JPH11202378A - Exposure controller and control method for the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an exposure controller and its control method capable of flexibly performing optimum exposure control with respect to each scene even in the case of exposure control performed when an image plane is divided with a fixed frame. SOLUTION: As to an electronic still camera 10; photometric areas divided based on a photometric value found by the photometric value calculating part 41 of an exposure control system 40 are arranged to plural groups by a grounding part 42, besides, the scene is discriminated by using an evaluation value for each group calculated by an evaluation value calculating part 43 on a scene discriminating part 44, so that the importance degree of each group can be calculated. An exposure value is decided by an exposure value calculating part 45, based on the importance degree, control is performed in accordance with the decided exposure value by an exposure control part 46, and a diaphragm and/or a shutter mechanism are controlled by a driving signal outputted in accordance with the exposure control. Therefore, the exposure adequate to a variety of scenes such as backlight, excessive front light, a scene with a flat background, and front light is performed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of measuring light incident on a field of view from an image pickup surface on which a light receiving element is arranged as a light measuring area, converting an obtained signal into a digital signal, and processing the signal into a digital signal. Exposure control device and method for performing exposure control suitable for the scene of the scene by performing the exposure control, particularly, it is suitable for use in silver halide cameras, video cameras, digital still cameras and the like that automatically perform exposure control is there.

[0002]

2. Description of the Related Art For example, in the case of portraits, landscapes, general photography, etc., in order to accurately record luminance (intensity of incident light) from a subject as image information, an image pickup system employs a photosensitive system. Exposure must be controlled so that the amount of light supplied to a solid-state imaging device, a film, or the like corresponding to a unit is appropriately adjusted. Exposure control is performed by controlling the aperture, shutter speed, and the like. The function of automatically performing this exposure control on the imaging system side is called automatic exposure control.

[0003] In this automatic exposure control, the whole screen average photometry using the average photometry value of the entire screen, the spot photometry using only the photometry value of a small area at the center of the screen, and the photometry value of the central portion of the screen are weighted and used. Center-weighted photometry to be evaluated, and the screen is divided into a plurality of fixed frame areas, photometric values are obtained for each area, a weighting factor is selected according to a pattern formed by the obtained photometric values, and the selected weighting factor and There is a method such as split photometry using a weighted average value obtained from each photometry value. For example, multi-pattern photometry and evaluation photometry are methods that have developed the split photometry.

[0004] However, even if exposure control is performed so that a part of a shooting scene (scene) is optimized, the luminance in the shooting scene may generally vary. For example, overexposure or underexposure occurs in other parts of the scene. In addition, none of the above-described photometric methods can perform effective exposure control for all shooting scenes. In particular, it is difficult to properly adjust the exposure to a main subject in a scene having a large unevenness in the brightness of the screen, such as backlight, over-direct light, or solid.

[0005] Some specific examples in which the problem of the exposure control is improved will be described. For example, Japanese Unexamined Patent Publication
No. 23, JP-A-4-71331, and JP-A-9-3714
In Japanese Patent Laid-Open Publication No. 5 (1993) -1995, basically, a method is used in which a screen is divided into grid-like areas, and the magnitude of luminance obtained from these areas is represented by a histogram (frequency distribution). JP Hei 4-3202
In Japanese Patent Laid-Open No. 23 and Japanese Patent Laid-Open No. Hei 9-37145, the frequency is determined based on the obtained luminance histogram. A neural network is used for the determination. In Japanese Patent Application Laid-Open No. Hei 4-213331, a target value of exposure control is obtained by calculating a tendency of a light beam state by performing fuzzy inference based on a luminance histogram.
Exposure control is based on the luminance histogram,
For example, it is determined whether the state is backlight, over-forward light, forward light, or the like.

[0006] Further, as an improvement example using the divided photometry,
For example, JP-B 5-34657, JP-A-4-310930,
There are JP-A-6-129906 and JP-A-6-148715. The screen has multiple areas (zones) for split metering
And each area is defined as a photometry area. In these improved examples, the subject luminance is basically obtained based on the luminance measured in each photometry area.

[0007] Further, as a feature of each improvement example, Japanese Patent Publication No. 5-34657 describes that a zone is concentrically divided, and the average brightness of the zones is weighted to calculate the subject brightness. I have. Japanese Patent Application Laid-Open No. 4-310930 discloses that, when colorimetric information is used to group photometric areas based on conditions in which adjacent photometric areas have similar colors, and when the center of gravity of the grouped areas is determined, the photometric areas are located at a relatively high position. It describes that the average luminance is obtained from the remaining groups excluding the high luminance group. JP 6-1299
Japanese Patent Application Publication No. 06-2006 discloses that a luminance value is calculated for each area or for each group according to the magnitude of a photometric output, and an appropriate exposure value is calculated for a scene based on information on the obtained brightness. Finally, Japanese Patent Application Laid-Open No. 6-148715 describes that a photometric area is grouped according to a photometric mode, and an appropriate exposure value of a scene is calculated.

[0008]

By the way, it is difficult to properly adjust the exposure of a main subject in a scene in which the luminance in the screen is largely uneven. In the above-mentioned split metering, exposure control can be performed to some extent in response to backlight or over-directed light.However, if the divided area is a fixed frame, the main subject or the screen component such as the sky is assumed to have a fixed frame. If they do not exactly match the area, the effect will not be exhibited.

Further, when the luminance photometric area is divided into a grid and a histogram is created based on photometric values from each of the divided areas to group the photometric areas, the state of the luminance distribution of the histogram is evaluated. By doing so, the backlight and over-direct light can be determined better than before, but since the determination is made within the fixed frame range, the fixed frame assumed by the main subject and the screen component such as the sky is assumed. There is a possibility that an erroneous determination is made that the area does not exactly match the area, and as a result, overcorrection is performed.

Further, as described in Japanese Patent Laid-Open No. 4-310930, the measurement area is divided according to color information without providing a fixed frame in the photometry area, and the divided areas are grouped together. Also, for example, it is possible to recognize the main subject in the scene including the sky and make an accurate determination of the scene, but since the conditions set at this time are very limited, the range applicable to the determination is narrow.

The present invention solves the above-mentioned drawbacks of the prior art, and provides an exposure control apparatus and an exposure control apparatus capable of flexibly and optimally performing exposure control for each scene even if the screen is divided by a fixed frame. The aim is to provide a method.

[0012]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, the present invention measures the luminance of incident light from an object scene by using an image pickup means in which an image pickup surface on which a light receiving element is arranged is used as a light measurement area. The output from the imaging means is converted into a digital signal by the A / D conversion means, and signal processing is performed on the basis of the obtained digital signal, whereby exposure control of the imaging means suitable for the scene of the scene is performed. In the exposure control device performed in
A photometric value calculating means for calculating a photometric value representing the luminance of the incident light for each block obtained by grouping a plurality of outputs from the D converting means, and a photometric value divided based on the photometric value obtained by the photometric value calculating means. A grouping means for grouping the areas into a plurality of groups, and an evaluation value indicating characteristics of the group based on the photometry values of each group, while grouping the photometry values from the photometry value calculation means according to the grouping means. Evaluation value calculation means for calculating the scene value, scene determination means for determining the scene of the scene based on the outputs from the evaluation value calculation means and the grouping means, and an object scene based on the result of the scene determination means. And an exposure value calculating means for calculating an optimum exposure value for the above scene.

Here, the grouping means includes frequency distribution investigating means for dividing the photometric value into a plurality of photometric value ranges, and the frequency distribution investigating means includes a size of one section for dividing the photometric value range, or a photometric value. It is preferable that the number of divisions in the range is arbitrary, and the divided photometric regions are grouped.

The evaluation value calculation means includes: a position evaluation means for calculating an average evaluation value in each group based on an evaluation value given to each block according to a distance from the center of the photometry area; It is desirable to have at least one of frequency evaluation calculating means for calculating the evaluation value and average photometric value evaluating means for calculating the evaluation value from the average of the photometric values of each group.

The grouping means selects one block arbitrarily selected from the groups collected by the frequency distribution checking means as a search reference candidate block in the group,
First memory means for storing information of the search reference candidate block, and a search reference candidate block initially stored in the first memory means are stored as a search reference block, and match the information of the search reference block. If there is a block adjacent to a block selected from among the search reference candidate blocks, the second block stores the information of the block, and if there is a block adjacent to the block selected as the search reference candidate block, the adjacent block is set as a search reference neighboring block. A third memory for storing information, a comparison and determination unit for comparing the information of the block stored in the second memory with information of a block near the search reference, and a search for the result of the comparison and determination. Check the spatial separation of groups based on the data in the second memory means storing the blocks that match the information of the reference block The loop separation checking means includes a first memory means, a second memory means, a third memory means, a comparison determining means, a group separation checking means, and a control means for controlling the frequency distribution checking means. It is even more advantageous to perform group classification by repeating the search in the imaging plane when the inspection result of the group separation inspection unit suggests the existence of a plurality of groups within the group, using the block position and the division information of the frequency distribution inspection unit. It is.

Further, the scene determining means compares the luminance of the subject in the object scene with the frequency distribution pattern obtained by the frequency distribution investigating means and the evaluation value calculated for each group by the evaluation value calculating means. Extremely bright backlit scenes, over-directed scenes in which the subject is extremely bright compared to the background brightness in the object scene, and photometric values in the object scene where the frequency of only a part of the metering value range remains It is preferable to distinguish between a scene with a flat background having a feature that is extremely high in frequency as compared with the range, and a scene with normal light with appropriate luminance of the subject and the background.

The scene discriminating means is a first subtraction for calculating a difference between a frequency in the photometric value range having the highest luminance obtained from the frequency distribution investigating means and a minimum frequency in the entire photometric value range as a difference on the upper side. Means, and second subtraction means for calculating a difference between a frequency in the lowest brightness photometric value range obtained from the frequency distribution investigating means and a minimum frequency in the entire photometric value range as a lower-side difference, Adding means for adding the difference on the lower side and the difference on the upper side; and third and fourth means for subtracting, from the output of the adding means, upper and lower thresholds used for determining a scene, respectively.
It is preferable to calculate an evaluation index used for correcting each scene.

When the divided group is further divided into a plurality of small groups, the scene discriminating means may preferentially use the evaluation value calculated from the smaller number of blocks in each small group.

The exposure value calculating means calculates the photometric value calculated for each group classified by the frequency distribution investigating means or the grouping means, the frequency of the photometric value, and the average evaluation value of the group calculated by the evaluation calculating means. Weighting factor calculating means for calculating a weighting factor for multiplying the photometric value based on the weighting factor, and exposure value determining means for calculating the sum of the results obtained by multiplying the weighting factor of the weighting factor calculating means and the photometric value for each group. desirable.

It is preferable that the exposure value calculation means corrects the weight coefficient of each group using the evaluation index obtained by the scene determination means.

The evaluation value calculating means includes an index of importance of a position occupied by each group in the photometry area, a frequency of each group,
Alternatively, it is advantageous to include storage means for storing an evaluation value corresponding to an average value of photometric values.

[0022] The evaluation value calculation means uses the evaluation value stored in the storage means as a reference, and further, based on an index of importance of the position occupied by each group in the photometry area and / or an index different from this index. It is preferable to correct the evaluation value of the storage means.

The exposure value calculating means corrects the photometric value in an area where the imaging surface is divided into a predetermined size based on the result of the scene determining means, and calculates the exposure value using the corrected photometric value. Is preferred.

According to the exposure control device of the present invention, the photometric areas divided based on the photometric values obtained by the photometric value calculating means are grouped into a plurality of groups by the grouping means, and the evaluation value calculating means evaluates each group. The scene value is calculated by the scene determination means, and the scene value is determined using the evaluation value of each group obtained by the scene determination means. The exposure value is calculated in accordance with the scene determined by the exposure value calculation means. Is controlled by the aperture and / or shutter mechanism. Thereby, for example, a backlight, an over-direct light, a scene with a flat background, a so-called solid, and a variety of scenes such as a direct light, accurately,
Exposure can be controlled accurately.

According to the exposure control method of the present invention, the luminance of the incident light from the object scene is measured using the image pickup surface on which the light receiving element is arranged as a photometry area, and a signal based on the photometry is converted into a digital signal. An exposure control method for controlling an exposure value obtained by performing signal processing based on a digital signal obtained from a digital signal to be suitable for a scene of an object scene. A photometric value calculating step of calculating a photometric value, a grouping step of grouping blocks based on the photometric value obtained in the photometric value calculating step, and a grouping based on the photometric values of the group compiled by the grouping step. An evaluation value calculation step of calculating an evaluation value representing the characteristic of the evaluation value obtained from the evaluation value and the grouping step obtained for each group from the evaluation value calculation step A scene determining step of determining a scene of the scene based on various frequencies; and an exposure value calculating step of calculating an optimal exposure value for the scene of the scene based on a result obtained by the scene determining step. It is characterized by:

Here, the grouping step includes a frequency distribution checking step of dividing the photometric value into a plurality of photometric value ranges, and further includes a step of dividing the photometric value range in the frequency distribution checking step.
The size of the section or the number of sections in the photometric value range is optional,
It is advantageous to group the divided photometric areas by the number of sections.

The evaluation value calculation step includes: a position evaluation step of calculating an average evaluation value within each group based on an evaluation value given to each block according to a distance from the center of the photometry area; The method may include at least one of a frequency evaluation calculation step of calculating an evaluation value and an average photometry value evaluation step of calculating an evaluation value from an average of photometry values of each group.

In the grouping step, one block arbitrarily selected from the groups collected by the frequency distribution checking step is used as a search reference candidate block and a search reference block in this group, and information on each block is stored in respective storage means. In the first information storage step, and one piece of information of the search reference candidate block stored in the first information storage step is read, and a plurality of blocks adjacent to the read search reference candidate block are located near the search reference candidate block. A second information storing step of storing the information of the search reference neighboring block in the storage means, determining whether the search reference neighboring block can be regarded as the same block as the search reference block, and continuing the comparison determination processing. A comparison judgment step for judging, and after the comparison judgment step is completed, And a group separation determining step of comparing the number of blocks that can be regarded as the same block and the total number of blocks when previously grouped into a group to determine group separation, and the comparison results of the group separation determining step match. It is desirable to repeat these series of processing until group classification.

The comparison determining step includes a neighborhood reading step of reading information of one search reference neighborhood block from the plurality of search reference neighborhood blocks stored in the second information storage step, and one search reference neighborhood extracted in the neighborhood reading step. A comparison / determination step of comparing and determining whether the information of the block is included in the information of the search reference block already stored, and a group determination of determining whether the search reference neighborhood block used in the comparison / determination step belongs to the selected group. When the search reference neighborhood block belongs to the selected group as a result of the step and this group discrimination step, the third information storage for storing the information as the search reference candidate block and the search reference block using the search reference neighborhood block determined by comparison. The process and the process destination to be shifted according to the number of search reference neighboring blocks after the third information storing process. It is advantageous to include a first determination step of disconnecting and a second determination step of determining, after the first determination step, a process destination to be shifted according to the number of search reference candidate blocks, whereby the spatial space of the group is reduced. Efficient separation.

In the exposure value calculating step, the image pickup surface is corrected to a photometric value in an area divided into a predetermined size based on the result of the scene determining step, and the exposure value is calculated using the corrected photometric value. It is desirable to do.

In the second information storing step, four to eight blocks located around the search reference candidate block selected in the first information storing step may be used as the search reference neighboring blocks.

According to the exposure control method of the present invention, the photometric regions divided by using the information obtained from the luminance distribution created in the frequency distribution investigating step are grouped into a plurality of groups in the grouping step, and each of the photometric areas is divided in the evaluation value calculating step. An evaluation value is calculated for the group, and in the scene determination step, a scene is determined using the evaluation value and the like, and an exposure value suitable for the scene determined in the exposure value calculation step is calculated, and according to the determined exposure value. The exposure control performed by the aperture and / or shutter mechanism to accurately and accurately control exposure to various scenes such as, for example, backlight, over-directed light, a scene with a flat background, and direct light. Can be.

[0033]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, an embodiment of an exposure control apparatus and a control method according to the present invention will be described in detail with reference to the accompanying drawings.

The exposure control apparatus according to the present invention measures the luminance of the incident light from the object scene while taking an image with the imaging surface on which the light receiving element is arranged as a photometry area, and measures the output obtained by the A / D.
A digital camera that performs exposure control suitable for the scene of the scene by performing signal conversion based on the obtained digital signal, and a silver halide camera that automatically performs exposure control. Used for video cameras, digital still cameras, and the like, it is possible to accurately and accurately apply to various scenes such as backlight, over-direct light, a scene with a flat background (hereinafter referred to as a so-called solid background), and normal light. There is a main feature of exposure control.

An exposure control apparatus according to the present invention will be described with reference to FIGS. FIG. 1 shows a schematic configuration when an exposure control device is applied to an electronic still camera 10, for example.

As shown in FIG. 1, the electronic still camera 10 includes an optical system 20, a signal processing system 30, an exposure control system 40, and a drive control system 50. The optical system 20 includes a lens 21, a diaphragm / shutter 22, and a CCD 23. This is a light incident mechanism for taking a desired subject image into the camera body and entering the image pickup surface of the CCD 23. The signal processing system 30 includes an A / D converter 31 and an image signal generator 32. The exposure control device corresponds to the exposure control system 40 shown in FIG. Exposure control system 40
Is a photometric value calculation unit 41, a grouping unit 42, an evaluation value calculation unit 4
3. It includes a scene determination unit 44, an exposure value calculation unit 45, and an exposure control unit 46. The drive control system 50 supplies a drive signal to the aperture / shutter 22 and the CCD 23 in accordance with the control from the exposure control system 40 to perform control.

As described above, in the optical system 20, the lens 21
Is an optical lens disposed substantially in front of a lens barrel (not shown). In FIG. 1, one lens having a predetermined focal length is advantageously used. Although not directly related to the present embodiment, zooming may be performed with the lens 21 movable. Further, a plurality of lenses may be used in combination. The incident light of the subject is supplied from the lens 21 to the imaging surface 23a of the CCD 23 through the diaphragm / shutter 22. CCD23
In the figure, light receiving cells for photoelectrically converting incident light are two-dimensionally arranged. The imaging surface 23a outputs an imaging signal from the light receiving cell to the signal processing system 30.

In the signal processing system 30, first, a photometric signal is supplied to an A / D converter 31. The A / D converter 31 converts the photometric signal into a digital signal and outputs the digital signal to the image signal generator 32, and also supplies the digital signal to the photometric value calculator 41 of the exposure control system 40. The image signal creating unit 32 performs signal processing on the supplied signal to an image signal of a television system and primary color signals of three primary colors R, G, and B. Image signal generator 32
Are output to a monitor or the like that visualizes such signal-processed signals, or are recorded on a recording medium.

In the exposure control system 40, the digitized image signal (image data) supplied from the A / D converter 31 is supplied to the photometric value calculator 41. The photometric value calculation unit 41
As shown in FIG. 2, the imaging data obtained from 23a is, for example, arranged in a grid pattern in M × N blocks B _ij (i = 0,..., M-1; j
= 0,..., N−1), and the image data of each block is combined to calculate the photometric value of the block. As a result, M × N photometric values are obtained from the imaging surface 23a.

The grouping section 42 includes a histogram creating section 42
a and a group classification unit 42b. The histogram creating unit 42a creates a histogram, considering the photometric values supplied from the photometric value calculating unit 41 in block units as a one-dimensional data array (see FIG. 3). The histogram creating unit 42a divides the level range of the photometric value supplied in creating the histogram into four sections D1, D2, D3, and D4, and indicates in which section the photometric value is included by the cumulative frequency. .

When dividing the level range of the photometric value, the histogram creating unit 42a determines how much the level range is set for one section or how many sections are set for the entire level range of the obtained photometric value. Can be arbitrarily selected.

The group classification unit 42b outputs the result supplied from the histogram generation unit 42a, that is, the divisions D1, D2, D3,
The imaging surface 23a is divided so that each block position and photometric value in D4 are treated as a set of data. In other words, there is no other way to group (classify) data belonging to the same section in a set of data thus obtained.
In the case of FIG. 4, the grouping unit 42 includes a histogram creation unit 42
It is set in advance so that the imaging surface 23a is treated as M × N = 64 blocks in a. By dividing the level range into four sections, the imaging plane 23a in FIG. 4 becomes four areas A1, A2, A3,
Divided into A4.

The evaluation value calculation unit 43 calculates an evaluation value based on the photometric values of the groups classified by the grouping unit 42. Here, the evaluation value is a numerical value representing the characteristic of each group. There are various methods for calculating the evaluation value. The evaluation value calculation unit 43 calculates the average evaluation value in each group based on the evaluation value given to each block according to the distance from the center of the photometric area by the first method, for example.
a, a frequency evaluation calculating unit 43b for calculating an evaluation value from the frequency of each group according to the second method, an average photometric value evaluating unit 43c for calculating an evaluation value from the average of the photometric values of each group according to the third method, etc. Contains. In the position evaluation unit 43a,
It is desirable to include at least one of the position evaluation unit 43a, the frequency evaluation calculation unit 43b, the average photometric value evaluation unit 43c, and the like. Among these evaluation methods, the position evaluation unit 43a uses a method that places importance on placing a main subject in the vicinity of the center of the screen when considering composition. in this way,
It is desirable to obtain an evaluation value by combining these methods in order to improve the accuracy of scene determination.

The position evaluator 43a provides a numerical value as an index of the importance of the position occupied by each group in the photometric area, that is, a numerical value indicating the spatial importance of a block according to the distance from the center of the screen (hereinafter, CPV: A lookup table 431 that stores a Center Priority Value) and a group CPV calculation unit 432 that calculates an average value for each group based on the CPV obtained from the lookup table 431. Look-up table 431, for example, numerical values according to the distance from the center C ₃₃ of the metering area is allocated in the case of the block as shown in FIG. 4 (see Figure 5). Also,
The group CPV calculation unit 432 calculates the CPV included in the group based on the CPV of each block, and calculates a numerical value indicating how close the group is to the center of the screen. The position evaluation unit 43a outputs the CPV of each block obtained from the lookup table 431 to the scene determination unit 44. The specific operation of the position evaluation unit 43a will be described later in detail.

Although not shown, the frequency evaluation calculating section 43b and the average photometric value evaluating section 43c also include a memory for storing an evaluation value corresponding to the frequency of each group and the average of the photometric values, respectively, and a numerical value from this memory. Includes a group average evaluation unit for calculating the average of the frequency and the photometric value for each group.
In this embodiment, the evaluation value calculation unit 43 sends the average values obtained by the position evaluation unit 43a, the frequency evaluation calculation unit 43b, and the average photometric value evaluation unit 43c to the scene determination unit 44. With these supplies, the determination accuracy in the scene determination unit 44 is further increased, and the adaptive range of the scene determination can be expanded.

Further, based on the evaluation value of the look-up table 431, the position evaluation unit 43a adaptively determines the importance of the position occupied by each group in the photometry area and / or an index different from the index. Alternatively, the evaluation value of the lookup table 431 may be corrected. As an index different from the CPV of each group, for example, the degree of rotation of the electronic still camera 10 and the aspect ratio of the screen are used.

The rotation of the electronic still camera 10 is detected by a sensor such as a mercury switch and supplied as an index indicating whether the camera is in a vertical or horizontal composition. When the index indicating the vertical composition is supplied, the lookup table 431 based on the horizontal composition is provided.
Is shifted to a lower position, for example, so as to reduce the influence of the sky. In this case, the upper part of the screen in the composition in the scene is likely to be empty.
The CPV is kept low, and the effect of the sky in backlit scenes can be reduced.

Further, a plurality of look-up tables may be prepared in advance and, for example, an operator may make a selection according to a scene. The numerical distribution of the CPV may be adaptively modified according to the change of the aspect ratio of the screen such as standard, wide, panorama, etc. supplied from the operation switch. With this configuration, it is possible to accurately capture the subject by evaluating the position in the screen and perform accurate exposure to the subject.

The scene discriminating section 44 includes a histogram creating section 42a
The scene of the object scene is determined based on the frequency distribution pattern obtained by the above and various evaluation values calculated for each group from the evaluation value calculation unit 43. As an example, the scene discriminating unit 44 sets the frequency in the photometric value range with the highest luminance (highlight group) among the data supplied from the histogram creating unit 42a to N _h , and the photometric value range with the lowest luminance (shadow group). The frequency at N _s , the minimum frequency N _min in the entire photometric range
It is represented by the variable In addition, a variable called N _{ud is used for} a difference between the frequency of the highlight group and the minimum frequency N _min (N _h -N _min ), and a variable called N _ld is used for the difference between the frequency of the shadow group and the minimum frequency N _min (N _s -N _min ). . In addition, for example backlight, excessive follow light, the threshold value T _u used for the determination of such solid backgrounds, are provided respectively to correspond to T _l in the variable N _ud and variable N _ld.

The scene discrimination unit 44, a subtraction function unit 44a for obtaining a difference N _ud, a subtraction function unit 44b for obtaining a difference N _ld, a summing function unit 44c for adding the obtained difference N _ud and difference N _ld, a subtraction function section 44d from the output of the summing function unit 44c draws a threshold T _u, includes a subtraction function section 44e subtracting the threshold T _l from the output of the subtraction function section 44d. These functional units are not shown here. The output of the subtraction function unit 44e is an evaluation index used for correcting each scene. Here, the evaluation index is indexed (index: hereafter, I _x
), And outputs the index _Ix to the exposure value calculation unit 45.

The scenes which can be determined by the scene discrimination section 44 include a backlit scene where the background is extremely bright compared to the brightness of the subject in the object scene, and an object whose extreme background is compared to the background luminance in the object scene. In bright over-directed scenes, in front-lit scenes in which the brightness of the subject and the background are appropriate, and in unprecedented object scenes, the frequency of some photometric value ranges is higher than that of the rest. There are scenes with a flat background, so-called solid backgrounds with extremely high features. The detailed determination will be described in the operation description.

When the divided group is further divided into a plurality of small groups, the scene discriminating section 44 may preferentially use the evaluation value calculated from the smaller number of blocks in each small group. . In this process, the subject having the larger number of blocks is emphasized. Index I _x may be corrected by the value of 0 to 1 by normalizing.

The exposure value calculator 45 includes a weight coefficient calculator 45a.
And an exposure function determining unit 45b. The weighting factor calculation unit 45a is the photometric value is calculated for each group classified by the histogram 42a or grouping classified 42b Gv _g [EV], the frequency N _g of the photometric value, a position evaluation unit of the evaluation calculation unit 43 Average rating of the group calculated by 43a
Each value of CPV _g is supplied. Weight coefficient calculator 45a
Calculates the final exposure value E from the sum of the result of multiplying these values and photometric values Gv _g as described above for each group. The exposure value calculator 45 determines the aperture and shutter speed from the calculated final exposure value E according to a program diagram.
The exposure value calculation unit 45 calculates the index obtained by the scene determination unit 44.
The final exposure value E 1 may be calculated by correcting the weight coefficient W _g of each group using I _x . The operation described later also describes this correction.

The exposure control section 46 outputs a control signal to the drive control system 50 based on the aperture and shutter speed values supplied from the exposure value calculation section 45. The drive control system 50 outputs a drive signal corresponding to the supplied control signal to the aperture / shutter 22 of the optical system 20 and the CCD 23, respectively. These components can appropriately adjust the amount of incident light according to the supplied drive signal. At this time, an electronic still camera
10, accurately determine whether the background of the scene or the scene of the subject corresponds to any scene such as backlight, over-direct light, solid background, etc.
Exposure control suitable for the determined scene is performed.

Next, the electronic still camera 10 to which the present invention is applied
Will be described with reference to the flowcharts of FIGS. 6 to 13 and the drawings such as screen patterns. FIG. 6 shows an operation in exposure control. Electronic still camera 10, for example, enters into any investigate whether to exposure mode for optimal image incident light from the scene (exposure control mode) while pressing the release button halfway, proceeds to step S10 I do.

In step S10, a control signal is supplied from the exposure control section 46 to the drive control system 50 in the exposure control mode so that exposure is performed at a preset aperture and shutter speed, and the flow advances to step S11.

In step S11, the aperture / shutter 22 and the CCD 2 related to the exposure are exposed according to the drive signal from the drive control system 50.
3 is controlled so that the CCD23 receives the incident light from the object scene. By this light reception, the incident light is photoelectrically converted by the CCD 23.

Next, in step S12, the output from the CCD 23 is supplied to the A / D converter 31 of the signal processing system 30 to be converted into a digital signal. This digital signal is output to the imaging surface 23 of the CCD 23.
This is the image data captured in a. In this case, the imaging surface 23
a corresponds to the photometry area. The image data is supplied to the image signal creation unit 32 of the signal processing system 30 and to the photometric value calculation unit 41 of the exposure control system 40.

Next, in step S13, the photometric value with the above-described settings is obtained by the photometric value calculation section 41 from the supplied image data (photometric value calculation step). Here, the photometric value calculation unit 41 divides the imaging surface 23a into a plurality of blocks, and calculates a photometric value for each block. The photometric value represents the luminance of incident light in one block. As shown in FIG.
For example, when the imaging surface 23a is divided into a plurality of blocks,
The imaging surface 23a is divided into M × N. As a result, the photometric value calculation unit 41 calculates M × N photometric values, and
Then, the process proceeds to step S14. As shown in FIG. 2, M = N =
The number of blocks in the case of 8 is 64.

In step S14, the supplied photometric values are sent to the histogram forming section 42a of the grouping section 42 to create a histogram of the photometric values. The histogram is obtained, for example, by dividing the entire photometric value range into four equal parts as shown in FIG. 3 and cumulatively counting photometric values falling within each of the sections D1, D2, D3, and D4. This is a group classification based on the level range of the photometric value.

In step S15, the group classification unit 42b sorts the data of the blocks in which the blocks are grouped for each of the groups into groups (grouping processing). Imaging surface
By grouping by photometric value division in 23a, four groups g (g = 0,..., 3) are obtained. Step S14, S15
Corresponds to a grouping process. After this processing, the procedure goes to step S16.

In step S16, the evaluation value calculating section 43 calculates an evaluation value representing the characteristic of the group based on the photometric values of the group thus collected (evaluation value calculating step). In this embodiment, a CPV (Center Priority Value) indicating how close the block is to the center of the screen is taken as a specific example of the evaluation value. As shown in FIG. 5, the CPV is assigned a large value to the block located at the center of the screen, and is set so as to approach zero as it approaches the periphery of the screen.
The evaluation value calculation unit 43 stores the CPV in, for example, a look-up table 431 in association with each block on the screen. The evaluation value calculation unit 43 obtains the CPV _g in the group by associating the classified group with the CPV in the lookup table 431. In general, the number of blocks included in a group is n, and the CPV of a block B _k in the group is
When expressed as (B _k ), the group CPV calculator 432 of the evaluation value calculator 43 divides the sum of the CPV (B _k ) of each block by n to obtain an equation (1)

[0063]

(Equation 1) Requires the group's CPV _g . Here, the subscript g indicates the number of the classified group. Equation (1)
For example, when the CPV _g of each group corresponding to the regions A1 to A4 shown in FIG. 4 is obtained, the group g = 0 and the group g =
3, CPV ₀ = 9/19 = 0.474 and CPV ₃ = 12/7 = 1.714 are obtained.

The evaluation value calculation step has been described in detail for the position evaluation step. However, the evaluation value calculation step is not limited to this step. The evaluation value is calculated from the frequency of each group (frequency evaluation calculation step), The evaluation value is calculated from the average value of the photometric values of (Average photometric value evaluation step).
Although at least one of the above-described steps must be used for calculating the evaluation value, it is more preferable to use a plurality of these steps. In the present embodiment, as shown in FIG. 1, the position evaluation calculation unit 43a, the frequency evaluation calculation unit 43b,
The result obtained by the average photometric value calculation unit 43c is supplied to the scene determination unit 44. As a result, as will be understood from the following description, the accuracy of the evaluation value is increased, and the applicable range of the scene determination is expanded.

Then, the process proceeds to a subroutine SUB1, in which a scene discriminating section 44 is determined based on a numerical value obtained based on various frequencies described later supplied from the histogram forming section 42a and an evaluation value of each group obtained from the evaluation value calculating section 43. To determine the scene of the scene (scene determination step). The types of scene determination include, for example, a backlight, an over-direct light, a scene having a solid background, and a normal-lit scene. In the subroutine SUB1, the index _Ix is calculated and supplied to the exposure value calculation unit 45 as described later. The index _Ix is used as an evaluation value indicating the degree to which the exposure value calculation unit 45 corrects the exposure value. After the end of the subroutine SUB1, the process moves to the subroutine SUB2.

In subroutine SUB2, subroutine SUB1
The exposure value calculation unit 45 calculates the optimum exposure value for the scene in the object scene based on the result obtained by (exposure value calculation step).
A weighting factor is calculated using the parameters obtained so far, and the weighting factor is multiplied by the photometric value of the group to calculate an exposure value (final exposure value). After calculating the exposure value, the aperture and the shutter speed are determined using the program value based on the exposure value. Step S1 after completion of subroutine SUB2
Go to 7.

In step S17, the exposure controller 46 is controlled based on the aperture and shutter speed obtained in the subroutine SUB2.
Outputs a control signal to the drive control system 50.

In step S18, the drive control system 50 supplies a drive signal to each part of the light incidence mechanism in accordance with the supplied control signal.

Finally, in step S19, the operation according to the supplied drive signal is performed by each unit to complete the optimal exposure setting. By controlling the exposure in accordance with such a procedure, it is possible to accurately determine various scenes and set an optimal exposure for the electronic still camera.

In order to describe the exposure control in the electronic still camera 10 in more detail, the above-described subroutine SUB
1 and SUB2 will be further described with specific examples. In the subroutine SUB1, the scene determination unit 44 uses the parameter value obtained based on the frequency as described above to determine the scene. The parameters include the frequency N _h of the photometric value range indicating the highlight group (highest luminance), the frequency N _s of the photometric value range indicating the shadow group (lowest luminance), the minimum frequency N _min of the entire photometric value range, The difference (N _h -N _min ) is N _ud , and the difference (N _s -N
_min) to N _ld, and, backlit, excessive follow light, T _u as a threshold that is set so as to correspond to N _ud and N _ld used for the determination of such solid background, there is a T _l.

First, as shown in FIG.
In step 10, the above-described parameters N _h , N _s , and N _min are set based on the result of the histogram. Also, threshold values _Tu and _Tl are set.

Next, in sub-step SS11, the difference (N _h -N _min )
Parameters by calculating the difference _{(N s -N min) N ud} , N ld
Is calculated.

In sub-step SS12, a process for determining whether or not the mode is bimodal is performed. Bimodal is called because it represents a low frequency in the middle area and a high frequency at both ends (highlight side and shadow side) of the middle area when expressed in a histogram, and forms two high peaks. ing. In this determination processing, (N _ud -T _u ) and (N _ld -T _l ) are calculated. When all the calculation results are positive values (Yes), it is determined to be bimodal, and the process proceeds to sub-step SS13. When both the calculation results are not positive values (No), the process proceeds to sub-step SS14.

[0074] Incidentally, the operations performed in this sub-step SS12 represents the evaluation value of the scene (the index I _x). Summarizing the above relationship, the index I
_x is generally the formula (2)

[0075]

[Number 2] represented by _{I x = N ud + N ld} -T u -T l ···· (2). The index _Ix is supplied to the exposure value calculation unit 45 as an evaluation value indicating a degree of correcting the exposure value. Index I _x may be supplied normalized.

Next, in sub-step SS13, it is determined whether this scene is backlit or over-directed. For this scene determination, the CPV _g of each group is used. At this time, the CPV _{g of the} group in the highlight group and the group of CPV _g in the shadow group classified in association with the histogram section are compared. When the CPV _{hg of the} highlight group is larger than the CPV _{sg of the} shadow group, it is considered as over- _directed . When the CPV _{sg of} the shadow group is larger than the CPV _{hg of the} highlight group, it is regarded as backlight.

Next, in sub-step SS14, it is determined whether or not the scene has a solid background. In this scene discrimination,
Discrimination is performed by using the histogram scene analysis and the group position (or spatial) evaluation value CPV _g . In this scene, it is known from these analyzes that there is often one outstandingly high-frequency segment. Here, the frequency of the highest frequency group in the analysis is N _max ,
CPV is set to CPV _max.
And CPV _else on average V _g. In this determination, it is also necessary to check whether the frequency corresponding to, for example, the background exists in the scene by using the predetermined threshold value T _base . CPV _max
It is also necessary to compare CPV _else and examine whether CPV _max is out of the center of the scene. In consideration of these, the condition for determining a scene having a solid background is that N _max is larger than a predetermined threshold T _base and CPV _else is larger than CPV _max . When this determination condition is satisfied (Yes), it is determined that the background is solid, and otherwise, it is determined that the light is normal. After this determination, the process proceeds to return and the subroutine ends.

The operation of the subroutine SUB1 will be described with reference to specific histograms shown in FIGS. The scene determination in FIG. 8, for example, each parameter value defined in the foregoing, when using the _{N h = 24, N s =} 30, N min = 2, T u = 18, T l = 10, N ud = 22, N _ld = 30 is obtained. Sub-step SS12
As is clear from FIG. 8, the scene in FIG. 8 is (N _ud -T _u )> 0, and
Since (N _ld -T _l )> 0, it is determined to be bimodal. This bimodal scene includes a scene such as a backlight or an over-directed light. The distinction between over-directed light and backlit light is made by comparing the highlight group and the shadow group classified by the histogram forming unit 42a with the CPV _{g of} each group, and comparing the highlight group CPV _hg and the shadow group described above. CPV _sg
I go from the size relationship.

FIG. 9 shows an example of determining a scene having a solid background which has not been regarded as a problem in the past. In this case, for example, according to the analysis described above, the frequency of the highest frequency group N _max = 48, the CPV of this group is CPV _max = 0.70 _{, and} the CPV _{g of the} remaining groups excluding this group is averaged to obtain the CPV.
_{When else} = 1.07 is obtained, since N _max is larger than the predetermined threshold T _base and CPV _else is larger than CPV _max , the scene determining unit 44 determines that the background is solid.

Next, the subroutine SUB2 will be described with reference to FIG. In sub-step SS20, the photometric value Gv _g of each group, the frequency N _g, the process proceeds to substep SS21 is set location estimate CPV _g, the index I _x, respectively.

In the sub-step SS21, an equation (3) obtained by multiplying the frequency N _g by the position evaluation value CPV _g

[0082]

[Mathematical formula-see original document] w _g = N _g * CPV _g ... (3), the weight coefficient w _g of each group is calculated. In this sub-step SS21, the normalization coefficient k for normalizing the weight coefficient w _g obtained here is also calculated by the equation (4).

[0083]

(Equation 4) Is calculated using The normalization coefficient k is multiplied by the weighting coefficient w _g to obtain the equation (5),

[0084]

[Number 5] sub-step to calculate the _{w gn = k * w g ····} (5) the weighting coefficient w _gn that has been standardized from
Proceed to SS22.

In the sub-step SS22, the standardized weight coefficient w _gn obtained in the previous step is corrected in order to perform optimal exposure control. This correction is performed using the evaluation value (index _Ix ) calculated according to the scene determined by the scene determination unit 44. Index I _x is a real value from 0 to 1 using normalized. The index I _x is given by equation (2)
As can be understood from FIG. 5, when the value greatly exceeds the threshold value, the degree increases, and when the value is not so different from the threshold value, the degree is small and proportional to the difference between the threshold value.

Assuming that the weighting factor corrected for the group regarded as important by the scene determination is W _g , the corrected weighting factor W _g is a value between the uncorrected weighting factor w _gn and a maximum of 1. You can ask for it. Therefore, the corrected weighting factor W _g is given by equation (6)

[0087]

(6) W _g = w _gn + (1-w _gn ) * I _x (6)

Incidentally, since the sum of the weighting factors needs to be 1, the weighting factors W other than the corrected weighting factors W
_{g_else} is given by equation (7)

[0089]

W _{g_else} = W _{g_else} -W _{g_else} * I _x ··· (7) After this calculation, the process proceeds to sub-step SS23.

In the sub-step SS23, the final exposure value E is calculated by the equation (8) using the exposure value corrected according to the scene determined by the scene determination section 44.

[0091]

(Equation 8) It is calculated by: However, when the uncorrected CPV _g in the previous sub-step SS22 has the same value regardless of the group, the final exposure value E is simply an average exposure value. By calculating the final exposure value E in this way, exposure can be adjusted to a region of a group having high importance in a scene, which is compatible with the scene. After the exposure value is obtained in this manner, the process proceeds to sub-step SS24.

In sub-step SS24, the aperture and the shutter speed are determined based on the calculated final exposure value using a program diagram. Based on the determined aperture and shutter speed, a control signal is transmitted from the exposure control unit 46 to the drive control system 50.
Output to After this process, the process returns to the subroutine SUB2 and ends.

As a more specific example of the subroutine SUB2, the calculation of the weight coefficient in the bimodal scene shown in FIG. 8 will be described using numerical values. When the parameters defined above were classified into groups g = 0 to g = 3, the data shown in Table 1 was obtained.

[0094]

[Table 1] Using the equation (3), the weight coefficient w _g for each group (that is, w ₀
= 15, w ₁ = 9.6, w ₂ = 3.0, w ₃ = 48). From equation (4), the normalized coefficient was k = 0.0132. A weighting coefficient w _gn standardized by applying Expression (5) is obtained for each group. As the normalized weight coefficient w _gn , the numerical value shown at the bottom of Table 1 was obtained for each group. When the final exposure value was obtained without correction, 14.224 [EV] was obtained by the integration of each group by the same calculation as Expression (8). As a comparison, as described above, the position evaluation value CPV _g does not depend on the group (that is, the spatial importance is equally the same), and the weighting factor is simply a value obtained by dividing the frequency by the sum of the frequencies. become. The final exposure value in this case is approximately 12.6 [EV]. Compared to the comparative example in which the final exposure value is the average photometric value, it is understood that the result of calculating the exposure value is subjected to exposure control by focusing on Group 3 in Table 1 having high importance.

Further, when the correction is performed in consideration of the scene discrimination, the weight coefficient W ₃ = 0.8905 of the group 3 and the weight coefficient W _{0 of the} other groups are obtained by using the equations (6) and (7). = 0.0594, W
₁ = 0.0381 and W ₂ = 0.012 were obtained, respectively. By substituting these weighting factors into equation (8) and calculating, the optimal final exposure value
E was 15.5 [EV]. The calculation of this corrected exposure value is
This indicates that the photometry of the photometry area 23a requires an exposure that is larger than the final exposure value 14.2 [EV] before correction and is closer to the group 3, and that a luminance difference from the group around the group 3 of interest can be obtained.

The exposure controller 46 determines an aperture and a shutter speed using a program diagram based on the calculated final exposure value, and outputs a control signal to the drive control system 50. When the drive control system 50 drives and exposes each unit, the electronic still camera 10 performs exposure suitable for this scene.

[0097] Thus, the exposure can be controlled according to the scene by determining the scene by integrating the photometry into the fixed area and correcting the weighting factor, thereby eliminating erroneous determination of the exposure by the photometry. Become like

Next, the electronic still camera 10 to which the present invention is applied
A description will be given of a determination as to whether the group put together in is composed of spatially divided small groups. The electronic still camera 10 has the configuration of the grouping unit 42 in FIG.
The process of the flowchart of 13 is changed.

In this case, the group classification section 42b of the grouping section 42 includes memories 421, 422, 423, a comparison determination section 424, a spatial arrangement investigation section 425, and a group classification control section 426. The memory 421 stores a histogram forming unit under the control of the group classification control unit 426 in the group.
One block arbitrarily selected from 42a is a search reference candidate block in this group, and is a memory for storing information of the search reference candidate block.

The memory 422 is a memory for storing the search reference candidate block initially stored in the memory 421 as a search reference block, and for storing information of a block that matches the information of the search reference block. In other words,
The result of the final search is a memory for storing only information of blocks in the same group. The memory 423 is a memory that, when there is a block adjacent to a block selected from the search reference candidate blocks, sets the adjacent block as a search reference neighboring block and stores information of the search reference neighboring block. The adjacent blocks are read from the histogram generator 42a to the memory 423 under the control of the group classification controller 426 and stored.

The comparison / determination unit 424 includes a group classification control unit 42
Under the control of 6, the information of the block near the search reference in the memory 423 is read out one by one and compared with the information of the block stored in the memory 422 and determined. And a group classification control unit
Under the control of 426, when the result belongs to the same group as the group classified by the histogram conversion but has not been stored in the memory 422 yet, the comparison / determination unit 424
Stores the information of the search reference neighborhood block in the memory 423 in the memory.
422. At this time, the memory 422 stores the data of this block under the control of the group classification control unit 426.

The space arrangement examination unit 425 examines the spatial separation of groups based on the data in the memory 422. Memory 422
Is compared with the number of blocks in the group selected by the histogram forming unit 42a. When the numbers of both blocks are the same, the group is one and is not spatially separated. When the number of blocks stored in the memory 422 is smaller than the number of blocks of the group selected by the histogram generator 42a, it is determined that the groups are spatially separated. This result is output as the result of group classification. A signal based on this result is output to a group classification control unit 426.
Supplied to

The group classification control unit 426 includes the memories 421, 4
22, 423, a comparison / determination unit 424, a spatial arrangement investigation unit 425, and a histogram generation unit 42a. In particular, when the signal based on the result of the investigation by the spatial arrangement investigation unit 425 suggests the existence of a plurality of groups in the group, the group classification control unit 426 performs control for repeating the search in the imaging plane 23a to perform the group classification. In other words, the comparison determination unit 424 compares the remaining blocks with the segment information, and repeats the determination.

At this time, the position information and the division information of the histogram forming section 42a are used as the block information. For example, a case where the screen (for example, the imaging surface 23a) shown in FIG. 12 is grouped into groups by photometry will be described with reference to the flowchart of FIG. Here, the photometry area 23a
Each block which is a component of the above is assigned classification information in which the obtained photometric values are classified by the histogram forming unit 42a, that is, information in which the levels of the photometric values are classified. The histogram generation unit 42 is controlled by the group classification control unit 426.
The information of the selected block is supplied from a to the group classification unit 42b. Here, the selected block is a small group a1 of the group A. Using this information, spatial group classification is performed in subroutine SUB3. This subroutine SU
B3 includes a first information storage step and a second
, A comparison determination step, and a group separation determination step. Thus, the subroutine SU
In B3, after the processing by histogram generation in the histogram generation unit 42a, the sub-step SS30 starts to further determine how blocks in the group are spatially related. That is, the subroutine SUB3 is a small group classification determination process.

In sub-step SS30, one block arbitrarily selected is set as a search reference candidate block and a search reference block in the group by the histogram generator 42a.
The information of each block is stored in the memories 421 and 422, respectively (first information storage step). This allows the memory 422
Stores a block as a small group a1. The information of this block functions as a reference for the group. Thereafter, the process proceeds to sub-step SS31.

In sub-step SS31, sub-step SS30
Is read, and a plurality of blocks adjacent to the read search reference candidate block are set as search reference neighboring blocks, and the information of the search reference neighboring blocks is stored in the memory 423 (second example).
Information storage process). 4 around the search criteria candidate block
Or 8 blocks. For example, if it is set so as to store the four surrounding blocks, four are stored in the memory 423 each time the information of the search reference candidate block is read. Here, reading of information from the memory naturally disappears from the memory.

Thereafter, the sub-step SS32 to the sub-step SS
Until 38, the processing of the comparison judgment step is performed. In the comparison determination step, while comparing the information of the search reference candidate block and the search reference neighboring block, it is determined whether the search reference neighboring block can be regarded as the same block as the search reference block,
It is determined that the comparison determination process is continued.

First, in the sub-determination step SS32, in the sub-step SS32, information on one search reference neighboring block is read from the plurality of search reference neighboring blocks stored in the memory 423 in the sub-step SS31, and the process proceeds to the sub-step SS33 (neighbor reading). Process). In sub-step SS33, sub-step
It is determined whether the information of one search reference neighboring block read by the SS 32 is included in the information of the search reference block already stored (comparison determination step). If it is determined that the selected search reference neighboring block contains the same information as the stored search reference block information (for example, position information) (Yes), it is determined that the search has been performed, and the process proceeds to sub-step SS36. . If the selected search reference neighboring block does not include information of a block that matches information of the stored search reference block (for example, position information) (No), the process proceeds to the next sub-step SS34.

In sub-step SS34, the previous sub-step
It is determined whether the search reference neighborhood block extracted in SS33 belongs to the selected group (group determination step). The determination as to whether or not a user belongs to a group is confirmed by, for example, a group number or division information. If the search reference neighboring block is information of a block that does not belong to the group A (No), the process proceeds to sub-step SS36. If the search reference neighboring block belongs to group A (Yes
), And proceed to sub-step SS35.

In the sub-step SS35, the sub-step SS34
The information is stored in the memories 421 and 422 as the search reference candidate block and the search reference block using the search reference neighborhood block from the data storage unit (third information storage step). The search reference neighborhood block is stored in the memory 421 so that the search reference neighborhood block is used as a new search reference candidate block for the search. Further, the memory 422 of the search reference neighborhood block includes:
In this case, the storage of the blocks collected as the small group a1 included in the preset group A is stored.

In sub-step SS36, the process destination to be shifted is determined according to the number of blocks near the search reference in memory 423 (first determination process). When data is stored in the memory 423 (Yes), the process proceeds to sub-step SS32. If there is no data in the memory 423 (No), the process proceeds to sub-step SS37.

In sub-step SS37, sub-step SS36
After this, the process destination to be shifted is determined according to the number of search reference candidate blocks in the memory 421 (second determination process). If data is stored in the memory 421 (Yes), the process proceeds to the sub-step SS31. In this case, the information of the search reference candidate block is read, and the subsequent processing is performed.
If no data is stored in the memory 421 (No), the process proceeds to sub-step SS38. As a result of the series of comparison determination processing, the number of obtained blocks that can be regarded as the same block is stored in the memory 422.

In sub-step SS38, the separation of groups is determined by comparing the number of blocks with the total number of blocks in a group (group separation determining step).
When the comparison result does not match (Yes), it is determined that another small group exists at the position where the spatial arrangement is shifted, and the memory is determined so far.
A block other than the block stored in 423 is set, and the flow shifts to sub-step SS30. If a series of these processes are repeated and the number of repetitions is counted until the comparison result of the group separation determination process matches, the number of small groups can be determined by this group classification process. When the comparison result matches (No), it is determined that there is no division of the small group, and the process proceeds to the return. After proceeding to return, this subroutine SUB3 ends.

By operating as described above, it is possible to know the presence or absence of another group that is spatially separated even though it is in the same section, and other spatially separated and thus arranged screens are formed. Since the group is very likely to be an object different from one of the objects, the CPV of each of the small groups will be different by obtaining the CPV according to the presence or absence of the spatially divided small groups. If a small group is reflected while the group is the same, the accuracy of scene determination can be improved.

Actually, for example, when it is clear that the same group A is divided into two small groups a1 and a2, the number of blocks in the group A is 12, CPV = 1.8, and the group a1
Then, the number of blocks is 10, CPV = 2.0, the number of blocks in group a2 is
2. When CPV = 0.8, a group a1 having a high composition ratio is prioritized and treated as a representative value of the group. This allows CPV
Could be changed to a value closer to the actual one.

Although the evaluation value CPV in the position evaluation unit 43a of the evaluation value calculation unit 43 in the above-described embodiment uses a look-up table, the evaluation value CPV is obtained by direct calculation so that it can be used more generally. You may. In this case, the block having the highest importance is specified in advance. The center coordinates of this block are B (X _C , Y _C ). If the coordinates of a block are
When B (X, Y), the distance D from the center coordinate is

[0117]

D = {(XX _C ) ² + (YY _C ) ² } ^1/2 ... (9) This distance D increases as the distance from the center increases. Based on this, the evaluation value CPV _g can be obtained by subtracting the distance D from a predetermined value. Also, this CP
A more flexible value can be obtained by multiplying V _g by a coefficient.

Next, the electronic still camera 10 to which the present invention is applied
Another embodiment will be described with reference to FIGS. 14 and 15. FIG. Here, the electronic still camera 10 performs photometry using a conventional photometry method. As a known divisional photometry method, for example, there is a method of dividing a screen (that is, the imaging surface 23a) into a central portion 230 and a peripheral portion 231 as shown in FIG.

Up to now, the luminance of the central part 230 and the peripheral part 231 has been measured by this division, scene determination has been made based on the luminance difference, and a weight coefficient for each divided area has been determined. The exposure value is determined by performing weighted averaging on each of the divided areas using the weight coefficients. In the case of scene determination, for example, when the photometric value of the central portion 230 is higher than the photometric value of the peripheral portion 231 and higher than a predetermined value set in advance, the scene is determined as over-ordered light, and Weighted averaging was performed by increasing the weight for the photometric value. Central 230
If a person corresponds to, the exposure could be controlled so that the person would be better captured. Conversely, the photometric value of the peripheral part 231 is
When the photometric value is higher than 230 and higher than a predetermined value set in advance, the scene is determined to be backlit, and the weight for the photometric value of the peripheral portion 231 is reduced to perform weighted averaging.
As a result, when a person corresponds to the central part 230, the person is prevented from being darkened.

Incidentally, in such a divided photometry system, since the divided area is a fixed frame, the effect is not exerted unless the shapes of the components of the screen match the inside of the fixed frame, so that an erroneous determination occurs as described. There was a case. For example, figure
When shooting an over-directed scene such as that shown in Fig. 15, a bright subject is placed at a position shifted from the center 230 of the screen. The part 230 cannot be recognized as bright. As a result, a sufficient difference in luminance between the central portion 230 and the peripheral portion 231 cannot be obtained, and the condition of over-directed light is not satisfied.

Therefore, the electronic still camera 10 equipped with an exposure control based on a divided photometry method for the central part 230 and the peripheral part 231 is mounted.
When the present invention is applied to the present invention, photometry is performed on each block obtained by dividing the screen into grid-like blocks as shown in FIG. Grouping processing is performed based on the photometric values in each block. The bright subject portion 232 is classified as a highlight group. This highlight group is set to a high value as having a positional significance due to the application of CPV. Also, based on the result of the histogram shape analysis performed when each block is grouped, the scene in the scene is determined to be over-directed. This indicates that the scene is over-illuminated and that the bright subject is a highlight group.

Based on this result, the photometric value Gv
When c is calculated, it is considered that it is sufficient to consider the over- _directed light and approach the average photometric value Gv _{h_av} of the highlight group, so the corrected photometric value C_Gvc of the central part 230 is the degree of the over-directed light. Overlung index ofl_Ix (over front lig
ht index)

[0123]

C_Gvc = Gvc + (Gv _{h_av} -Gvc) * ofl_Ix (10) The overordering light index ofl_Ix is an evaluation value of 0 or more and 1 or less. The overordering light index ofl_Ix is calculated based on the magnitude of the bimodal frequency, the magnitude of the CPV of the highlight group, and the like. C_Gvc = Gvc when the overordering light index ofl_Ix is 0
When the overordering light index ofl_Ix is 1, C_Gvc = _{Gvh_av} . In this manner, the correction allows the photometric value of the central portion 230 to be highly evaluated. As a result, the central part 23
The difference in luminance between 0 and the peripheral portion 231 is obtained. The correction may be performed for the peripheral portion 231 by the same method. By performing the same exposure control as in the past based on the photometric values of the two regions obtained in this way, the existing hardware assets can be used as they are without making the erroneous determination that occurred in the conventional split photometry method. Exposure can be controlled by utilizing this. An electronic still camera 10 according to another embodiment includes:
As a result, the exposure control can be flexibly performed by calculating the optimum exposure value even for the exposure control in the fixed frame.

With this configuration, the photometric areas divided based on the obtained photometric values are grouped into a plurality of groups, and the scene is determined using the evaluation values calculated for each group. In addition to scene discrimination, for example, using the calculated importance of each group, an exposure value is calculated based on these, exposure control is performed in accordance with the determined exposure value, and output is performed in accordance with this control. By controlling the aperture and / or shutter mechanism by the drive signal, it is possible to accurately control various scenes such as a backlight, an over-direct light, a scene having a flat background, a scene having a so-called solid background, and a front light. Since exposure control can be performed accurately, it is possible to perform optimal exposure control for each scene, and to always obtain a high-quality image without, for example, overexposure and underexposure. Kill.

[0125]

As described above, according to the exposure control apparatus and the control method of the present invention, the photometric areas divided based on the photometric values obtained by the photometric value calculating means are grouped into a plurality of groups by the grouping means. Further, a scene is determined by using the evaluation value calculated by the evaluation value calculating unit for each group as a scene determining unit, and the exposure value is calculated and determined by an exposure value calculating unit based on information such as the determination of the scene. Exposure control means performs control in accordance with the exposure value. By controlling the aperture and / or shutter mechanism with a drive signal output in accordance with the exposure control, for example, backlight,
Exposure control can be performed accurately and accurately for various scenes, such as over-directed light, scenes with a flat background, so-called solid backgrounds, and direct light. , And a high-quality image without, for example, overexposure or underexposure can be obtained.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a schematic configuration of an embodiment when an exposure control device according to the present invention is applied to an electronic still camera.

FIG. 2 is an arrangement pattern diagram showing an arrangement of divided blocks on an imaging surface in the CCD of the optical system shown in FIG. 1;

FIG. 3 is a histogram showing photometric values obtained by the electronic still camera of FIG. 1 in terms of cumulative frequency for each section range.

FIG. 4 is a diagram showing a group pattern when the histograms of FIG. 3 are spatially displayed and grouped for each group.

5 is a distribution pattern diagram of a CPV value of a look-up table used in the position evaluation value calculator shown in FIG.

FIG. 6 is a main flowchart showing a procedure of exposure control in the electronic still camera of FIG. 1;

FIG. 7 is a flowchart illustrating a procedure of a subroutine SUB1 of FIG. 6;

8 is a histogram showing an example of a bimodal photometric value obtained when an image of a subject is captured by the electronic still camera in FIG.

9 is a histogram illustrating an example of a photometric value having a solid background obtained when an image of a subject is captured by the electronic still camera in FIG.

FIG. 10 is a flowchart illustrating a procedure of a subroutine SUB2 of FIG. 6;

FIG. 11 is a schematic block diagram illustrating a configuration of a grouping unit used in a modification of the above-described embodiment.

FIG. 12 is a diagram showing an example of a group pattern when the histograms used in the modification of FIG. 11 are spatially displayed and grouped for each group.

FIG. 13 is a flowchart illustrating a procedure for classifying small groups in the group pattern of FIG. 12;

FIG. 14 is a schematic diagram showing a state of a divided screen used for exposure control in another embodiment of the electronic still camera to which the present invention is applied.

15 is a schematic diagram showing an example of a screen configuration when a bright subject is arranged on the screen of FIG. 14 at a position shifted from the center.

[Explanation of symbols]

 10 Electronic still camera 20 Optical system 30 Signal processing system 40 Exposure control system 41 Metering calculation unit 42 Grouping unit 43 Evaluation value calculation unit 44 Scene discrimination unit 45 Exposure value calculation unit 42a Histogram conversion unit 42b Group classification unit

Claims (19)

[Claims] 1. An image pickup device in which the luminance of incident light from an object scene is measured by an image pickup surface having a light receiving element as a light measurement area, and an output from the image pickup device is converted into a digital signal by an A / D converter. The exposure control device performs the exposure control of the imaging unit suitable for the scene of the object scene by performing signal conversion based on the obtained digital signal. A photometric value calculating means for calculating a photometric value representing the luminance of the incident light for each block in which a plurality of outputs from the D converting means are grouped, and division is performed based on the photometric value obtained by the photometric value calculating means. Grouping means for grouping the photometric regions into a plurality of groups; and photometric values from the photometric value calculating means for each group according to the grouping means, and representing characteristics of the groups based on the photometric values of each group. Evaluation Evaluation value calculation means for calculating the scene value, scene determination means for determining a scene of the object scene based on outputs from the evaluation value calculation means and the grouping means, and the scene determination means based on the result of the scene determination means. An exposure control device, comprising: an exposure value calculation unit that calculates an exposure value optimal for a scene of a scene. 2. The exposure control device according to claim 1, wherein said grouping means includes a frequency distribution investigating means for dividing said photometric value into a plurality of photometric value ranges, wherein said frequency distribution investigating means comprises: An exposure control apparatus, wherein the size of one section for dividing a range or the number of sections of the photometric value range is arbitrary, and the divided photometric areas are grouped. 3. The exposure control device according to claim 1, wherein the evaluation value calculation means is configured to calculate an evaluation value within each group based on an evaluation value given to each of the blocks in accordance with a distance from a center of the photometry area. Position evaluation means for calculating an average evaluation value, frequency evaluation calculation means for calculating an evaluation value from the frequency of each group, and average photometric value evaluation means for calculating an evaluation value from the average of the photometric values of each group. An exposure control device comprising at least one of the following. 4. The exposure control device according to claim 2, wherein the grouping unit selects one block arbitrarily selected from the group collected by the frequency distribution investigating unit as a search reference candidate block in the group,
First memory means for storing information on the search reference candidate block; and a search reference candidate block initially stored in the first memory means being stored as a search reference block, and matching with the search reference block information. A second memory unit for storing block information; and a block adjacent to a block selected from the search reference candidate blocks, the adjacent block being a search reference neighboring block, Third memory means for storing information of the following; comparison / judgment means for judging the information of the block stored in the second memory means and the information of the block near the search reference; The spatial memory of the group based on the data in the second memory unit storing the block that matches the information of the search reference block. Controlling the first memory means, the second memory means, the third memory means, the comparison determining means, the group separation checking means, and the frequency distribution checking means. Using the division information of the position and frequency distribution investigating means of the block as the information, when the inspection result of the group separation investigating means suggests the existence of a plurality of groups in the group, An exposure control apparatus characterized in that a search is repeated to perform group classification. 5. The exposure control device according to claim 1, wherein the scene determining unit determines a frequency distribution pattern obtained by the frequency distribution investigating unit, and a frequency distribution pattern obtained by the evaluation value calculating unit for each group. On the basis of the evaluation value calculated in the above, a backlight scene where the background is extremely bright compared to the luminance of the subject in the object scene, and the subject is extremely bright compared to the background luminance in the object scene A front light scene, a flat background scene having a feature that the frequency of only a part of the photometric range is extremely high compared to the rest of the photometric range in the object scene, and the brightness of the subject and the background. An exposure control device that determines an appropriate front light scene. 6. The exposure control device according to claim 1, wherein the scene discriminating means includes a frequency in a photometric value range with the highest luminance obtained from the frequency distribution investigating means and a total photometric value range. A first subtraction means for calculating a difference between the minimum frequencies of the light measurement values as a difference on the upper side, a frequency in a photometry value range having the lowest luminance obtained from the frequency distribution investigation means, and a minimum frequency in the entire photometry value range. Second subtraction means for calculating the difference of the lower side as a difference on the lower side, addition means for adding the difference on the lower side and the difference on the upper side, An exposure control apparatus comprising: third and fourth subtraction means for subtracting lower thresholds, respectively, and calculating an evaluation index used for correction of each scene. 7. The exposure control device according to claim 1, wherein the scene discriminating unit determines whether each of the divided groups is divided into a plurality of small groups. An exposure control apparatus characterized in that an evaluation value calculated from the smaller number of blocks in a small group is used preferentially. 8. The exposure control device according to claim 1, wherein the exposure value calculating means calculates a photometric value calculated for each group classified by the frequency distribution investigating means or the grouping means. Weighting coefficient calculating means for calculating a weighting coefficient by which the photometric value is multiplied based on the frequency and the average evaluation value of the group calculated by the evaluation calculating means; And an exposure value determining means for calculating a sum of the results of multiplication for each exposure. 9. An exposure control apparatus according to claim 8, wherein said exposure value calculating means corrects the weight coefficient of each group using an evaluation index obtained by said scene discriminating means. Control device. 10. The exposure control device according to claim 1, wherein the evaluation value calculating means includes an index of importance of a position occupied by each group in the photometry area, a frequency of each group, or the frequency of each group. An exposure control device including a storage unit that stores an evaluation value corresponding to an average value of photometric values. 11. The exposure control device according to claim 1, wherein the evaluation value calculation unit uses an evaluation value stored in the storage unit as a reference,
Further, the exposure control device adaptively corrects the evaluation value of the storage means based on an index of importance of a position occupied by each group in the photometry area and / or an index different from the index. 12. The exposure control device according to claim 1, wherein the exposure value calculating means converts a photometric value in an area where the imaging surface is divided into a predetermined size based on a result of the scene determining means. An exposure control apparatus, wherein an exposure value is corrected and an exposure value is calculated using the corrected photometric value. 13. A method for measuring the brightness of incident light from an object scene by using an image pickup surface on which a light receiving element is disposed as a light measuring area, converting a signal based on the light measuring into a digital signal, and based on the obtained digital signal. An exposure control method for controlling an exposure value obtained by performing signal processing so as to be suitable for the scene of the object scene, the method comprising: converting the digital signal into a luminance of the incident light for each block obtained by dividing the photometric area. A photometric value calculating step of calculating a photometric value representing the following; a grouping step of grouping the blocks into groups based on the photometric value obtained in the photometric value calculating step; and a photometric value of the group collected by the grouping step. An evaluation value calculation step of calculating an evaluation value representing a characteristic of a group based on the evaluation value obtained for each group from the evaluation value calculation step and the grouping step A scene determining step of determining the scene of the scene based on various frequencies obtained; and an exposure value calculating step of calculating an optimal exposure value for the scene of the scene based on a result obtained by the scene determining step. And an exposure control method. 14. The control method according to claim 13, wherein the grouping step includes a frequency distribution investigating step of dividing the photometric value into a plurality of photometric value ranges, and the method further comprises: The size of one section for dividing the photometric value range or the number of sections of the photometric value range is arbitrary, and the divided photometric areas are grouped by the number of sections. 15. The control method according to claim 13, wherein the evaluation value calculating step includes averaging in each group based on an evaluation value given to each of the blocks according to a distance from a center of the photometry area. A position evaluation step of calculating an evaluation value, a frequency evaluation calculation step of calculating an evaluation value from the frequency of each group, and an average photometric value evaluation step of calculating an evaluation value from an average of photometric values of each group. An exposure control method comprising: 16. The control method according to claim 13 or 14, wherein the grouping step includes selecting one block arbitrarily selected from the groups collected by the frequency distribution checking step in the group. And a search reference block, a first information storage step of storing the information of each block in respective storage means, and reading out one piece of information of the search reference candidate block stored in the first information storage step, A second information storing step of setting a plurality of blocks adjacent to the read search reference candidate block as a search reference neighboring block, and storing information of the search reference neighboring block in a storage unit; Determining whether or not the block can be regarded as the same block, and determining whether to continue the comparison determining process. After the completion of the comparing and judging step, the number of blocks that can be regarded as the same block obtained by the comparing and judging processing is compared with the total number of blocks that have been put into a group in advance to determine group separation. An exposure control method, comprising repeating a series of these processes until the comparison result of the group separation determination step matches, thereby performing group classification. 17. The control method according to claim 16, wherein the comparing and judging step includes reading information of one of the search reference neighboring blocks from the plurality of search reference neighboring blocks stored in the second information storing step. A reading step; a comparing and judging step of judging whether or not the information of one search reference neighboring block extracted in the neighboring reading step is included in the information of the search reference block already stored; A group discriminating step of judging whether or not the search reference neighboring block used in step 1 belongs to the selected group; and, if the search reference neighboring block belongs to the selected group as a result of the group discriminating step, A third information storing step of storing information as the search criterion candidate block and the search criterion block; A first judging step of judging a process destination to be shifted according to the number of the search reference neighboring blocks after the storing process; and a process destination to be shifted according to the number of the search reference candidate blocks after the first judging process. And a second determining step of determining the exposure control method. 18. The control method according to claim 13, wherein in the exposure value calculating step, the image pickup surface is corrected to a photometric value in an area divided into a predetermined size based on a result of the scene determining step. And an exposure value is calculated using the corrected photometric value. 19. The control method according to claim 16, wherein, in the second information storing step, four blocks located around a search reference candidate block selected in the first information storing step in a search reference neighboring block. An exposure control method characterized by using eight to eight blocks.