposite direction than the center distance measurement value, select the center distance measurement value. If the value is larger in the opposite direction than the value, increase the rate of selecting the center distance measurement value. " Then, the act of “selecting those with a high selection rate” is performed.

In other words, the closer the left distance measurement value is to 1 m, the higher the correct rate of the left distance measurement value.

・ The closer the center distance measurement value is to 1 m, the higher the correct rate of the center distance measurement value.

・ The closer the right distance measurement value is to 1 m, the higher the right distance measurement value is.

・ The closer the left distance measurement value is to the center distance measurement value, and the closer the right distance measurement value is to the center distance measurement value, the higher the correct rate of the center distance measurement value. .

・ The farther the right distance measurement value is from the center distance measurement value, and the closer the left distance measurement value is from the center distance measurement value, the higher the correct rate of the center distance measurement value. .

Based on the above logic, the distance measurement value to be selected is determined based on the total rate of selection rates of the three parties.

As an embodiment, 120, 121 and 123 are output circuits (details of the circuit configuration will be described later) having a function of outputting a higher level signal to the signal lines 123, 124 and 125 as they are closer to 1 m described above. As a result, if the left distance measurement value is close to 1 m, the level of the signal line 123 is high, and as a result, a large amount of current representing "the rate at which the left distance measurement value is correct" is supplied to the signal line 129 through the resistor 126. Similarly, if the center distance measurement value is close to 1 m, the level of the signal line 124 is high, and as a result, a large amount of current representing "the rate at which the center distance measurement value is correct" flows through the resistor 127 through the signal line 130. If the right distance measurement value is close to 1 m, the level of the signal line 125 is high, and as a result, a large amount of current representing "the right distance measurement value is correct" flows through the signal line 131 through the resistor 128.

Next, paying attention to the difference in distance between the left and right sides, as in FIG. 5, the degree when the left distance measured value is close to the signal distance 133 and the right distance measured value is far from the output circuit 81, 83 through the multiplier 132 is obtained. , Through the resistor 134, a large amount of current representing "the rate at which the center distance measurement value is correct" flows through the signal line 130. Similarly, the right distance measurement value is close to the signal line 136 from the output circuits 82 and 84 through the multiplier 135,
In addition, the degree when the distance measurement value on the left side is far is obtained, and a large amount of current representing "the rate at which the distance measurement value at the center is correct" is supplied to the signal line 130 through the resistor 137.

Reference numerals 138, 139, and 140 denote resistors, and these resistors have a voltage value that is an added value of currents that represent “probability of correctness” by each of the above calculations. 141 is a signal line corresponding to the one with the highest voltage selected from the signal lines 129 to 131.
A peak detection circuit in which any one of 142, 144, and 146 is set to a high level, and the detailed circuit configuration will be described later.

As described above, when the distance measurement value on the left side is most correct, the signal line 142 becomes high level and the analog switch 143
Is turned on, and the distance measurement value via the output line 31 is selectively output as lens drive information. Similarly, when the center distance measurement value is most likely to be correct, the signal line 144 becomes high level,
The analog switch 145 is turned on and the distance measurement value via the output line 32 is selected as the lens drive information. When the distance measurement value on the right side is most likely to be correct, the signal line 146 becomes high level, the analog switch 147 is turned on, and the distance measurement value via the output line 33 is selectively output as lens drive information.

To give priority to logic like the condition of "But" mentioned above is to change the function in each output circuit 81-84, 120-122 or change the value of resistors 134, 137 and resistors 126-128. Can be achieved with.

In this way, for each logic, the "correct rate" is added to make a comprehensive determination, which means that the sum of a plurality of logics is performed in an analog manner, and independence is not required between the logics. The judgment does not have to be strict. That is, it means that a logic with some contradiction may be used, and there is an advantage that various logics can be added in the same manner.

In the above description, the term "rate" is qualitatively the same as the probability, but it is calculated independently without performing the above-mentioned independence test, and is normalized. It is different from “probability” in that it can be calculated including points that are not present and cases that are gray (ambiguous).

Here, an example for realizing an input / output characteristic like the output circuit 120 of FIG. 8 will be shown.

FIG. 9 shows an example of the characteristic, and shows the characteristic that the output value shown on the vertical axis 372 is output by the function shown by the straight lines 373a and 373b for the input shown on the horizontal axis 371. That is, until the input reaches the value indicated by the point 374 (1 m in the embodiment), the output gradually increases like a straight line 373a, and if there is more input, the point
The input of 374 may be used as a peak and gradually decreased as shown by a straight line 372b. FIG. 10 shows an example of a circuit configuration for realizing this, in which an input is added to the signal line 359,
If its value is lower than point 374, it is placed on signal line 360 at point 374.
Since the voltage corresponding to is applied to the diode 351,
352, differential amplifier 355, diode 356 and signal line 362
When a predetermined voltage is applied to the adder 358, an output as shown by a straight line 373a in FIG. 9 is generated on the signal line 364, and when the input to the signal line 359 is higher than the point 374,
Since the voltage corresponding to the point 374 is applied to the signal line 361, the diodes 353 and 354, the differential amplifier 363, the diode
The function of 357 and the adder 358 produces an output on the signal line 364 as shown by the straight line 373b in FIG.

FIG. 11 shows a configuration example of the peak detection circuit 141 of FIG. 8. When the input through the input line 151 is the maximum, the output line 154 is at the high level and the input through the input line 152 is the maximum. In this case, the output line 155 is set to high level, and when the input through the input line 153 is maximum, the output line 156 is set to high level. That is, input line 1
When the input through 51 is the maximum, the input through the input line 151 is larger than the input through the input line 152, so the output of the comparator 157 is high level, and the input through the input line 151 is the input line 151. Since it is larger than the input through 153, the output of the comparator 158 also becomes high level. Therefore, the output of the AND gate 159 becomes high level, that is, the output line 154 becomes high level. Also input line 153
When the input through the input line 153 is the maximum, the input through the input line 153 is larger than the input through the input line 151, and therefore the output of the comparator 160 becomes high level, and the input through the input line 153 is the input line 152. The output of the comparator 161 also becomes a high level because it is larger than the input through the. Therefore, the output of the AND gate 162 becomes high level, that is, the output line 156 becomes high level. When the input through the input line 152 is maximum, the outputs of the comparators 157 and 161 and the AND gates 159 and 162 are low level, so that the output of the NOR gate 163 is high level, that is, the output line 155 is high level.

FIG. 12 is a flow chart when the embodiment of the analog circuit of FIG. 8 is constructed by a microcomputer or the like.
At, for example, initializing a register that holds each distance measurement value, proceed to # 22, and determine how much the distance measurement values on the left, center, and right sides differ from "1 m" that is considered as a short distance in this embodiment. The weighting is performed (weighting increases as it approaches 1 m) in accordance with (the absolute value of) the difference (corresponding to the output of the output circuits 121-123 in FIG. 8). Next, in # 23, paying attention to the difference between the center distance measurement value and the left and right distance measurement values, the degree to which the left distance measurement value is close and the right distance measurement value is far from the center distance measurement value, or the right distance measurement value The central distance measurement value is weighted according to the degree to which the distance value is close and the left distance value is far (see FIG.
(Corresponding to the output of 2,135) (the weight is increased as the left and right are separated in different directions). Then proceed to # 24,
Of the weighted left, center, and right distance measurement values, the one with the highest weight is output as the lens drive information at this time (corresponding to the output of the peak detection circuit 141 in FIG. 8).

FIG. 13 shows an embodiment in which the flow shown in FIG. 12 is programmed, in which the left, center and right distance measurement values are input as variables to L, C and R, and the , And output the selected distance to a variable called OUT. It is an example of a FORTRAN-like program that is a type of machine language used when coding a computer.

In Fig. 13, "L", "C", "R" of LR, CR, RR
Is a variable "left side distance measurement value", "center distance measurement value", "right side distance measurement value", and "R" in the latter stage of each is "correct" for left, center and right distance measurement values. "Rate". α, β
Represents the weighting function, and α is the left, center, and right distance measurement values of 1 m.
Is a function that becomes the maximum when, and β is a function that gives a greater weighting as the difference between the center distance measurement value and the left and right distance measurement values becomes larger than 2 m. Also, ABS means that the absolute value is selected, max1 means that the maximum value is selected, and therefore ABS (L-1m) means that the absolute value of the difference obtained by subtracting 1 m from the distance measurement value on the left is obtained,
max1 ((L−C) −2m, 0) means that the difference between the calculated distance difference between the left distance measurement value and the center distance measurement value and the difference between 2m and “0” is selected to be large. Also, EQ means equal.

 The above FIG. 13 will be described in more detail.

Needless to say, in the part corresponding to # 21 in Fig. 12, each variable is initialized, and in the part corresponding to the next # 22, the left, center, and right distance measurement values are measured according to the difference from 1 m. Are weighted (using the function α), and variables LR, CR, and RR each containing a ratio that is likely to be correct are set. In the part corresponding to the next # 23, with respect to the central distance measurement value weighted by the function α, depending on the degree to which the left and right distance measurement values deviate in different directions (2 m is used as a reference in the embodiment). Further weighting (using the function β) is performed, and the variable CR is set. Then, in the part corresponding to # 24, each of the variables LR and C
The most weighted R or RR is selected. Specifically, for example, if LR and max1 (LR, CR, RR) are equal, the variable L, that is, the left distance measurement value is output as lens drive information.

On the program, the act of selecting the peak is changed to the condition that has the same value as "max1" to facilitate the program.

Next, the embodiment of FIG. 8 is applied to the subtraction method (that is, the deduction method).
Fig. 14 shows an example of the test carried out at.

The output of 120 increases the level of the signal line 123 as the left side is closer to a predetermined short distance, but the central and right sides of the inverting amplifier 170 absorb current from the signal lines 130 and 131 by resistors 171 and 172 in order to reduce the correct rate. Similarly, the closer the center is to a predetermined short distance, the higher the level of the signal line 124, but the left and right sides of the inverting amplifier 173 absorb the current from the signal lines 129 and 131 by the resistors 174 and 175 in order to reduce the correct rate. Similarly, the closer the right side is, the higher the level of the signal line 125 is, but the central and left sides of the inverting amplifier 176 absorb the current from the signal lines 129 and 130 by the resistors 177 and 178 in order to reduce the correct rate.

Also, the output from the signal line 136 indicating the ratio of the left side being far from the center and the right side being close to the center, the resistance is passed through the inverting amplifier 179.
180,181 absorbs the current from the signal lines 129,131, which means the left and right ratios, and conversely, the signal line 133 output, which shows the ratio where the right is far from the center and the left is close, the resistors 183 and 184 mean the left and right ratios through the inverting amplifier 182. Signal line 129,131
Drains the current from.

The subtraction method for absorbing the current as in the embodiment of FIG.
Even if it is used together with the addition method shown in the figure, it has almost the same function, and there is an advantage that the numerical value (voltage) does not become excessive when calculating the total sum of operations.

FIG. 15 shows another embodiment, in which the “rate” is added in the embodiment of FIG. 8 and the “rate” is subtracted in the embodiment of FIG. Although addition and subtraction are performed on the signal line having the "correct rate", the present embodiment is a method of performing addition only using a plurality of signal lines.

For example, use 3 signal lines of 0%, 50%, 100% ・ When the logic result is a strong negative, add a predetermined value to the 0% signal line ・ The logic result is a weak negative (somewhat different ) …… Add a predetermined value to the 0% and 50% signal lines ・ When the logical result is unknown (I am confused about the answer) …… Add a predetermined value to the 50% signal line ・ Affirmative result (which However, when the logical result is affirmative, the specified value is added to the 100% signal line. The percentage of the center of gravity is obtained from the value of the signal line, and the maximum position of the center of gravity (%) is selected.

This calculation has a drawback in that the number of calculations increases as compared with the above-mentioned method. However, since the above-mentioned method carries out the synthesis of the logic with each one signal line, it means that the degree of affirmation or negation is synthesized. That is, it is impossible to distinguish between the "probable rate" when there are multiple logical results whose logical results are unknown and the "probable rate" when there is no confusing logical result. For that reason, you may accidentally choose the gray result, which is confusing in other words. In the method explained this time, if there are many logical results that get lost, the rate of the signal line of "50%" increases,
Negative of "0%" and affirmative of "100%" become "25%" and "75%" in calculating the center of gravity. That is, there is an advantage that the calculation is performed in such a manner that the denial and the affirmation are diluted, and the calculation including the above-mentioned hesitation, weak affirmation, and weak denial is possible.

The developed version of this form is the so-called "FUZZY theory" that has been widely used in recent years.

Fig. 15 shows 1 in each of the "probable percentages" in Fig. 8
This is an example in which the number of signal lines that were lines is 5 lines of 0%, 25%, 50%, 75%, and 100%.

The closer the left side is, the higher the level of the signal line 123 and the higher the output of the amplifier 200 as described above. As shown by 201, the single resistance 126 in FIG. 8 is a resistance block in which five signal lines with "probable correctness" are coupled with different resistance values, and 100% of the signal lines are often small resistances. Is configured so that a small amount of current flows through the 0% signal line with a large resistance. The degree of closeness in the center is similar for amplifier 20
2. The resistor block 203 supplies the signal to five signal lines that determine the central ratio. Similarly, the right side rate is determined by the amplifier 204 and the resistance block 205 to determine whether or not the right side is a short distance.
It is supplied to two signal lines.

As described above, the resistance difference of the resistance block is logically correct with the idea that “the left distance measurement value should be absolutely correct” when the difference distance measurement value is logically determined to be a short distance. It is like a "ratio" of the idea that "other than the difference distance measurement value is correct, that is, it may not be selected." Therefore, the resistance ratios in the resistance blocks 206, 401 that supply current to the five signal lines that determine the central rate from the signal lines 133 and 136 may be different from the resistance ratios of the resistance blocks 201, 203, 205. In particular, as in the case of the logic described above, the resistance blocks 206, 401 are used because of the priority logic configuration such as "when ~"
Has a lower resistance on the 100% side and a higher resistance on the 0% side.

It should be noted that the average resistance value in the resistance block is low resistance in the case of strong logic due to the strength of the entire logic, that is, the weight of the logic result.

As described above, the signal line groups of three selected rates output the rates to be selected to the respective signal lines of the signal lines 208, 209 and 210 through the center of gravity calculating unit 207 which calculates the respective centers of gravity. Then, the maximum of the respective rates to be selected on the left side, the center, and the right side is obtained by the peak detection circuit 141, and by the high level of any of the signal lines 142, 144, 146, the distance measurement values of the left side, the center, and the right side are detected by the analog switches 143, 145, 147. Either of them is selectively output as the lens drive information.

FIG. 16 shows a specific configuration example of the center of gravity calculation unit 207.

As shown by 220, a signal line having a rate to be selected from 0 to 100% (in the above description, 5% of 0%, 25%, 50%, 75% and 100% is used).
The signal line of the book) is connected from the left side of the resistor 221 in the figure to a position corresponding to the ratio. This allows the signal lines 222, 22
The current is shunted to 3 at a ratio according to the connection position. The shunted current is operated by the divider 224 to calculate the output according to the shunt ratio, and the voltage is output to the signal line 225. Therefore, 1
If the current flows only at the position of 00%, the voltage of 100 is output, and if the current flows only at the position of 50%, the voltage of 50 is output. This gives a "selected rate" signal based on integrated logic.

As described above, by providing a plurality of signal lines that mean 0 to 100%, it is possible to operate even the gray result of the logic result. This is especially effective when the number of things affected by the logic is different, that is, when the left and right rates are 1 logic each and the central rate is 3 logic, as in the previous example. Become. That is, in the calculation of an accurate probability, each probability result must be normalized in each logic. In the above example, if the central rate is positive, unknown, or negative, the central probability will increase, decrease, or decrease, and the probability of other cases must decrease, increase, or increase (depending on the logical inclusive relation). I won't. Moreover, in the unknown logic, the decrease and rise must be different from the rise and decrease at the time of affirmation and denial. Even more simply, unless the probability of each case is manipulated accurately and intricately according to the logic result (especially when there are many cases and gray logic), the calculation accuracy cannot be maintained.

However, in the above-described example, by providing a signal line of 50%, for example, the positive or negative rate can be thinned for each rate. Also, since each positive and negative has a distribution constant having a value in 50% of the signal lines, the conclusion is to compare the centroids of these distributed ones even if the number of logics that affect them is different. , This number will be in a normalized form.

FIG. 17 shows a flowchart in the case where the above-described FIG. 15 embodiment is configured by a microcomputer, etc., in which the initial setting is made in # 31, and then the left, center, and right distance measuring values are set in # 32. Check how much difference there is from "1 m", which is regarded as a short distance, and weight each in the array according to this difference (15th
(Resistance blocks 210, 203, 205) and pay attention to the difference between the center distance measurement value and the left and right distance measurement values in # 33, the left distance measurement value is close to the center distance measurement value, and the right distance measurement value is Weighting (in resistance blocks 206 and 401 in FIG. 15) in the array of the central distance measurement values is performed according to the distance degree or the right distance measurement value is near and the left distance measurement value is far. Next, the procedure proceeds to step # 34, and the center of gravity in each of the weighted left, center, and right distance measurement values is calculated. Then, in # 35, the one with the largest center of gravity is selected and output (FIG. 15, peak detection circuit 141).
Corresponding to the output of).

FIG. 18 is an example of programming the flow of FIG. Note that here, 0 to 100% of the signal lines are 0 to 1 of the array.
It is calculated as an index ($) of 00. In addition, the resistance blocks 210, 203, 205 and 206, 401 are arranged in the arrangement of α $ and β $.
The function in is realized.

 FIG. 18 will be described in more detail.

Needless to say, in the part corresponding to # 31 in Fig. 17, the initial setting of each variable is performed.In the part corresponding to the next # 32, the left, center, and right distance measurement values are determined according to the difference from 1 m. Are weighted (by multiplying the function α over 0 to 100) into variables LR, CR, and RR each containing a ratio that is likely to be correct. In the next portion corresponding to # 33, with respect to the central distance measurement value weighted by the function α, the weighting (function β is set to 0 in the array) is further performed according to the degree to which the left and right distance measurement values deviate in different directions. Multiply over ~ 100) to make the variable CR. In the portion corresponding to # 34, the center of gravity in each array is obtained based on the added value of the left, center, and right distance measurement values weighted from 0 to 100. Then, in the portion corresponding to # 35, the one with the largest center of gravity is selected and output.

FIG. 19 is an embodiment which further conceptually expresses the embodiment of FIG. 15 described above.

In FIG. 19, 230 is an output circuit having a function similar to the output circuits 120 to 122 of FIG. 15 that “outputs a larger value as it approaches 1 m”, and from here on the left side via signal lines 231 to 233. The signal should be close to 1m on the center and right side. 234 is an output circuit having a function that should change the rate according to each distance measurement degree, and the resistance blocks 201, 203, 203 of FIG.
Corresponding to 205, 235 to 237 are function circuits (corresponding to the amplifiers 200, 202 and 204 in FIG. 15) for multiplying the function output and the degree output. Reference numeral 238 is an arithmetic unit for adding the change of the rate due to the logical result to the rate so far, and corresponds to the current addition by the resistance block in FIG. 239 is an output circuit having a function of “outputting a larger value farther from the center” similar to the output circuits 81 and 84 of FIG. 12, and 240 is similar to the output circuits 82 and 83 of FIG. It is a functional circuit that outputs a relatively large value. Reference numeral 241 is a multiplier corresponding to "and" in the same logical sentence as the multipliers 132 and 135 in FIG. Reference numeral 242 is an output circuit having a function for changing the rate based on the same logic as the resistance blocks 206 and 401 in FIG. 15, and 243 is an arithmetic unit for multiplying the function output and the degree output. 245 is an arithmetic unit that changes the value of the rate according to the function result.

After that, one of the distance measurement values is selectively output as the lens drive information by the same operation as in FIG.

Here, the fuzzy theory will be described with reference to FIG. The output circuits 230, 239, 240 mean the conditional membership function in the fuzzy condition, and the output circuits 234, 242 are the fuzzy consequent membership functions. In the above description, the Larsen method is used for the operation of outputting the consequent membership function from the conditional membership function performed by the operators 235, 236, 237, 243, 244. Similarly calculator 238, 24
The rule composition performed in 5 is equivalent to the algebraic sum operation (the above was simply added, but theoretically it is limited to the maximum "1" operation), and the condition performed by the multiplier 241 The algebraic product was used for the combination of.

As another example, there are various methods (extracted from the paper by Dr. Mizumoto of Osaka Electro-Communication University at the 5th Knowledge Engineering Symposium) as shown in the following table, and it is clear that these methods are effective.

Among the operations that output the consequent part membership function from the conditional part membership function, the one that uses the sum operation for the composition of rules, that is, the idea of adding a ratio that seems to be correct (add "0" if it is completely unknown) There are the following things.

On the contrary, the one that uses the product operation for the composition of the rule, that is, "1" when it is completely unknown, and the idea that the membership function of the consequent part emerges as the condition is satisfied, becomes next. There is one.

FIG. 20 shows a flow chart in the case where the above-mentioned FIG. 19 embodiment is configured by a microcomputer, etc., and in # 41, the left, center, and right distance measurement values are determined depending on how far from 1 m. Each of them is weighted, and in the next # 42, the central distance measurement value is weighted according to how far the left and right are apart, and in # 43, the weighting of each distance measurement value is performed. Select and output the largest one.

FIG. 21 shows the flow of FIG. 20 already described by fuzzy logic. Here, “short distance” is a function that takes the maximum “1” in “1 m” described above, and α is “0 to A function that gently takes "0 to 1" for "100%", β is, for example, "50 to 100%"
On the other hand, it is a function that suddenly takes "0 to 1".

The difference between the functions α and β makes a difference in the strength of the two logics.

 FIG. 21 will be described in more detail.

In the part corresponding to # 41 in Fig. 20, the left, center, and
Weighting is performed on each of the right distance measurement values, and the rate at which each distance measurement value is likely to be correct for this weighting is calculated by the function α, and these are set as variables LR, CR, RR. In the part corresponding to the next # 42, the difference between the left and right distance measurement values and the center distance measurement value is positive and large to some extent, and the difference between the center distance measurement value and negative is large to some extent , The center and the right distance measurement values are weighted according to the relative relationship of far (near), medium, and near (far), and the ratio at which the central distance measurement value is likely to be correct is given to this weighting. It is calculated by the function β, and this is set as the variable CR. In the part corresponding to # 43, the weighted variable
Select and output the larger of LR, CR, and RR.

In all of the above examples, the distance measurement results are left, center,
A distance measurement value in any one of the right distance measurement areas is selected. Next, a method of taking a medium value of these distance measurement values, that is, an intermediate value as a distance measurement result will be described. FIG. 22 shows an embodiment for realizing this, which will be explained as a modified form of the analog operation of FIG.

Since a signal indicating that it is better to use the central distance measurement value for the signal line 113 is output, the result is output between the central distance measurement value and the closest distance measurement value in an analog manner. 26
0 is a variable resistor for this purpose. When the level of the signal line 113 is zero, the closest distance measurement value via the signal line 38 is selected, and when the level of the signal line 113 is maximum, the central distance measurement via the output line 32 is performed. The value is servo-variable according to the level of the signal line 113 so that the value is selected. As a result, when the level of the signal line 113 is the intermediate level, the intermediate distance between the closest distance and the center distance measurement value is output.

By outputting an intermediate value in this way, when the depth of field is shallow, it may not be possible to focus on either, but if there is a certain depth of field as in general photography, both You can get a good photo that is in focus.

FIG. 23 shows an embodiment in which an intermediate value is output from the system shown in FIGS. 8 and 14.

The operation up to the calculation of each rate in the signal lines 129, 130, 131 is almost the same as described above. Reference numeral 261 is a normalization circuit for normalizing and outputting the signals of these three rates to three signal lines of 262, 263, 264 so that the sum of the three rates becomes "1". Multipliers 265,266,26 for these three signals
The weighted average of each distance measurement value is obtained by multiplying each distance measurement value at 7 and added at the adder 268, and output as lens drive information. As a result, a photograph in which everything is in focus can be obtained by the comprehensive judgment of the three parties.

FIG. 24 shows a specific circuit configuration example of the normalization circuit 261 shown in FIG.

271,272,273 are positive rate signals, and multipliers 274,275,276
The signal on line 280 is multiplied by
Output to 77,278,299. Reference numerals 281, 282 and 283 are resistors, which add three signals via these resistors, and the inverting amplifier 284 controls the signal line 280 so that the sum becomes "1".

FIG. 25 shows a form in which an intermediate value is output in the calculation of the distribution probability shown in FIG. The part up to 210 is the 15th
Similar to FIG. 23, the parts from 261 to 268 are for calculating the weighted average as in FIG.

By calculating the weighted average after calculating the rates as shown in Figures 23 and 25, the advantage of being able to correctly output the selection of the three parties by the calculation of the complex rate no matter how the proportion of each of the three parties changes There is.

Up to this point, the exclusion of the adjacent obstacle such as the ground has been determined by the value of the measured distance and the difference information thereof, but other information can also be an effective determination means.

Therefore, first, an example of determination based on the photometric value (brightness) will be described. FIG. 26 shows an arrangement example of photometric sensors in a camera.
Reference numeral 300 is a camera body, and reference numerals 1, 2 and 3 are distance measuring units shown in FIG. 1, which measure distances in the directions of 301, 302 and 303, respectively. 304,305,306
Is a known photometry unit, which performs photometry in the directions of 301, 302 and 303, respectively.

FIG. 27 is a diagram showing the photographing frame of the camera and the distance measurement and photometry directions. With respect to the photographing view angle 307, each measurement direction is not only widened to the left and right as indicated by 301, 302, and 303, but also widened up and down. This is because in shooting, the screen may be shot vertically as shown in FIG. 28 due to the framing request, and at that time as well, the horizontal distance (in the vertical position) is measured.

Of course, the accuracy will be better if the score is increased to 3 points or more, that is, 5 points as shown in FIG. 29, for example. However, increasing the number of focus detection points is technically (costly) difficult and should be selected depending on the application of the camera. However, the determination of an obstacle such as the ground is necessary in any case, so in this case, the following description will be made with three points.

FIG. 30 shows this embodiment and the like, in which the camera is controlled by measuring distance measurement values and brightness in three directions.

An average photometric unit 308 calculates the average of three photometric values from the photometric units 304 and 306, and outputs the average to the signal line 309. Reference numeral 310 denotes an arithmetic circuit for selectively outputting the distance measurement value as lens drive information, and the number of the above three photometric values and the average distance measurement value of the signal line 309 are increased as inputs, as compared with those described above.

FIG. 31 shows a flow chart in the case where the above embodiment is configured by a microcomputer, etc., and in # 51, the left, center, and right distance measurement values are weighted according to how far they are from 1 m. Then, at # 52, the central distance measurement value is weighted according to how far the left and right sides are apart. In the next step # 53, the one-side distance measurement value is weighted according to the one-side short distance on the left and right and the opposite side is bright. This is a short distance on one side,
If the brightness in the distance measuring direction on the opposite side is brighter than the average brightness, it is highly likely that the direction is pointing to the sky (sky), so there is a high possibility that the metering direction on the one side is measuring the ground. Therefore, the weighting is attempted based on this. Then, in # 54, the one with the greatest weighting is selected and output from the respective distance measurement values.

FIG. 32 describes the flow of FIG. 31 in fuzzy logic. The parts corresponding to # 51, # 52 and # 54 in FIG. 31 are # 41, # 42 and # 43 in FIG. The details are omitted because they are the same. In the part corresponding to # 53, for example, even if the weighting is large due to the distance measurement value on the right side being close to the short distance, if the light measurement value on the left side is brighter than the average brightness, the variable RR The weighting is reduced by the function γ.

Here, FIG. 33 shows an example of the consequent membership function represented by α, β, and γ. The function indicated by β in Fig. 33 shows "pretty" to remove the ground from the distance difference 10
This is a 0% function. In addition, the function indicated by γ represents a weak denial that is “not so good” by a function that is almost flat in order to exclude something that looks like the ground from the brightness.
Based on the above logic, the arithmetic circuit 310 in FIG. 30 selectively outputs the distance measurement information that seems to be the main subject to the signal line 311, and causes the AF drive system 5 to adjust the focus position of the lens. In addition, this action selects which of the left, center, and right seems to be the main subject, and the same selection mechanism as that for selecting the distance is used to determine the main subject from the left, center, and right photometric values. The photometric value in a specific direction is output to the signal line 312. A comparator 313 compares the average photometric value with the photometric value in a direction that seems to be the main subject, and outputs a high-level signal to the signal line 314 when the main subject is considerably darker than the average photometric value, assuming that the subject is backlit. Reference numeral 315 is a strobe discrimination unit, and as is well known, the signal line 316 is set to a high level to illuminate the strobe 317 when the subject is backlit or dark, and the signal line 3 is otherwise provided.
Through 18 the proper aperture value is indicated to the aperture drive system 319 by automatic photometry. In addition, 320 is an analog switch, strobe 3
When 17 is made to emit light, the diaphragm drive system 319 is controlled to match the strobe 317 based on the distance measurement value of the signal line 311. As a result, correct distance measurement and correct metering can be performed even if there is a subject other than the center and the subject is backlit.

The above is an example using the brightness of the subject, but an example using another input is shown.

34 and 35 are diagrams for explaining another embodiment,
Figure 34 shows the camera in a horizontal position.
Hold with your hand 331 and use the release button 332 when taking a picture. FIG. 35 is also a view when shooting in a vertical position, and in this case, an example in which a release button 333 is used so that the user can similarly grip and release easily.

In this way, when the release button 333 is used, since the vertical position shooting is performed, it is highly likely that the distance measurement on the left side is measuring the ground as indicated by 301 in FIG. 28, and other than that is indicated by 303 in FIG. 27. It is highly possible that the right distance measurement is measuring the ground. A signal SW1 'indicating that the release button 333 is used is sent to the arithmetic circuit 310 via the signal line 334 of FIG. 30 as an input for selecting distance measurement.

FIG. 36 is a flowchart for realizing whether or not shooting is performed by using the release button 333 when the arithmetic circuit 310 is configured by a microcomputer or the like. In # 61, the left, center, and right distance measurement values are from 1 m. Weighting is performed according to how far the distance is, and in the next step # 62, the central distance measurement value is weighted according to how far the right and left are apart. In the next step # 63, the distance measurement value on one side is weighted according to whether or not one side is short-distance and vertical position shooting is performed. This is because if one side is close-range and the shooting at this time is vertical position shooting, there is a high possibility that the distance measuring direction on the one side is measuring the ground, so weighting will be performed based on this. To do. Then, in # 64, the one with the largest weighting of the respective distance measurement values is selected and output.

FIG. 37 describes the flow of FIG. 36 by fuzzy logic. The parts corresponding to # 61, # 62 and # 64 of FIG. 36 are # 41, # 42 and # 43 of FIG. Since it is the same, the details are omitted in this embodiment. In the part corresponding to # 63, for example, even if the weighting is set large because the distance measurement value on the right side is close to the short distance, if the shooting at this time is vertical shooting, the variable PR The weighting is reduced by the function γ.

Fig. 38 shows an example of using another function for upside down judgment.

Reference numeral 335 is a disc rotatable about the shaft 336, and a weight 337 is formed in a part of the disc 335, and the weight 337 is arranged on the lower side. Reference numeral 338 denotes a non-contact encoder fixed to the camera, which detects the attitude of the camera by reading the code of the code plate 339 coaxial with the disk 335. By inputting the posture information ANG from here through the signal line 340 to the arithmetic circuit 310 as shown in FIG. 30, even when shooting upside down without limiting the operation members as in the previous example. It can correctly detect the “direction like the ground”.

FIG. 39 shows a flow chart when the arithmetic circuit 310 is configured by a microcomputer, etc., and there is this angle signal.
In # 71, the left, center, and right distance measurement values are weighted according to how far they are from 1 m, and in the next # 72, how much left and right are separated from the center distance measurement value. Weighting is performed according to the degree of. In the next step # 73, one side is a short distance, and the distance measurement value on one side is weighted according to the posture information of the camera. This also intends to perform weighting based on the same inference as in the embodiment regarding whether or not the vertical position photographing is performed. Then, in # 74, the one with the largest weighting of the respective distance measurement values is selected and output.

FIG. 40 describes the flow of FIG. 39 by fuzzy logic. The parts corresponding to # 71, # 72 and # 74 of FIG. 39 are # 41, # 42 and # 43 of FIG. 21. Since it is the same, the details thereof will be omitted even in this embodiment. In the part corresponding to # 63, for example, even if the weighting is large due to the distance measurement value on the right side being close to the short distance, if the right side distance measurement direction is lower than the left side distance measurement direction, By such inference, the weighting of the variable RR is reduced by the function γ.
That is, the possibility of the ground is determined by two rules based on the binary information indicating which of the left and right distance measuring directions faces downward.

Even in the case of performing distance measurement in five directions as shown in FIG. 29, it is possible to eliminate the most ground-like object.

Next, an example of inputting slightly different information (for selecting a distance measurement value) based on the fuzzy theory is shown below.

In FIG. 30, the input of the strobe use / non-use information ST of the signal line 316 may be added to the distance measurement selection of the arithmetic circuit 310, and the distance selection logic may be changed based on this. Due to the limited output of the strobe 317, even if the subject is far away, it can only be taken as a photograph up to a medium distance (because the amount of light is insufficient). For this reason, it is better for the photograph to be focused on the distance (arrival distance) that the light can reach.

FIG. 41 shows a flowchart for realizing this when the arithmetic circuit 310 is configured by a microcomputer or the like.
At 81, weighting is performed according to how far the left, center, and right distance measurement values are from 1 m, and in the next # 82,
The central distance measurement value is weighted according to how far the left and right sides are apart. In the next step # 83, weighting is performed according to each distance measurement value depending on the distance at which the strobe light reaches. This is because it is better to focus on the distance at which the light reaches, as described above, so that weighting is performed according to the distance at which the light reaches. Then, at # 84, the one with the greatest weighting is selected and output from the respective distance measurement values.

FIG. 42 describes the flow of FIG. 41 in fuzzy logic. The parts corresponding to # 81, # 82 and # 84 in FIG. 41 are # 41, # 42 and # 43 in FIG. The details are omitted because they are the same. In the part corresponding to # 83, for example, with respect to the weighting calculated depending on whether or not the distance measurement value on the right side is close to the short distance, the weighting is performed by the function δ as the distance measurement value on the right side approaches the reach distance of the strobe light. I try to make it bigger.

Here, the “arrival distance” is a membership function that becomes zero at a distance of 5 m or more as shown in FIG. 43, and the consequent part δ is a membership function meaning weak affirmation as shown in FIG. 44. good.

As a similar condition, there is a remote control (hereinafter referred to as remote control) reception signal Rem shown on the signal line 350 in FIG. Some cameras, like television sets, also allow release by a remote control device. However, this device is also limited in the reach distance like the strobe 317, and there is a high possibility that the subject is also within the reach distance when instructed by the remote controller. For this reason, similar to FIG. 42, it is possible to realize in almost the same manner by replacing "strobe lighting" with "remote control reception".

By these, it is possible to prevent erroneous distance measurement of a distant point, and for example, even if the distance measurement point is deviated from the main subject, there is little extreme defocus, that is, an overall good picture can be obtained.

FIG. 45 is another example of the membership function of the reach when using the remote control. Since it is said that the remote control is often used from 1m to 5m, the rate of less than 1m and more than 5m is lowered.
This is because the remote control shooting is an unnatural action at a distance of less than 1 m, so it has the effect of preventing malfunctions at less than 1 m.

In this way, the parameter "distance frequently used" can be freely introduced.

Next, an example in which different information is considered using fuzzy theory is shown.

In the examples so far, the membership function of the distance difference from the center which is the output of the output circuit 239 in FIG. 19 is a function that increases as the distance from the center increases, as shown in FIG. 46. However, as a practical evaluation, the aperture is small,
If the focal length of the taking lens is short, the depth of field becomes deep, so either focus can be set at 2m, so long as it is 3m or more, the difference becomes noticeable.

Therefore, in the distance measurement selection in the arithmetic circuit 310 in FIG. 30, according to the focal length information f input through the signal line 360 and the aperture signal AV input through the signal line 309,
The membership function related to the above-mentioned distance difference is, for example, “F
It is recommended to switch to a shape that is insensitive to the distance difference, as shown in Fig. 47, such as when narrowing down from 8 "or when" f <50 mm ".

Conversely, if the aperture is shallow or the focal length is long, the
As shown in Fig. 48, it is better to be sensitive to the difference in distance.

In order to realize such a thing, the method of exchanging the membership functions according to the conditions as described above is a simple method.

As another method, the distance difference may be multiplied by “focal length × aperture value” and the degree of blur on the film may be changed to the same shape and introduced into the membership function. Easy to introduce).

FIG. 49 shows a flow chart in the case where this is configured by a microcomputer or the like. In # 91, the degree of distance between the left and right with respect to the central distance measurement value, the focal length at that time, and the aperture Weighting is performed according to the state.

To introduce this relationship in a fuzzy way:

As shown in FIG. 50, two series of inference formulas are prepared depending on whether the focal length is long or short. And "positive and large" prepares the membership function as shown in Fig. 46, and "positive and excessive" prepares the membership function as shown in Fig. 47. And when the focal length is "long",
As shown in 361, the function that has a maximum focal length of "1",
When it is "short", it is a function which becomes "1" at the shortest focal length as shown by 362 in FIG. Further, as shown by 363 in FIG. 52, the function with the aperture "open side" is a trapezoidal function in which the open side is "1" from a certain value, and the "small aperture side" is the function shown with 364. .

In this way, preparing two series of functions for the conditions of the focal length and the aperture means switching at a specific focal length in the binary logic, but in fuzzy logic like the weighted average of the conditions. It has an advantage that it works correctly even if the branch conditions are not exactly complementary as shown in FIGS. 51 and 52.

Further, in general, the frequency of the shooting distance depends on the focal length. Fig. 53 shows an example of frequency in the case of general focal length, Fig. 54
FIG. 55 shows an example of the frequency at a telephoto with a long focal length, and FIG. 55 shows a flowchart showing an example in which this is realized by a microcomputer or the like. That is, at # 101, weighting is performed according to the degree of matching the frequency of use of the shooting distance according to the focal length.

As described above, the fuzzy logic is characterized by the fact that when the focal length is long and the focal length is long, the short-distance shooting frequency is low. As shown in FIG. By adding a weak affirmation by the function δ, the probability of false ranging can be reduced. For example, if the left distance measurement area is frequently used at a short focal length and the focal distance at that time is short, the weighting of the left distance measurement value is increased.

According to the present embodiment, the left side, the center, and the right side of the photographing screen are distance-measured, and the possibility that the distance measurement value of an adjacent obstacle such as the ground is the distance measurement value is calculated from among the obtained distance measurement values. Since the determination is made based on the difference information, the camera attitude information, the respective photometric information obtained corresponding to the multi-point distance measuring positions, and the weighting amount of the lens driving information calculation of the distance measuring information is considered. It is possible to perform lens drive control suitable for only the subject, and it is possible to give a photograph in focus.

Further, since the above is automatically performed, it is possible to give a photographed image to the subject even with free framing, and it is possible to make the exposure control similarly suitable. In other words, it is possible to realize an AF camera that is in focus and exposed in most cases without being aware of the distance measuring field.

(Effects of the Invention) As described above, according to the present invention, it is possible to provide a camera photographing condition determination device or a camera that can easily determine more appropriate photographing conditions.

[Brief description of the drawings]

FIG. 1 is a schematic block diagram showing one embodiment of the present invention, and FIG.
FIG. 4 is a flow chart showing the operation when the embodiment of FIG. 1 is constructed by a microcomputer, etc., FIG. 3 is a circuit diagram by zone comparison for realizing the embodiment of FIG. 1, and FIG. Circuit diagram, FIG. 5 is a circuit diagram which also realizes this in an analog manner, FIGS. 6 and 7 are input / output characteristic diagrams of the output circuit 81 of FIG. 5 and circuit diagrams for realizing the same,
FIG. 8 is a circuit diagram by probability addition using a distance difference for realizing the embodiment of FIG. 1, and FIGS. 9 and 10 are input / output characteristic diagrams of the output circuit 120 of FIG. 8 and for realizing the same. FIG. 11, FIG. 11 is a circuit diagram showing a configuration example of the peak detection circuit in FIG. 8, FIG. 12 is a flowchart showing the operation when the configuration shown in FIG. 11 is configured by a microcomputer, and FIG. FIG. 14 is a circuit diagram by probability subtraction using a distance difference for realizing the embodiment of FIG. 1, FIG.
FIG. 16 is a circuit diagram based on the same probability array, FIG. 16 is a circuit diagram showing a configuration example of the center of gravity detection unit in FIG. 15, and FIG. 17 is a flowchart showing the operation when the embodiment of FIG. 15 is configured by a microcomputer, FIG. 18 is a diagram showing an example of programming it, FIG. 19 is a block diagram showing a fuzzy embodiment of FIG. 15, FIG. 20 is a flow chart showing the operation when it is configured by a microcomputer, etc. FIG. 20 is a diagram showing an example of programming the FIG. 20 embodiment, FIG. 22 is a circuit diagram showing an example of intermediate value output by analog calculation using distance difference to realize the FIG. 1 embodiment, FIG. Is also a circuit diagram showing an example of intermediate value output by probability calculation, FIG. 24 is a circuit diagram showing a configuration example of a normalization circuit in FIG. 23, and FIG. 25 is a circuit diagram showing the distance difference in order to realize the embodiment in FIG. FIG. 26 is a circuit diagram showing an example of intermediate value output by a random array, FIG. 27 is a diagram showing an example of arrangement of distance measuring and photometric units in another embodiment, FIGS. 27 to 29 are diagrams showing examples of distance measuring point positions in the photographing screen, and FIG. 30 is another embodiment of the present invention. Fig. 31 is a schematic block diagram showing
FIG. 30 is a flowchart showing the operation when the embodiment is configured by a microcomputer, etc., FIG. 32 is a diagram showing an example of programming it, FIG. 33 is a diagram showing consequent part membership functions shown in FIG. 32, 34 and 35 are views for explaining the vertical position and horizontal position release operations of the camera in another embodiment of the present invention, and FIG. 36 shows the operation when configured by a microcomputer or the like using camera attitude information. Flowchart 37, shown
FIG. 38 is a diagram showing an example of programming it, FIG. 38 is a diagram explaining another mechanism for generating posture information of the camera, and FIG.
Figure is a flow chart showing the operation when it is configured by a microcomputer etc., Figure 40 is a diagram showing an example of programming it, and Figure 41 is a case where it is configured by a microcomputer using the reach distance of strobe light. Fig. 42 is a flowchart showing the operation, Fig. 42 is a diagram showing an example of programming it,
43 and 44 are views showing membership functions used in the embodiment of FIG. 42, FIG. 45 is a view showing membership functions of remote control reception signals, and FIGS. 46 to 48 are memberships based on distance difference. FIG. 49 shows a function, FIG. 49 is a flow chart showing an operation in the case of being configured by a microcomputer etc. using aperture value information and focal length information, FIG. 50 is a diagram showing an example of programming it, FIG. 51 and FIG. 52 is number 50
FIG. 53 is a diagram showing a membership function of aperture value information or focal length information in the embodiment, FIGS. 53 and 54 are diagrams showing membership functions at different focal lengths, and FIG. 55 is a graph showing the use frequency of focal lengths. FIG. 56 is a flow chart showing the operation in the case of being configured by a microcomputer or the like, and FIG. 56 is a diagram showing an example of programming it. 1,2,3 ... Photometric unit, 4 ... Arithmetic circuit, 304,305,306
…… Metering unit, 308 …… Average metering unit, 310 ……
Arithmetic circuit, 315 ... Strobe discrimination unit, 319 ... Aperture drive circuit, 332, 333 ... Release button, 337 ... Weight, 338
...... Non-contact encoder, 339 …… Code plate.

 ─────────────────────────────────────────────────── ─── Continuation of front page (72) Eiji Nishimori Eiji Nishimori Kanagawa Prefecture Kawasaki City Takatsu-ku 770 Shimonoge Office Tamagawa Plant Canon Inc. In-house (72) Inventor Yasutoshi Ichida 770 Shimonoge, Takatsu-ku, Kawasaki-shi, Kanagawa Canon Inc. Tamagawa Plant (56) References JP-A-59-193307 (JP, A) JP-A-60-233610 (JP) , A) JP-A-1-179115 (JP, A)

Claims (2)

(57) [Claims] 1. An input means for inputting information about a subject, a first calculating means for performing a fuzzy calculation using the information about the subject input to the input means as an input value of a predetermined membership function, and the first calculating means. Second calculating means for obtaining the center of gravity value of the calculation result of the calculating means, and photographing condition determining means for determining the photographing condition based on the information about the subject obtained from the input means and the calculation result of the second calculating means. An imaging condition determination device for a camera, comprising: 2. An input means for inputting information about a subject, a first computing means for performing fuzzy computation using the information about the subject input to the input means as an input value of a predetermined membership function, and the first computing means. Second calculating means for obtaining the center of gravity value of the calculation result of the calculating means, and photographing condition determining means for determining the photographing condition based on the information about the subject obtained from the input means and the calculation result of the second calculating means. A camera having: