JP4523629B2 - Imaging device - Google Patents

Description

  The present invention relates to an image pickup apparatus having an automatic exposure control function and maintaining an optimum exposure level of a feature region. A photometric block effective for exposure control is obtained using a result of detecting a feature region from a picked-up image. The present invention relates to an imaging apparatus that selects and performs exposure control.

  Camera exposure control is performed by dividing the imaging screen into a plurality of photometric blocks, and selecting and analyzing the luminance value for each photometric block according to the subject's condition and the photographer's intention. . For example, in exposure control for center-weighted metering, the exposure control amount is determined using the luminance integrated value at the center of the imaging screen.

  Further, in a conventional imaging apparatus, there is an apparatus that optimizes an exposure amount for exposure control by using, for example, an area where a face is detected as a photometric area (for example, Patent Document 1).

Japanese Patent Laying-Open No. 2003-107555 (paragraphs 0062 to 0089, FIG. 8) Patent No. 3,478,201 The patent document 2 will be mentioned later.

  In general exposure control, the average brightness value including the subject imaged in the photometry block and the background is used as the exposure amount, so it is easily affected by the exposure status of the background light, and the subject appears bright when the background is dark. When the background is bright, the subject's exposure level fluctuates due to the influence of background light, such as when the subject appears dark, and there is a problem that the accuracy of advanced image processing such as recognition processing at the later stage is difficult to stabilize.

  Further, in conventional weighted metering or the like, since exposure control is performed with the value of a luminance value at one point in the subject, there is a problem that the exposure value fluctuates due to variations in the luminance value at one point within the subject.

  Further, in the exposure control method using face detection as in the conventional example described in Patent Document 1 described above, when reflected light such as glasses enters the face detection area, the influence of the reflected light from the glasses is affected. As a result, the exposure value fluctuated.

  The present invention has been made to solve the above-described problems, and can stably detect an exposure value of a subject and can stabilize a processing result of advanced image processing such as authentication. An object is to provide an imaging apparatus having control.

The imaging apparatus of the present invention
Imaging means;
Integration means for obtaining an integrated value for each of a plurality of predetermined photometric blocks each forming part of an imaging screen of the imaging signal of the imaging means;
Feature area detection means for detecting a photometric block constituting the feature area in the imaging screen from the imaging signal of the imaging means;
Generating a histogram of the luminance distribution in the photometric block from the imaging signal of the imaging means, and detecting the photometric block in which the histogram is biased to high luminance or low luminance as a specific block ;
The exposure evaluation using the integrated value of the photometric block included in the characteristic area detected by the characteristic area detection unit, other than the photometric block included in the specific area detected by the specific area detection unit Exposure control means for calculating a value and controlling the imaging means using the exposure evaluation value.

  According to the present invention, since the photometric block obtained from the detection result of the feature region having features such as a human face, finger fingerprint, and characters on the paper is used for exposure control, the exposure evaluation value of the feature region can be stably obtained. It is done.

  Even when the brightness of the background changes, the exposure evaluation value of the feature region can be obtained stably.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing an outline of an imaging apparatus according to Embodiment 1 of the present invention. As shown in FIG. 1, the imaging apparatus according to the first embodiment includes an imaging unit 1, an integration unit 3, a feature region detection unit 4, a singular region detection unit 5, and an exposure control unit 2.

The imaging means 1 includes a solid-state imaging device having an array of photoelectric conversion elements, and outputs a digital signal corresponding to the image signal of the subject.
The integrating unit 3 integrates the digital signal output from the imaging unit 1 for each of a plurality of predetermined photometric blocks (photometric windows) each forming a part of the imaging screen.
The feature area detection unit 4 detects a feature area in the imaging screen, for example, a face portion when the subject is a person, from the digital signal output from the imaging unit 1.
The singular area detecting means 5 detects the luminance distribution for each of the photometric blocks (photometric blocks included in the characteristic area) constituting the characteristic area detected by the characteristic area detecting means 4 from the digital output of the imaging means 1 and detects it. On the basis of the result, a photometric block having a specific luminance distribution, for example, a portion having extremely high luminance is detected locally. Examples of such a locally high brightness portion include a nose portion that is bright (shining) or a spectacle portion that reflects light. For example, a locally low-luminance portion such as a dirty portion or a sunglasses portion may be detected.

  The exposure control means 2 selects a photometric block to be used for exposure control according to the detection result of the characteristic area detected by the characteristic area detection means 4 and the detection result of the specific area detection means 5, and integrates the selected photometric blocks Using the integration result of the means 3, exposure control is performed.

  FIG. 2 is a diagram showing a more specific configuration of the imaging apparatus of FIG. In FIG. 1, the imaging apparatus according to the first embodiment includes an imaging unit 1, a digital signal processing unit 10, a timing generation unit 11, a feature region detection unit 4, a singular region detection unit 5, and an exposure control unit 2. A system control unit 15, a display unit 12, a recording unit 13, a data memory 16, and a program memory 17. The data memory 16 is composed of, for example, SDRAM, and the program memory 17 is composed of, for example, flash memory.

  The imaging unit 1 includes a lens 6, a solid-state imaging device 7, an analog signal processing unit 8, and an AD conversion unit 9.

  The digital signal processing means 10 performs signal processing on the digital output of the AD conversion means 9 and includes a portion that functions as the integration means 3 described above. The digital signal processing unit 10 also has a function of displaying a captured image on the display unit 12 and a function of writing data such as a captured image in the storage unit 13 such as an external external storage device.

  The system control means 15 controls each part of the image pickup apparatus shown in FIG. 2, and also performs initial control at the time of power activation of the entire image pickup apparatus, end control at power-down, and control of the entire system other than exposure control. .

  The feature region detection unit 4, the specific region detection unit 5, the exposure control unit 2, and the system control unit 15 are realized by software, for example, a CPU 18 that operates according to a program stored in the program memory 17.

The lens 6 of the imaging unit 1 forms an image of the reflected light from the subject on the solid-state imaging device 7.
The solid-state image sensor 7 photoelectrically converts the intensity (light quantity) of light incident on each pixel from the subject and takes it out as an imaging signal AA.
The analog signal processing unit 8 performs correlated double sampling processing (CDS) on the imaging signal AA output from the solid-state imaging device 7 and outputs an amplified (AGC) analog signal AB.
The AD conversion means 9 digitally converts the output signal AB of the analog signal processing means 8.

  The timing generation unit 11 generates timing signals BB, BC, BD, BE for performing synchronization processing of the solid-state imaging device 7, the analog signal processing unit 8, the AD conversion unit 9, and the digital signal processing unit 10.

The timing generation unit 11 controls the shutter speed (reciprocal of the accumulation time) by changing the accumulation time of the solid-state imaging device 7 based on the control signal CA of the exposure control unit 2.
The gain of the analog signal processing means 11 is controlled based on the control signal CB from the exposure control means 2.
In this way, exposure control can be realized by controlling the accumulation time or shutter speed of the solid-state imaging device 7 and the gain of the analog signal processing means 8.

  The integration unit 3 is configured in the digital signal processing unit 10 and integrates the output of the AD conversion unit 9 for each predetermined photometric block based on the output of the AD conversion unit 9. This integration result is accumulated in the data memory 16 via the CPU bus 19 and further input to the exposure control means 2 for use in exposure control.

  Here, the predetermined photometry block indicates a unit in which imaging signals serving as a basis for detecting an exposure evaluation value for performing exposure control are integrated.

The photometric block is configured by dividing the entire screen as shown in FIG. 3, for example.
In the example shown in FIG. 3, the entire screen is divided into 16 photometric blocks (photometric windows) in the horizontal direction and 12 in the vertical direction.
Each photometric block has a number corresponding to the horizontal position from the upper left corner of the screen (horizontal block number (also called “horizontal block coordinates”), and a number corresponding to the vertical position (vertical block number (“vertical block coordinates”). And B (i, j) (where i is a horizontal coordinate, i = 0-15, j is a vertical coordinate, j = 0-11).

When the number of pixels of the solid-state image sensor 7 is 640 horizontal pixels and vertical lines, each photometric block has 40 pixels in the horizontal direction and 40 lines in the vertical direction.
The position of each photometric block is represented by two-dimensional coordinates (i, j) consisting of a combination of the horizontal block coordinates and the vertical block coordinates.
The coordinates of each photometric block may be represented by the coordinates of the pixel at the upper left corner.

As shown in FIG. 4, the solid-state imaging device 7 of the imaging unit 1 includes an RG line 71 that repeats RGRG, R, and G pixels, and a GB line 72 that repeats GBGB, G, and B pixels. A color filter having a structure generally called a Bayer array is provided.
Here, the G color notation followed by R and G pixels is Gr, and the G color notation followed by G and R pixels is Gb. Any color filter arrangement other than those described above may be used as long as it is a color filter that can obtain R, B, and G color signals.
In the above-described Bayer array color imaging and the 40 × 40 pixels of each photometry block (basic photometry block), R color, Gr color (G color in the same line as R color), B color, Gb Each pixel of color (G color in the same line as B color) occupies 1/4, and therefore the number of pixels of each color component included in each photometric block is 20 × 20 pixels.

  As the solid-state image sensor 7, a three-plate type in which R color, G color, and B color filters are arranged in front of one solid-state image sensor 7 may be used. The present invention can also be applied to a monochrome solid-state imaging device 7 in which no color filter is arranged.

  Further, as the solid-state imaging device 7, an imaging device in which pixels on a matrix are arranged, such as a CCD or a CMOS sensor, can be used.

  The feature area detection means 4 detects, for example, a face area FA in units of photometric blocks based on the digital signal of the image output from the digital signal processing means 10 and stored in the data memory 16, and the feature area Outputs information (identifying information) representing a photometric block constituting the FA, for example, block coordinates (i, j) of the photometric block constituted by a combination of a horizontal block number and a vertical block number.

In the case of the example shown in FIG. 3, the characteristic area detecting means 4 uses the photometric blocks B (5, 4), B (6, 4), B (7, 4), B (8, 4) that constitute the characteristic area FA. , B (9, 4),
B (5,5), B (6,5), B (7,5), B (8,5), B (9,5),
B (5, 6), B (6, 6), B (7, 6), B (8, 6), B (9, 6),
B (5, 7), B (6, 7), B (7, 7), B (8, 7), B (9, 7)
Is output from the feature region detection means 4.

  Instead of outputting block coordinates indicating the position of the photometric block as described above as information representing the photometric block, coordinates indicating the position in the imaging screen of the pixel located at the upper left corner of the photometric block (horizontal position) , A combination of vertical position) may be output as coordinates of the photometric block.

  As a feature region detection method, for example, a face may be detected by detecting a color portion (for example, a skin color portion) specified in advance, and other known methods, for example, brightness of the feature region A method for detecting a change in distribution, a method for detecting a geometric feature of a subject (a face ellipse), and a relative position of an image component within a feature region (for example, eyes, nose, mouth which are components of a face) Or the like can be used.

  The feature area detecting means 4 may be configured in the digital signal processing means 10 instead of being realized by software, and realized by hardware logic such as ASIC and FPGA connected to the CPU bus 19. It's also good.

  The singular area detection means 5 generates a histogram of the luminance distribution for each of the photometry blocks detected by the feature area detection means 4, and based on the generated histogram, the photometry block that should not be used for exposure control, that is, the photometry that should be excluded Detect blocks.

In the case of the example shown in FIG. 3, the characteristic area detecting means 4 uses the photometric blocks B (5, 4), B (6, 4), B (7, 4), B (8, 4) that constitute the characteristic area FA. , B (9, 4),
B (5,5), B (6,5), B (7,5), B (8,5), B (9,5),
B (5, 6), B (6, 6), B (7, 6), B (8, 6), B (9, 6),
B (5, 7), B (6, 7), B (7, 7), B (8, 7), B (9, 7)
In response to this data, the singular area detecting means 5 generates a histogram of photometric block units, and based on the histogram, photometric blocks that should not be used for exposure control, that is, photometry to be excluded Detect blocks.

  For example, a part where the luminance is extremely high or a part where the luminance is extremely low is detected as a specific region to be excluded. In the example shown in FIG. 3, the photometric blocks B (7, 5) and B (7, 6) are portions of the nose NS that are devoted and are detected as photometric blocks to be excluded, and the block B (7, 5 ) And B (7, 6) are output.

In the configuration shown in FIG. 2, the exposure control means 2 is supplied via the CPU bus 19 with the integrated value for each predetermined photometric block obtained by the integrating means 3 and stored in the data memory 16.
The exposure control means 2 is also supplied via the CPU bus 19 with information representing the photometric block constituting the characteristic area FA detected by the characteristic area detecting means 4, for example, the coordinates (i, j) of the photometric block. Information indicating a photometric block having a specific luminance distribution detected by the specific area detecting means 5, for example, block coordinates (i, j) of the photometric block is supplied via the CPU bus 19.

  The exposure control means 2 receives the information representing the photometric block constituting the characteristic area from the characteristic detecting means 4 and the information representing the photometric block to be excluded from the singular area detecting means 5, and based on these, The photometry block used for exposure control is selected, the shutter speed and the gain are obtained based on the integrated value for each selected photometry block, and the exposure of the imaging unit 1 (the shutter speed of the solid-state imaging device 7 and the analog signal processing unit 8). Gain).

  In the exposure control, generally, the gain is set to be high by the control signal CB under dark illumination light. Under bright illumination light, the shutter speed is set to be high by the control signal CA.

Hereinafter, the operation of the imaging apparatus will be described in more detail with a focus on exposure control.
The reflected light from the subject forms an image on the solid-state image sensor 7 by the lens 6.

  Reflected light from the subject imaged on the solid-state image sensor 7 is accumulated for a predetermined exposure time determined by exposure control. The image pickup signal AA can be output by transferring the accumulated charges using a vertical transfer clock and a horizontal transfer clock. Here, in the case of a CCD, in addition to a horizontal transfer clock and a vertical transfer clock, a clock for reading out charges from a photodiode constituting the solid-state imaging device 7 to the vertical CCD and an exposure time set by a shutter speed are set. A clock such as a SUB clock for discharging the accumulated charges exceeding the substrate is supplied from the timing generation unit 11 to the solid-state imaging device 7.

  Subsequently, the image signal AA from the solid-state image sensor 7 is subjected to CDS processing by the analog signal processing means 8. In the CDS process, the black level and the signal level of the imaging signal AA are detected by the sample hold control signal BD of the timing generation unit 11 to generate the sample signal AB. The sample signal AB is digitally converted by the AD conversion means and output as a digital signal AC.

The integrating means 3 of the digital signal processing means 10 obtains an R color integrated value, a G color integrated value, and a B color integrated value for each photometric block from the digital signal AC, and from these, an exposure evaluation value YA for each photometric block is obtained. It calculates | requires by the following formula | equation (1).
YA (i, j) = kr * Ar (i, j) + kg * Ag (i, j) + kb * Ab (i, j) (1)

Ar (i, j) is the average value of the R color of the photometric block (i, j),
Ag (i, j) is the average value of the G color of the photometric block (i, j),
Ab (i, j) indicates the average value of the B color of the photometric block (i, j).
As the average value Ag (i, j) of G color, the average value of Gr color pixels may be used, the average value of Gb color pixels may be used, Gr color pixels and Gb color pixels. An average value of pixels may be used.
kr, kg, and kb are coefficients for obtaining the average luminance from the three primary colors of RGB, and are, for example, values of kr = 0.299, kg = 0.487, and kb = 0.114. Note that the present invention is not limited to the above-described coefficients, and is defined by a color space standard adopted by a system that performs image processing.

  From the digital signal processing means 10, an image signal is sent to the data memory 16 via the CPU bus 19 in the YCbCr or RGB image format, and these data are stored in the data memory 16.

The feature area detection means 4 detects the feature area FA from the image signal held in the data memory 16 based on the above-described feature area detection method, and outputs information representing the photometric blocks constituting the feature area FA.
In the case of the example shown in FIG. 3, the feature area detection means 4 uses the photometric blocks B (5, 4), B (6, 4), B (7, 4), B (8, 4), B (9, 4). ),
B (5,5), B (6,5), B (7,5), B (8,5), B (9,5),
B (5, 6), B (6, 6), B (7, 6), B (8, 6), B (9, 6),
B (5, 7), B (6, 7), B (7, 7), B (8, 7), B (9, 7)
Outputs information indicating.

Information representing a photometric block constituting the detected feature area FA, for example, a block number or coordinates representing the block, is output to the exposure control means 2 via the CPU bus 19. This block number may be temporarily stored in the data memory 16 or may be stored in a memory space in the exposure control means 2.
For example, when the feature area FA is limited to a rectangle, among the photometric blocks constituting the feature area FA, the coordinates of those located in the upper left corner of the feature area FA (5, 4) and the coordinates located in the lower right corner (9, 7) may be output as information indicating the position of the floor start point and end point of the feature area FA.

  As described above, the coordinate value of each photometric block may be expressed by the coordinates of the pixel at the upper left corner. In this case, the block B (5, 4) of the block B (5, 4) located at the upper left corner of the rectangular feature area FA is used. The coordinates (360, 280) of the pixel in the upper left corner and the coordinates (360, 280) of the pixel in the upper left corner of the block B (9, 7) located at the lower right corner of the feature area are the start point and end of the feature area FA, respectively. Output as information indicating the position of a point.

  The singular region detection unit 5 generates a histogram representing the luminance distribution of each of the photometric blocks detected by the feature region detection unit 4, and determines whether the region is a singular region based on the histogram. The singular region mentioned here means a part having a locally high special brightness such as a nose part or a spectacles part, or a part having a locally low special brightness such as dirt or sunglasses.

In the case where the nose NS portion is covered, in FIG. 3, the histogram representing the luminance distribution of the photometric blocks B (7, 5) and B (7, 6) corresponding to the nose NS portion is as shown in FIG. The frequency (frequency) of the high luminance level (class) becomes high. The other photometric blocks of the feature area FA correspond to the skin color part (not measured), and the histogram has a high frequency of the intermediate luminance level as shown in FIG. Therefore, the difference in histogram distribution is determined statistically and empirically, the photometric blocks B (7, 5) and B (7, 6) are excluded, and other photometric blocks,
B (5, 4), B (6, 4), B (7, 4), B (8, 4), B (9, 4),
B (5,5), B (6,5), B (8,5), B (9,5),
B (5, 6), B (6, 6), B (8, 6), B (9, 6),
B (5, 7), B (6, 7), B (7, 7), B (8, 7), B (9, 7)
By performing exposure control using, the subject is in a stable exposure state.
Is used for exposure control.

  In an example of an algorithm for removing a photometric block having a high local brightness, for example, a photometric block B (7, 5) or B (7, 6) having a light, a relatively small number based on a histogram for each photometric block. In the photometry block (of the first predetermined number or less), the number of classes whose frequency exceeds a predetermined value in the high luminance level range is equal to or greater than the second predetermined number, and the frequency is predetermined in the intermediate luminance level range. When it is determined that the number of classes exceeding the value is equal to or smaller than the third predetermined number, it is determined that the photometric block is a unique photometric block with extremely high brightness locally.

  On the other hand, it is desirable to remove from the exposure control not only a high-brightness signal such as “Tekari” but also a low-brightness photometric block such as dirt or sunglasses. In that case, for example, in a comparatively small number (fourth predetermined number or less) of photometric blocks, the number of classes whose frequency exceeds a predetermined value in the range of the low luminance level is the fifth predetermined number or more, In the range of the intermediate luminance level, if the number of classes whose frequency exceeds a predetermined number is equal to or smaller than the sixth predetermined number, it is determined that the light metering block is a peculiar photometric block with extremely low luminance locally.

As shown in FIG. 7, a similar problem may occur in the case of a person wearing glasses. That is, even when the illumination light is reflected on the glasses EG, the luminance of the glasses EG portion is brightened, so that a high luminance distribution is increased as in the histogram shown in FIG. As shown in FIG. 7, photometric blocks B (5, 4) and B (9, 4) excluding the photometric block of the spectacles EG portion.
B (5, 5), B (9, 5)
B (5, 6), B (6, 6), B (7, 6), B (8, 6), B (9, 6)
B (5, 7), B (6, 7), B (7, 7), B (8, 7), B (9, 7)
By performing exposure control using, the subject is in a stable exposure state.

The exposure control unit 2 obtains the integration result of the block coordinates of the selected photometry block based on the integration result of the integration unit 3, and obtains the basic exposure evaluation value ΣEX by the following equation (2).
In the case of the example shown in FIG. 3, the integration result of the block coordinates of the selected photometry block is obtained, and the basic exposure evaluation value ΣEX is obtained by the following equation (2).
ΣEX = {YA (5,4) + YA (6,4) + YA (7,4) + YA (8,4) + YA (9,4) + YA (5,5) + YA (6,5) + YA (8,5 ) + YA (9,5) + YA (5,6) + YA (6,6) + YA (8,6) + YA (9,6) + YA (5,7) + YA (6,7) + YA (7,7) + YA (8,7) + YA (9,7)} / n
... (2)
In the above equation (2), n is the total number of selected photometric blocks (“18” in the example of FIG. 3).

The calculation of the above equation (2) can also be expressed by the following equation (3).
ΣEX
= {Α (0,0) × YA (0,0) + α (1,0) × YA (1,0) +... + Α (15,0) × YA (15,0)
+ Α (0,1) × YA (0,1) + α (1,1) × YA (1,1) +... + Α (15,1) × YA (15,1)
+ ...
+ Α (0,11) × YA (0,11) + α (1,11) × YA (1,11) +... + Α (15,11) × YA (15,11)} / n
... (3)
It becomes.

  In the above equation (3), α (i, j) is a block coordinate selection control coefficient, α (i, j) = 1 for a selected block, and α (i, j) for a non-selected block. = 0. YA (i, j) is the exposure evaluation value of the photometric block B (i, j), n is the total number (for example, 18) of photometric blocks with α (i, j) = 1, and n = Σα (i, j j).

Therefore, in the case of FIG. 3, based on the detection result of the characteristic region and the detection result of the singular region,
α (5,4), α (6,4), α (7,4), α (8,4), α (9,4), α (5,5), α (6,5), α (8,5), α (9,5), α (5,6), α (6,6), α (8,6), α (9,6), α (5,7), α ( 6, 7), α (7, 7), α (8, 7), and α (9, 7) are 1, and the other α (i, j) is 0.
If the total number n of effective photometric blocks is set to a power of 2, division processing can be performed by bit shift, and ΣEX can be processed at high speed regardless of hardware processing or software processing. It can be carried out.

ΣEX indicates an exposure evaluation value of an effective photometric block.
The exposure control is performed so as to converge to a preset exposure target value. As a convergence method at this time, a feedback type exposure control method for feeding back and converging the difference between the exposure evaluation value and the exposure target value, or until the difference between the exposure evaluation value and the exposure target value becomes zero. For example, a feedforward type exposure control method disclosed in Patent Document 2 described above, which controls the address of the LUT used for control, can be used.

  Since the exposure evaluation value for each photometric block is used, the control method of the exposure control is changed regardless of the total number n of photometric blocks obtained from the detection result of the characteristic region detection unit 4 and the detection result of the specific region detection unit 5. Can be controlled without.

  As described above, it is possible to perform appropriate exposure control by removing the singular region where the luminance distribution is significantly different from that of the other photometric blocks by the singular region detecting means 5.

  For example, as described above, the subject's nose or the like may be bright depending on how the illumination is turned on. When the luminance information of this part is used as the integrated value, the average luminance of the exposure control is shifted in the bright direction, so that the face image after the exposure control may fall into an underexposed state. By excluding regions with extremely high brightness locally, appropriate exposure control is possible.

  The same applies to the case where the spectacles reflect the illumination light. When the image is captured, the eye information is lost. In such a case as well, although underexposure occurs, appropriate exposure control can be performed by excluding the photometric block of the glasses portion and using other photometric blocks as shown in FIG.

  Appropriate exposure control can be performed by excluding a portion having extremely low brightness locally as a unique region.

  By maintaining the optimal exposure state for the subject, the contrast and brightness of the subject can be kept constant, for example, an image signal with stable brightness and contrast can be obtained even in subject recognition processing, Recognition accuracy can be improved. In addition, stable recognition accuracy can be obtained even in various lighting environments.

  In the example shown in FIG. 2, the singular region detection means 5 is configured by software, that is, as a part of the CPU 18 that operates according to the program in the program memory 17, but instead is configured in the digital signal processing means 10. It's also good. When configured as a part of the digital signal processing means 10, it is not necessary to send all data using the CPU bus 19, and a histogram can be generated at high speed without considering the bus transfer efficiency.

  On the other hand, when the singular area detection means 5 is configured as a part of the CPU 18, it can be realized by combining advanced image processing in addition to generating a histogram, and the exposure accuracy detection value of the characteristic area can be increased.

  For example, when glasses or the like are present in the feature region, the outline of the glasses can be detected. As a detection method, edge detection is performed to obtain edge information. For example, if the edge information is continuous, exists in the feature area (face) or around the face, and has a shape such as an ellipse or a rectangle, the area surrounded by the continuous edge information is in glasses. Therefore, it is possible to improve the detection accuracy of luminance information such as not using a photometric block including glasses.

FIG. 9 shows a flowchart of exposure control according to the present invention.
When the exposure control is started, a feature region is detected from the captured image in the feature region detection step ST1. This process is performed by the feature region detection means 4.
Next, in the feature region determination step ST2, it is determined whether or not the feature region has been detected. This process is performed by the system control means 15.
When a feature region is detected in the feature region detection step ST2 (YES in ST2), the luminance distribution for each photometric block in the feature region is determined, and a photometric block having a unique distribution should be excluded (singular This is detected as a region photometric block (ST3). This process is performed by the unique region detecting means 5.
Next, a photometric block is selected based on the information representing the photometric block constituting the feature area detected in step ST3 and the information representing the photometric block constituting the specific area detected in step ST5 (ST4). . In this case, among the photometric blocks determined to constitute the feature region in step ST3, those other than the photometric block determined to constitute the singular region in step ST5 are selected.
If no feature area is detected in step ST2 (NO in ST2), a photometric block predetermined in step ST5 is selected.
In exposure control step ST6, exposure control (determination of shutter speed of solid-state image sensor 7 and gain of analog signal processing means 8 based on luminance information of the photometric block selected in step ST4 or the photometric block selected in step ST5 is performed. Decision). The processing of steps ST4, ST5 and ST6 is performed by the exposure control means 2.

As described above, the imaging apparatus according to the first embodiment has means for obtaining a photometric block based on the feature area FA corresponding to the feature portion of the subject, so that the subject can maintain the optimum exposure condition. it can.
In addition, since the singular area detecting means 5 excludes the photometric block whose luminance distribution is significantly different from that of other photometric blocks, it is possible to avoid the influence of the nose, the reflecting glasses, dirt and sunglasses.

The schematic configuration of FIG. 1 can be configured by another system configuration instead of the configuration shown in FIG.
For example, the image pickup means 1 provided with a stop on the incident side of the lens in FIG. 2 can also be used. In that case, it is also possible to control the degree of aperture opening as part of the exposure control.
The presence or absence of an aperture is determined based on the specification of the system configuration of the imaging apparatus. When there is control by the aperture, in addition to controlling the accumulation time of the solid-state imaging device 7 and the gain in the analog signal processing means 8 so that the luminance evaluation value of the image photometry block becomes a predetermined average luminance level, control of the aperture The only difference is that it is added (by the CPU 18).

  In the above example, the feature area detection unit 4 selects only the photometry blocks constituting the feature area. However, when the area of the area detected as the feature area is small, the surrounding photometry blocks are also selected. And it is good also as using for exposure control.

It is a block diagram which shows schematic structure of the imaging device of Embodiment 1 of this invention. It is a block diagram which shows the imaging device of FIG. 1 in detail. It is a figure which shows an example of the photometry block selected by a characteristic area | region detection means and a specific area | region detection means. It is a figure which shows arrangement | positioning of the pixel of each color component in a Bayer type | mold arrangement | sequence. It is a figure which shows an example of the histogram which shows the luminance distribution of the photometry block B (7, 5) and B (7, 6) corresponding to the nose which FIG. It is a figure which shows an example of the histogram which shows luminance distribution of blocks other than peculiar block B (7,5) and B (7,6) among the blocks in the characteristic area | region of FIG. It is a figure which shows the other example of the photometry block selected by the characteristic area | region detection means and the specific area | region detection means. Photometric blocks B (6,4), B (7,4), B (8,4), B (6,5), B (7,5), B corresponding to the portion of the shining glasses in FIG. It is a figure which shows an example of the histogram which shows the luminance distribution of (8,5). 3 is a flowchart illustrating an operation of the imaging device of FIGS. 1 and 2.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 Imaging device, 2 Exposure control means, 3 Accumulation means, 4 Feature area detection means, 5 Specific area detection means

Claims (1)

Imaging means;
Integration means for obtaining an integrated value for each of a plurality of predetermined photometric blocks each forming part of an imaging screen of the imaging signal of the imaging means;
Feature area detection means for detecting a photometric block constituting the feature area in the imaging screen from the imaging signal of the imaging means;
Generating a histogram of the luminance distribution in the photometric block from the imaging signal of the imaging means, and detecting the photometric block in which the histogram is biased to high luminance or low luminance as a specific block ;
The exposure evaluation using the integrated value of the photometric block included in the characteristic area detected by the characteristic area detection unit, other than the photometric block included in the specific area detected by the specific area detection unit calculating a value, imaging equipment that is characterized by having a exposure control means for controlling the imaging means with the exposure evaluation value.

Images