JPH0332174A - Automatic exposure adjustment device - Google Patents

Abstract

PURPOSE:To attain proper correction in response to the position and shape of a main object without incurring unstable pattern or unnaturality by using fuzzy inference for the state evaluation of the pattern. CONSTITUTION:A switching control circuit 25 and a switching circuit 26 applies time division to an output of an LPF 22 and gives the result to integration circuits 31-36 respectively, which outputs the brightness evaluation value to a simple average circuit 68, normalizing circuits 51-56 and weighting circuit 61-66. Then a priority decision circuit 57 sets a few rules to decide the priority of each area of the pattern in response to the brightness distribution of the pattern in advance and uses fuzzy inference to decide the priority depending on the condition. Thus, the pattern is evaluated with the inference based on a few rules decided experimentally in advance without detailed condition setting corresponding to various patterns and the optimum priority is decided and proper exposure adjustment is attained.

Description

DETAILED DESCRIPTION OF THE INVENTION (a) Field of Industrial Application The present invention relates to an imaging device such as a video camera that automatically adjusts exposure.

(B) Conventional Technology In video cameras, control of the brightness level of a captured video signal using aperture, gain, etc., or so-called exposure adjustment, is a very important issue along with focus control.

Conventionally, as this automatic exposure adjustment mechanism, a method of detecting the average brightness level, peak value, etc. of the photographing ff1L screen and controlling the aperture and the gain for the captured video signal based on these has been popular. In this method, if there is a high-brightness part such as a light source in the screen, or if the background is dark, the main subject may not be properly exposed due to the surroundings.

In order to solve this problem, for example, Japanese Patent Laid-Open No. 62-11036
9<HO4N5/243) has been proposed. This method takes advantage of the fact that the main subject is likely to be located in the center of the screen, and divides the image capture screen into the center and other peripheral areas, obtains the brightness level of each area, and then calculates the brightness level of each area. The exposure is adjusted based on the ratio to obtain the appropriate exposure for the main subject in the center of the screen.

Figure 1E15 is a block diagram of a system using such an approach.

The incident light passes through a lens (1), the amount of light is adjusted by an aperture mechanism (2), and then photoelectrically converted by an imaging circuit (3) and output as a captured video signal.

This captured video signal is amplified by a variable gain amplifier (4) and then sent to a video circuit. At this time, the imaging circuit (
The output of step 3) is compared with a target brightness level by a comparator (5), and the aperture mechanism (2) is controlled using this differential voltage.

On the other hand, the captured video signal is sent to the area selection circuit (19), and the switching signal for area separation obtained by the synchronization separation and switching control circuit (12818) allows the signal of the priority area to be used for level detection. The signal of the non-priority area is inputted to an integrating circuit (20) which is a digital integrator, and the signal of the non-priority area is inputted to an integrating circuit (21), and each of them is integrated for l fields.

The outputs of both integration circuits are supplied to a division circuit (15), and the ratio between them is sent to a gain control circuit (16) and an aperture target brightness level itIJ9g circuit (17). Both control circuits are
Based on the result obtained by the division circuit (15), correction is performed by varying the target brightness level of the aperture and the gain of the variable gain amplifier (4).

(c) Problems to be Solved by the Invention In such a center-priority method, the setting of the priority area changes greatly, so if correction is performed accordingly, the brightness level of the entire screen will fluctuate greatly.

On the other hand, if the priority area is made larger, stability will be achieved, but not only the main subject but also the background will be included in the priority area depending on the position and shape of the main subject, making it impossible to obtain sufficient correction.

In order to improve this, it may be possible to subdivide the priority area and make the priority of each area variable to accommodate various screens. However, actual screens are diverse and have various brightness distributions, and numerically dividing conditions and setting values for all cases is subject to severe restrictions in terms of system capacity and processing ability.

(d) Means for Solving the Problems The present invention sets a small number of rules in advance to determine the priority of each area of the screen according to the brightness distribution of the screen, and when determining the priority based on these conditions, It is characterized by using fuzzy inference.

(E) Effect The present invention has reached its conclusion as described above, so by using fuzzy inference, without setting detailed conditions corresponding to various screens, it is possible to solve the problem based on a small number of rules determined experimentally in advance. The screen is evaluated by inference and the optimal priority is determined.

(f) Example An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a circuit block diagram of the device of this embodiment.

The incident light passes through a lens (1), the amount of light is adjusted by an aperture mechanism (2), and then photoelectrically converted by an imaging circuit (3) and output as a captured video signal. This captured video signal is amplified by a variable gain amplifier (4) and sent to a video circuit, and is also amplified by a variable gain amplifier (4) and sent to a video circuit.
It is fed to an integrator (80).

L P F (22) extracts the low frequency component of the luminance signal in the captured video signal and outputs it to the subsequent switching circuit (26).

The synchronization separation circuit (23) extracts vertical and horizontal synchronization signals from the captured video signal, and the subsequent switching control circuit (25) extracts the vertical and horizontal synchronization signals from the CO of the imaging circuit (3).
Based on the fixed oscillator output used to drive D, the third
A switching signal is issued for screen division over six areas (AI) to (A6) in the figure.

The switching circuit (26) receives the switching signal and switches each region (
Switches sequentially according to A1) to (A6), L P F
(22) The output is time-divided for each region by this switching circuit (26>) and is supplied to integration circuits (31) to (36), respectively.

The integrating circuits (31) to (36) each include an A/D converter (27) that A/D converts the output of the switching circuit (26) as shown in FIG. circuit(
28) A digital integrator that is controlled by an adder (29) that adds the output and a latch circuit (28) that latches the added output, so that the low frequency component of the luminance signal within the corresponding area is A/D conversion is performed at the sampling period, and this A/D conversion data is integrated over one field period. Here, the integration circuit (31) is
The integrated value for one field of the low-frequency component of the luminance signal within the area (A1) is output to the memory (41), and the following is similarly applied to the areas (A2) (A3) (A4) (A5) (A6). The integral value of one field of the luminance signal within the integration circuit (32) (
33) (34) (35) (36) to memory (42) respectively.
)(43)(44)(45)(46). The latch circuit (28) is reset every 1 fields, and each memory retains the data immediately before the reset of each latch circuit, and the data is updated every field.

By the way, the areas (A1) to (A6) are respectively (Sl) to (S6), and the area (Al) is located at the center of the screen as shown in FIG. ).Furthermore, there is an area (A2) around this area (A2).
3) to (A6) are arranged.

When the integration for one field, which is one screen, is completed, the latest integrated value for one field in each area held in the memory (41) to (46) is calculated as the luminance evaluation value ( Y
l) to (Y6) are the subsequent simple average circuits (37),
It is output to each normalization circuit and each weighting circuit.

The normalization circuits (51) to (56) convert the brightness evaluation values (Yl) to (Y6) in each region into respective areas (Sl) to (S6
), and the brightness evaluation value per unit area of each region is divided by the normalized brightness evaluation value (V 1 ) to (V6) (however, V
1=Y1/St, V2=Y2/S2, . . . ).

The priority determination circuit (57) determines each normalized luminance evaluation value (vl
) to (V6), the priority (weight) of each area is determined. The priority determination process in this priority determination circuit (57) is executed according to a flowchart as shown in FIG. Specifically, the following six rules are used.

[Rule (1)] rifVl and v2 are close andV 1 and v3 are not close then area (AI) (A2) is given priority.'' [Rule (2)] rifVl and V2 are not close andV 1 and ■3 are close then area ( A1), (A3) Priority” [Rule (3)
] rif Vl and V2 are not close, and Vl and V3 are not close, then area (A1) is given priority.'' [Rule (4) rif Vl and v2 are close, and Vl and v3 are close.
hen area (AI), (A2), (A3) priority'' [Rule (5)] rif max (Vi) Same priority for all Lhen areas with small (i=1 to 6)'' [Rule (6)] rif max (V i ) is not small and the simple average value is small then give priority to the area below the simple average value.'' These rules are as shown in Figures ¥E6 to 11.
Conditions such as “close” and “small” are “V2/VIJ
rmax (Vi)'' and is defined by a membership function for each input variable, and has the priority (wik) of each area as a conclusion part. In addition, the inference is the usual m
This is carried out using the in-max method.

Next, each rule will be explained in detail.

[Rule (1)] is defined by membership functions such as those shown in FIGS. 66a and 66b. Figure 6(a) shows rV
This is a membership function for the input variable (V 2 /V 1 ) that indicates the degree to which condition (1) of rule (1) “l and V2 are close” is satisfied. That is, in order to determine the degree of closeness indicating how close the normalized brightness evaluation value (Vl) of area (A1) and the normalized brightness evaluation value (V2) of area (A2) are, the input variable is set to V2. /Vl, and the membership value (u,,) is obtained by substituting the input variable (V2/Vl) in the latest field into the chevron-shaped membership function that takes the maximum value when V2/V1=1. Seek. still,
When V2/V1=1, the membership value (u,,) is maximum.

Figure 6(b) shows the input variable (V3/
is a membership function for Vl). That is, in order to determine the degree to which the normalized brightness evaluation value (Vl) of area (A1) and the normalized brightness evaluation value (V3) of area (A3) are not close, the input variable is set to V3. /Vl
Then, the input variable in the latest field (
V3/Vl) by substituting the membership value (
U,) is found. Note that when V3/V1=1, the membership value (u+s) is minimum. In this way, Figure 6 (a
)(b), the membership value (u, , )(u.

, ) will be calculated. Note that this calculation is based on the second
This corresponds to S T E P (100) in the flowchart in the figure.

The membership value (u,,) (u+t) is ST
In E P (101), the minimum value of both, that is, the smaller membership value is selected as the degree of fulfillment (Ul) of rule (1). In the example of FIG. 6, since u, ,<u,, U1=u, is set.

The operations of S T E P (100) and (101) described above are also executed for the remaining five rules.

[Rule (2)] is defined by valley-shaped and mountain-shaped membership functions as shown in Fig. 77a and (b), and as in the case of Fig. 6, the rule (2) is that rVl and V2 are not close. )
For condition (1), the membership value (u, ,) is closer than (a), and rVl and V3 are closer.''
2) Membership value (U,) for condition (2)
is determined from (b), and in 5TEP (101), the smaller membership value (u, , ) (ust) is selected as the degree of fulfillment (U2) of rule (2). In the example of FIG. 7, since u++>ust, IJ2=u*s is set.

[Rule (3)] is defined by a valley-shaped membership function as shown in Figure 88a and (b), and as in the case of Figure 6,
The condition of rule (3) that "rVl and v2 are not close" (
Rule (3) that the membership value (us+) for 1) is less than (a), and rVl and V3 are not closer.
The membership value (u,,) for condition (2) is found from (b), and in S T E P (101), the smaller membership value (u,,) (u,,) is determined by the rule ( 3) is selected as the degree of establishment (U,). In the example of FIG. 8, since u3+<u, it is set to U,=u, .

[Rule (4)] is defined by a mountain-shaped membership function as shown in Figure 99a and (b), and rVl and V2 are close.
The membership value (u, ,) for condition (1) of rule (4) is closer than (a), and the membership value (U, ,) for condition (2) of rule (4) is that rVl and V3 are closer. ,,) can be found from (b), and S T E
Membership value (us+)(u
, , ) is selected as the degree of fulfillment (U, ) of rule (4). In the example of FIG. 9, U.

Since 1>u4m, it is set as 04=u.

[Rule (5)] As shown in Figure 10, the maximum value (max (Vi)
) (where i = 1 to 6) is the input variable, and this m
It is defined by a membership function shown by a simple decreasing straight line indicating the degree to which ax(Vi) is small, and once max(Vi) is determined, the membership value (U, , ) is uniquely determined. Note that this membership value (ul) is max (Vi)
becomes smaller as it becomes larger. S T E P (
101), there is only one membership value for rule (5), so after rule (5) is established (U5)
is set to U5=ug+.

[Rule (6)] is as shown in Figure 11 (a) and (b),
Similar to rule (5), the membership function has a simple increasing straight line with max (Vi) as an input variable, and the membership function with a simple decreasing straight line with all normalized luminance evaluation values (Vl) to (■6) as input variables. Defined by ship function. That is, in the membership function of FIG. 11(a), rmax
Condition (1) of rule (6) that “(Vi) is not small”
), the membership value (uo) can be determined if max(Vi) is determined as an input variable in order to determine the degree to which max(Vi) is not small. Note that this membership value (u,,) becomes smaller as max(Vi) becomes smaller. In addition, in the membership function in Figure 11(b), the rule that "the simple average value is small" (
If the simple average value is determined as an input variable to determine the degree to which the simple average value (2,) is small in condition (2) of 6), the membership value (U,) can be determined. Note that this membership value (uo) becomes smaller as the simple average value becomes larger. S T E P (101)
Then, select the smaller of the membership values (u,,) and (u,,), and the probability (U,) of rule (6) is U
,=u,.

is set.

As described above, it is determined in S T E P (102) that calculation of the degree of fulfillment (Uf) (i = 1 to 6) for all rules in S T E P (too) (tol) has been completed. Then, in STE P (103), the priority (Wk) (k=1 to 6) for each area is calculated. This priority (Wk) is calculated by weighted averaging of the conclusions based on the degree of establishment of each rule, as shown in the following equation.

In this formula (A), wik is the priority for each area regarding each rule, and is determined individually for each rule.

For example, regarding rule (1), “area (Al), (
In order to numerically indicate "prioritize A2)," the priorities (W, ) to (w
,,) is w, , = wth 83 W + s = W t 4 = W + s = W
+m=1 is set in advance. That is, the priority of area (AI) (A2) for rule (1) with respect to other areas is set to three times. Note that this priority setting is based on experiments conducted in advance.

Regarding rule (2), in order to indicate "prioritize areas (A1) and (A3)" as the conclusion part, the priority of each area (W,) to (W, ,) is w, , = w, ,=3 Wl g = W 1 += WI S ” W !
−=1 is set in advance.

Regarding rule (3), "give priority to area (A1)"
In order to show as a conclusion, the priority of each area (W,,)
〜(W3.) is w3. =3 W 22 = Wz 3 = W3 (°W!, =W,
, = 1 is set in advance.

Regarding rule (4), "area (A1), (A2),
In order to show "Prioritize (A3)" as the conclusion part, the priority of each area (W,,) to (W,,) is W 41 =
W 4! = W 4 m = 3w, 4=w
,,=w4. = 1 in advance.

Regarding rule (5), "All areas have the same priority"
In order to show this as a conclusion, the priority of each area (wo) to (W□) is V/5 l= W @ 1 = W @ s= Vl
It is set in advance that r @ 4 = Wth = W l @ = 1.

Regarding rule (6), ``In order to show as a conclusion that j gives priority to areas below the simple average value, the priority of each area (
W□) to (W,) are preset as 2=1 to 6, and if Vt<simple average value (2,) then wB=3Vs>simple average value (Zl) then w,=1. There is. For example, the normalized brightness evaluation values (
vl) to (v3) are larger than the simple average value (2,), VlF g 4 = Vlr 4 s = W s s
= 3W a r = W * * = W @ s
= 1. Note that the simple average value (Zl) is calculated by a simple average circuit (37) as described later. Considering the priority (Wk) considering all rules using the priority of each area in each rule set in this way using the examples in Figures 6 to 11, for area (A1), the formula (A) becomes.

In this formula (B), =u +13"u s*・3◆u, 13◆uo・3"u
i+'l+u 4t4 =u++"uTomoyuki"u3+"u4proverb◆Since u#l"u@1, the priority (Wl) of the area (A1) is W+=(3u+ ++3utt+31J)+"3114
*”Llst+umt)/(u++”ust”ust
“u<*”ust”ust).Similarly, the priority (W
, ) to (W,) are calculated as follows. The priority (Wk) of each area determined by fuzzy inference for all rules in this way is sent to the weighting circuits (61> to (66). The weighting circuits (61) to (66) determine the priority (Wk) for each area ( W,) to (Y,) are weighted and so-called priority processing is performed.In other words, each brightness evaluation value (Y,) to (Y6) is multiplied by the priority of the corresponding area (Wl) to (W,). Yi-Wi (i=1~
6> is calculated. All of the luminance evaluation values weighted in this way are supplied to a weighted average circuit (67). The weighted average circuit (67) includes weighting circuits (61) to (66).
) The added value of the output is divided by the sum of the products of each priority and area to output a weighted average value (Z, ). That is, calculate. Note that Si (i=1 to 6) indicates the area of each region.

The simple average circuit (68) adds up all the respective luminance evaluation values (Y i ) and calculates this added value as the area of the entire screen (S++S
Snow+...S,) to obtain a simple average of the entire screen. Note that this simple average value (Z,) is calculated based on each brightness evaluation value (Yi
), weighting circuits (61) to (66) assign priority (W
l) to (W,) are all weighted as “1”,
This value is equivalent to the value calculated by equation (C) using the weighted average circuit (67).

The simple average value (2,) calculated as above and the weighted average value (Z,) are divided! (69), m”
The division of Z, /ZH is done, and this division value (m)
are a gain control circuit (70) and a target level control circuit (71).
) is entered.

The gain control circuit (70) supplies a target level (P) to a comparator (5) that controls the gain of the variable gain amplifier (4). This target level (P) is the optimal target that can obtain the optimal exposure for the image capture screen when m=1, that is, when the simple average value (Zl) and the weighted average value Zl) are equal and the brightness distribution of the image capture screen is not considered. It is set at level (P6) and follows the division value (m) which is a correction value so as to always satisfy P=-mP.

Therefore, as a result, the weighted average value (2,
) becomes the optimal target level (Po).
) will change.

The comparator (5) integrates the captured video signal over a sufficiently long time constant (for example, one field period) and outputs the output of the integrator (90) indicating the brightness level of the corresponding field and the target level (P). This comparison output is supplied to the variable gain amplifier (4) so that the integrated output reaches the target level (P).
By controlling the gain so as to match , AGC is applied to the video signal in consideration of weighting processing.

The target level control circuit (71) supplies a target level (Q) to a comparator (72) that controls the aperture amount of the aperture mechanism (2), and this target level (Q) is equal to the target level (
Similarly to P), when the division value (m) satisfies the condition of m = 1, it is set to the optimal target level of Q - q o,
The weighted average value (m) changes to satisfy the formula Q = m q o, and as a result, the weighted average value (
The target level (Q) changes so that Z, ) always matches the optimal target level (qo).

The comparator (72) is connected to the target level (Q) and the integrator (80
) output, and this comparison output is compared with the aperture mechanism (
2), and based on this comparison output, the aperture mechanism (2)
the aperture mechanism (2) so that the integral output indicating the brightness level of the relevant field matches the target level (Q).
The aperture amount is controlled. Note that the time constant of the integrator (80) is equal to that of the integrator (90), and is set so that the aperture mechanism (2) does not follow instantaneous changes in the captured video signal.

As mentioned above, the variable gain amplifier (4) and the aperture mechanism (2)
The target level (
P )(Q ) changes according to the weighted average value (Z, ) that has been subjected to weighting processing, so the variable gain amplifier (
4) and optical exposure adjustment using the aperture mechanism (2), weighting processing is fully taken into consideration. For example, if the simple average value (Zl) of the entire screen is "120", the average value (Z If the brightness is "100", sufficient brightness is obtained over the entire screen, but rules (1) to (6)
If we focus only on the area that must be prioritized, we will not have enough brightness and the central area will be dark, so the division value (m) will be m-1, 2. Therefore, the target levels (P) (Q') are P=mP, , Q=
mq, and as a result, the gain of the variable gain amplifier (4) also increases, and the aperture amount of the aperture mechanism (2) also decreases.
Optimal exposure adjustment is made for the priority area.

Next, how the rules (1) to (6) affect exposure adjustment will be explained for each rule. Rules (1) to (4) are the basics of priority processing, and when the brightness evaluation values of areas (At), (A2, and A3) are close to each other, they work to increase the priority of that area. do.

For example, as in the prior art, the areas (AI) (A2) (A3) with the highest probability of containing the subject are simply given the same priority as the areas (A4), (A5), and (A6). When the subject (S) is photographed in a backlit situation as shown in Fig. 5, the bright background appears only in the area (A2), making it impossible to correct the subject (S) appropriately. Therefore, if rules (1) to (4) are applied, the area (Al
) (A3) are both dark and only the area (A2) is bright, so the normalized luminance evaluation value (Vl) (V2) (V3] 1m is V
1 #V 3 = #V 2 stands, rule (1)
+7) Conditions (1) (2), Condition (2) of Rule (3),
Since condition (1) of rule (4) is difficult to overcome, rule (
Only the probability of 2) is extremely high in the area (AI) (A3
) have a higher priority, and these areas (AI) (A3)
Focusing on the subject (S) that fits in this subject (S)
) is the optimum exposure condition. These series of rules are effective when the subject is backlit or in excessive direct sunlight.

Rule (5) corresponds to the case where the entire screen is dark, and when the maximum value of the normalized brightness evaluation value is not large, priority processing is not performed and the average value of the screen is used as the representative value. Furthermore, as a rule when the entire screen is dark, the following rule (5)'' can be substituted for rule (5).

[Rule (5)'] “The IF aperture is wide open.

Then all areas have the same priority.'' This rule (5) is based on the aperture mechanism (2) that reduces the darkness of the image capture screen.
If the aperture is quite open, that is, the aperture is quite large, the imaging screen is assumed to be dark, and the priority is set the same for all areas, making unnecessary corrections possible. It works to suppress The membership function of rule (5) with this degree of openness as an input variable is illustrated in Figure 12, and the priority of each area is w6.
, = w , , =w, =w, =w□=w, , =
It becomes l. At this time, to detect the opening degree of the diaphragm, the driving voltage value for operating the diaphragm mechanism (2) is set as shown in Fig. 14 (A/D
The converter (200) performs A/D conversion and the priority determination circuit (
57), or the diaphragm mechanism (2) is driven by a stepping motor that can count the rotor position, and the degree of opening corresponds to the number of steps in the opening direction of this motor, and further Various methods can be considered, such as separately providing a sensor to detect the degree of opening. Note that the amount of aperture of the aperture mechanism (2) changes in inverse proportion to the voltage value of the comparator (72). That is, the degree of opening changes in proportion to the voltage value. Note that it is also possible to use both rules (5) and (5)' as rules corresponding to the case where the entire screen is dark.

Rule (6) corresponds to the case where something extremely bright like a light source enters the screen, and the normalized brightness evaluation value (Y i )
When the simple average value (Zl) is small and the screen as a whole is dark even though the maximum value of Z1 is not small, priority is given to areas with low brightness. Here, if a light source, etc. is included in the screen, the simple average value (2,) will be small and the screen will be dark, even though the area containing this light source has a sufficient normalized brightness evaluation value. Put it away. In this case, by increasing the priority of areas with low normalized brightness evaluation values that do not include light sources, the influence of the light source is reduced and the exposure of the main subject is improved.

Note that the division of regions and the setting of each rule are not limited to this embodiment, and various forms can be considered. In addition, the switching circuit in Figure 1 (
26) Or division! It goes without saying that the operation (69) can be processed by software using a microcomputer.

Further, in the above embodiment, the optimum target level (Po) of the imaging screen without considering the luminance distribution is set in advance, and the division value (m) which is the ratio of the simple average VL (Zl) to the weighted average value (Z8) is set in advance. Integral that indicates the brightness level of the captured video signal by multiplying it by the optimal target level (P, ) as a correction value! (90
) The optimum exposure control is achieved by comparing the output with the output of
) to the weighted average value (Z, ) and the target level memory (91
) stored in the target level (P6) (however, P, T
is a value obtained by digitizing the target level (P, )) by a comparator (92), and based on the comparison result, the gain of the variable gain amplifier (4) is controlled, and the aperture mechanism (2 ) It is also possible to perform electrical and optical exposure adjustment by controlling the aperture amount. For example, the weighted average value (
zl) is smaller than the target level (Pl), it is assumed that the image capture screen is underexposed compared to the optimal exposure state after considering the luminance distribution, and the gain of the variable gain amplifier (4) is adjusted so that Z8=P. is raised and the aperture mechanism (2
) to increase the brightness, and conversely, when the weighted average value (Z, ) is larger than the target level (Pl), it is assumed that the exposure is overexposed compared to the optimal exposure state, and Z, = P
The brightness can be reduced by lowering the gain of the variable gain amplifier (4>) and increasing the aperture amount of the aperture mechanism (2) so that the brightness becomes 6'.

(G) Effects of the Invention As described above, according to the present invention, by using fuzzy inference to evaluate the state of the screen, a small It is possible to evaluate the screen using inferences based on the rules of numbers.

Furthermore, it is possible to perform appropriate correction according to the position and shape of the main subject without causing instability or unnaturalness on the screen.

[Brief explanation of drawings]

1 to 12 relate to an embodiment of the present invention, in which FIG. 1 is an overall circuit block diagram, FIG. 2 is a flowchart, FIG. 3 is an explanatory diagram of screen division, and FIG. 4 is a main circuit. The block diagram, Figure 5 is a diagram showing an example of the imaging screen, and Figure 6 is the rule (
Figure 7 is an explanatory diagram of rule (2), Figure 8 is an explanatory diagram of rule (3), Figure 9 is an explanatory diagram of rule (4), and Figure 10 is an explanatory diagram of rule (5). FIG. 11 is an explanatory diagram of rule (6), FIG. 12 is an explanatory diagram of rule (5), and FIGS. 13 and 14 are circuit block diagrams of other embodiments of the present invention. , FIG. 15 is a circuit block diagram of a conventional example. (31) (32) (33) (34) (35) (36)・
... Integration circuit, (61) (62) (63) (64) (6
5) (66)...Weighting circuit, (70>...Gain control circuit, (71)...Target level control circuit, (57)...
)...priority determination circuit

Claims (4)

[Claims] (1) Brightness level detection means that detects the brightness level of a video signal and determines a provisional representative value of the brightness level of the entire imaging screen; and correction that determines the amount of correction of the provisional representative value according to the brightness distribution within the screen. and an exposure control means for controlling exposure by changing an exposure adjustment amount according to the provisional representative value and the correction amount, and using fuzzy inference in determining the correction amount in the correction amount determination means. An automatic exposure adjustment device featuring: (2) A brightness evaluation value detection means that divides the imaging screen into a plurality of areas and detects the brightness level of each area as an exposure evaluation value, and calculates a temporary representative value representing the brightness level of the entire imaging screen from the exposure evaluation value. representative value calculation means; correction amount determining means for determining a correction amount for the temporary representative value according to the relative relationship of each exposure evaluation value; and changing an exposure adjustment amount according to the temporary representative value and the correction amount; An exposure control means for controlling exposure is provided, and the correction amount determining means uses at least an exposure evaluation value of each region or a value arithmetically obtained from the exposure evaluation value as an input variable, and uses the correction amount as a conclusion part to perform fuzzy inference. An automatic exposure adjustment device characterized by performing the following. (3) a brightness level detection means for detecting the brightness level of the video signal and determining a representative value of the brightness level of the entire imaging screen; and an exposure control means for controlling the exposure by changing the exposure adjustment amount according to the representative value. An automatic exposure adjustment device comprising: using fuzzy inference to determine a representative value in the brightness level detection means. (4) Exposure evaluation value detection means that divides the imaging screen into a plurality of areas and detects the brightness level of each area as an exposure evaluation value; priority determination means that determines the priority of each area; an arithmetic unit that performs a weighting process on the exposure evaluation values of each area to obtain a representative value of the brightness level of the entire screen according to the determined priority; and an exposure control unit that performs exposure control in accordance with the representative value; The automatic exposure adjustment device is characterized in that the priority determination means uses at least an exposure evaluation value of each region or a value arithmetically obtained from the exposure evaluation value as an input variable, and performs fuzzy inference using the priority as a conclusion part. .