JP2003319242A - Camera - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve operability of a camera. <P>SOLUTION: Based upon positions of a zoom lens 12 and a focus lens 14, a distance to a main object existing within a field is measured and based upon the position of the focus lens 14, an interval between the focus lens 14 and an image sensor 20 is measured. A face area to be occupied by a face of a figure on a photo-detection plane of the image sensor 14 is specified based upon the measured distance and interval. On the other hand, a flesh-colored area in the field is detected based upon a photographed image signal. A photographing mode is determined based upon a size of the flesh-colored area included in the face area and a size of the flesh-colored area out of the face area, and the field is photographed in the determined photographing mode. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a camera,
In particular, the present invention relates to a camera that is applied to a digital camera and shoots the scene in a shooting mode according to the scene.

[0002]

2. Description of the Related Art As a conventional camera of this type, there is one having a portrait mode suitable for photographing a human face. In this shooting mode, the program diagram is corrected so that the aperture is opened and the exposure time is shortened, and the white balance adjustment gain is corrected so as to suppress the change in the skin color of the person. As a result, it is possible to photograph a face with a healthy expression in a state where the background is blurred.

[0003]

However, in the prior art, it was necessary to manually set the photographing mode, and there was a problem in operability.

Therefore, a main object of the present invention is to provide a camera which can improve operability.

[0005]

According to the present invention, in a camera for photographing in a photographing mode according to a scene, a first measuring means for measuring a distance to a main subject existing in the scene, a face of a person. Based on the relationship between the face area and the skin color area, the specifying means for specifying the face area to be occupied by the object from the scene based on the distance measured by the first measuring means, the detecting means for detecting the skin color area of the scene, The camera is characterized in that the camera is provided with a deciding means for deciding a photographing mode.

[0006]

[Function] The first distance to the main subject existing in the object scene is
When measured by the measuring means, the specifying means specifies the face area that the face of the person should occupy from the scene based on the measured distance. On the other hand, the skin color region of the object scene is detected by the detection means. The shooting mode is determined by the determining means based on the relationship between the face area and the skin color area, and the scene is shot in the determined shooting mode.

Since the size of a person's face can be estimated in advance,
When the distance to the main subject is measured, the face area can be identified from the object scene. The shooting mode is determined based on the relationship between the face area thus identified and the skin color area of the scene.

When the zoom lens is provided, the distance to the main subject may be measured based on the position of the zoom lens.

Preferably, the optical image of the object field that has passed through the focus lens is incident on the light receiving surface of the image sensor, and the distance between the focus lens and the light receiving surface is measured by the second measuring means. The specifying unit specifies the face area of the object scene image projected on the light receiving surface based on the distance to the main subject and the distance between the focus lens and the light receiving surface.

Preferably, the determining means determines the size of the first skin color area included in the face area and the size of the second skin color area outside the face area, and based on the sizes of the first skin color area and the second skin color area. To set the shooting mode to portrait mode.

By outputting the message corresponding to the photographing mode determined by the determining means, the operator can easily determine whether the photographing mode is the desired mode.

[0012]

According to the present invention, the photographing mode is determined based on the relationship between the face area specified based on the distance to the main subject and the skin color area of the field.
The operability of the camera can be improved.

The above-mentioned objects, other objects, features and advantages of the present invention will become more apparent from the following detailed description of the embodiments with reference to the drawings.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Referring to FIG. 1, a digital camera 10 of this embodiment includes a zoom lens 12 and a focus lens 1.
4, including a diaphragm mechanism 16 and a shutter mechanism 18. The optical image of the object field is transferred to the image sensor 2 through these members.
It is incident on the light receiving surface of 0. The number of effective pixels of the image sensor 12 is about 4 million pixels, and there are 2300 pixels and 1740 lines in the horizontal and vertical directions of the light receiving surface. Light-receiving surface is Cy (cyan), Ye (yellow), Mg
(Magenta) and G (green) are covered by a color filter (not shown) arranged in a mosaic pattern,
Each pixel of the raw image signal generated by photoelectric conversion is C
It has color information of y, Ye, Mg or G.

When the power is turned on, the CPU 32 displays a real-time moving image (through image) of the subject on the monitor 44 by ⅙ to the TG (Timing Generator) 30.
The vertical thinning-out reading is instructed to set the horizontal zoom magnification and the vertical zoom magnification of the zoom circuit 34 to "1/4" and "1". The TG 30 performs thinning-out reading on the image sensor 20, whereby Cy, Ye, ...
.. and 2300 pixels.times.290 lines of raw image signal including alternating lines of Mg, G ...
It is output from the image sensor 20 at a rate of 1 frame every / 30 seconds.

The raw image signal of each frame output from the image sensor 20 is subjected to noise removal and level adjustment by the CDS / AGC circuit 22. The A / D converter 24 converts the raw image signal output from the CDS / AGC circuit 22 into a digital signal. When the power is turned on, the switch SW1 is connected to the terminal S1 and the raw image signal output from the A / D converter 24 is input to the signal processing circuit 26 via the switch SW1.

The signal processing circuit 26 is constructed as shown in FIG. Each pixel forming the raw image signal is Cy, Ye,
Since it has only one color information of Mg and G, first, the color separation circuit 26a complements the color information that each pixel lacks. The RGB conversion circuit 26b performs RGB conversion on the complementary color image signal output from the color separation circuit 26b, and the white balance adjustment circuit 26c outputs the RGB conversion circuit 26b.
White balance adjustment is performed on the primary color image signal output from the.
The white color-adjusted primary color image signal is converted into a YUV signal by the YUV conversion circuit 26d. The generated YUV signal has a ratio of Y: U: V = 4: 2: 2.

The primary color image signal output from the white balance adjusting circuit 26c is also input to the integrating circuit 26e, and the Y signal forming the YUV signal output from the YUV converting circuit 26d is also applied to the integrating circuit 26f. Referring to FIG. 3, the field (screen) is divided into 16 in each of the vertical direction and the horizontal direction, and 256 blocks are formed on the screen. Each block has a vertical position number i
(= 0 to 15) and horizontal position number j (= 0 to 15) are assigned.

The integrating circuit 26d integrates each of the R signal, G signal and B signal forming the primary color image signal for each block, and the integrating circuit 26f integrates the Y signal for each block. As a result, 256 integrated values r of the R signal are obtained.
(I, j), 256 integrated values g (i,
j) and 256 integrated values b (i,
j) is output from the integrator circuit 26e every frame period, and 256 integrated values y (i, j) regarding the Y signal are output from the integrator circuit 26f every frame period.

Returning to FIG. 1, the YUV signal output from the signal processing circuit 26 is applied to the zoom circuit 34. The YUV signal has a resolution of 2300 pixels × 290 lines, and the horizontal zoom magnification and the vertical zoom magnification of the zoom circuit 34 are set to “1/4” and “1”. A YUV signal of x290 lines is output.

The YUV signal output from the zoom circuit 34 is written in the display image area 38a (see FIG. 4) of the SDRAM 38 by the memory control circuit 36, and thereafter,
It is read from the display image area 38a by the memory control circuit 36. The resolution of the read YUV signal is converted into 575 pixels × 580 lines by the pseudo framing circuit 40, and the converted YUV signal is encoded into a composite video signal of 640 pixels × 480 lines by the video encoder 42. The encoded composite video signal is supplied to the monitor 44, and as a result, a through image of the subject is displayed on the screen.

The 256 integrated values y (i, j) output from the integrating circuit 26f shown in FIG. 2 are fetched by the CPU 32 and set in the register rgst1. Since the integrated value y (i, j) is generated every frame period, the set value of the register rgst1 is updated every frame period.

When the zoom button 56 is operated, a corresponding status signal is sent from the system controller 52 to the CPU 32.
Given to. The CPU 32 controls the driver 28a,
As a result, the zoom lens 12 moves in the optical axis direction.
The zoom magnification of the through image displayed on the monitor 44 changes in response to the operation of the zoom button 56.

When the shutter button 54 is pressed halfway, a corresponding status signal is sent from the system controller 52 to the CPU.
32. The CPU 32 uses the shutter button 54
The 256 integrated values y (i, j) output from the integrating circuit 26f after half-pressing are set in the register rgst2.
As a result, integrated values y (i, j) of two consecutive frames are obtained in the registers rgst1 and rgst2. C
The PU 32 calculates the possibility that the object scene is a sports scene based on the integral value y (i, j) thus obtained.

When the determination of the possibility of the sports scene is completed, the CPU 32 adjusts the focus. The focus lens 14 is moved in the optical axis direction by the driver 28b and set at the in-focus position. When the focus adjustment is completed, the CPU 32 causes the integrated values r (i, j), g (i, j) and b (i, j) output from the integration circuit 26e and the integrated value y (output from the integration circuit 26e. i, j)
Is set in the register rgst3. The integrated values r (i, j), g (i, j), b (i, j) and y for one frame
When the capture of (i, j) is completed, the CPU 32 determines the possibility of portrait scene, the possibility of evening scene, and the possibility of night scene.

The possibility of a portrait scene is that the distance from the zoom lens 12 to the main subject, the distance between the focus lens 14 and the image sensor 20, and the integral values r (i, j) and g (set in the register rgst3. i,
j) and b (i, j). Also,
The possibility of a sunset scene is calculated based on the integrated values r (i, j), g (i, j), b (i, j) and y (i, j) set in the register rgst3. Further, the possibility of a night scene is calculated based on the integrated value y (i, j) set in the register rgst3.

When the respective possibilities of the sports scene, the portrait scene, the evening scene and the night scene are calculated in this way, the scene with the highest possibility is determined as the scene scene. The camera setting, that is, the shooting mode is changed according to the confirmed scene, and the monitor 44 displays a message corresponding to the confirmed scene.

When the sports scene is confirmed, the CPU 32
Corrects the program diagram for exposure adjustment so that a moving subject can be captured clearly. As a result, the shooting mode becomes the sports mode. When the portrait scene is confirmed, the CPU 32 corrects the program diagram so that the background is blurred, and corrects the white balance adjustment gain so as to suppress the change in the skin color of the person. As a result, the shooting mode becomes the portrait mode. When the sunset scene is confirmed, the CPU 32 corrects the program diagram so that the distant view is captured clearly, and corrects the white balance adjustment gain so that the change in the color of the sunset can be suppressed. As a result, the shooting mode becomes the evening scene mode. When the night scene is confirmed, the CPU 32 corrects the program diagram so that the illumination becomes prominent. As a result, the shooting mode becomes the night view mode.

Whichever scene is set, the program diagram is corrected. Therefore, after the scene determination is completed, the CPU 32 adjusts the aperture amount and the exposure time based on the integrated value y (i, j) set in the register rgst2 and the corrected program diagram.

When the shutter button 54 is fully pressed after such exposure adjustment is completed, a corresponding status signal is given from the system controller 52 to the CPU 32.
The CPU 32 executes a shooting process. Specifically, CP
The U 32 commands the TG 30 to perform the main exposure, and when the main exposure by the TG 30 is completed, the driver 28 d drives the shutter mechanism 16. The incident light on the image sensor 20 is blocked by driving the shutter mechanism 16. CP
U32 also outputs a raw image signal for one frame obtained by the main exposure from the image sensor 20.
Instruct G30 to read all pixels. By this, 2
A raw image signal of 300 pixels × 1740 lines is read from the image sensor 20 by the interlaced scan method.

This raw image signal is supplied to the CDS / AGC circuit 2
2 and the A / D converter 24 through the memory control circuit 36
To the SDRAM 3 by the memory control circuit 36.
8 raw image areas 38b (see FIG. 4).
Since the raw image signal of 2300 pixels × 1740 lines is an interlaced scan signal, the raw image area 38b
The odd field signal is stored in the first half of the above, and the even field signal is stored in the latter half of the raw image area 38b. That is, odd field areas and even field areas are formed in the raw image area 38b.

After the writing to the raw image area 38b is completed, the memory control circuit 36 alternately reads out the raw image signal from the odd field area and the even field area line by line. As a result, the interlaced scan signal is converted into the progressive scan signal. The switch SW1 is connected to the terminal S2 when the shutter button 54 is fully pressed. Therefore, the raw image signal read by the memory control circuit 36 is given to the signal processing circuit 22 via the switch SW1. The signal processing circuit 22 executes a series of processes of color separation, RGB conversion, white balance adjustment, and YUV conversion, thereby generating a YUV signal (main YUV signal) of 2300 pixels × 1740 lines.

The horizontal zoom ratio and the vertical zoom ratio of the zoom circuit 34 are set to "1/4" and "1/6", respectively, when the shutter button 54 is fully pressed. Therefore, the YUV output from the signal processing circuit 22
The signal resolution is 5 from 2300 pixels x 1740 lines.
It is converted into 75 pixels × 290 lines. Zoom circuit 34
The YUV signal of 575 pixels × 290 lines output from the memory control circuit 36 is written in the display image area 38a shown in FIG. After that, the same processing as that for displaying the through image is performed, whereby the freeze image at the time when the shutter button 54 is operated is displayed on the monitor 4.
4 is displayed.

2300 output from the signal processing circuit 26
The main YUV signal of pixel × 1740 lines is also directly supplied to the memory control circuit 36, and the memory control circuit 36
Is written in the main image area 38c (see FIG. 4) of the SDRAM 38. When writing is completed, CPU3
2 creates a reduced YUV signal of 160 pixels × 120 lines based on the main YUV signal of 2300 pixels × 1740 lines. Specifically, the CPU 32 accesses the SDRAM 38 via the memory control circuit 36 and generates a reduced YUV signal by software processing. The generated reduced YUV signal is written in the reduced image area 38d (see FIG. 4) of the SDRAM 38.

The memory control circuit 36 reads out the main YUV signal and the reduced YUV signal from the SDRAM 38, and supplies each YUV signal to the JPEG codec 46. J
The PEG codec 46 compresses the supplied main YUV signal and reduced YUV signal according to the JPEG format, and generates a compressed main YUV signal and a compressed reduced YUV signal. The compressed main YUV signal and the compressed reduced YUV signal generated by the memory control circuit 36 are SDRA.
It is written in the compressed main image area 38e and the compressed and reduced image area 38f (see FIG. 4) of M38.

When the photographing process is completed in this way, the CPU 3
2 executes a recording process. Specifically, the CPU 32
Accesses the SDRAM 38 via the memory control circuit 36 and outputs the compressed main YUV signal and the compressed reduced YUV signal to the compressed main image area 38e and the compressed reduced image area 3.
Read from each 8f. The CPU 46 further records the read compressed main YUV signal and compressed reduced YUV signal in the recording medium 50 in a file format. The recording medium 50 is detachable, and the recording medium 50 is accessed via the I / F 48.

The CPU 32 is specifically shown in FIGS.
2, FIG. 15 to FIG. 16, FIG. 22 to FIG. 24, FIG. 29 to FIG.
The control program corresponding to the flowchart shown in FIG.
The control program is stored in the ROM 58.

First, the display system is activated in step S1 shown in FIG. Specifically, the TG 30 is instructed to perform thinning-out reading, and the horizontal zoom magnification and the vertical zoom magnification of the zoom circuit 34 are set to "1/4" and "1". As a result, a through image of the subject is displayed on the monitor 44.

In step S3, TG 30 to 30 fps
It is determined whether or not the vertical synchronizing signal has been generated, and if YES, the integrated value fetching process 1 is executed in step S5. As a result, 256 integrated values y (i, j) individually corresponding to the 256 blocks shown in FIG. 3 are stored in the register rgst.
Set to 1. In step S7, the shutter button 54
Is pressed halfway, and in step S9 it is determined whether the zoom button 56 has been operated. When the zoom button 56 is operated, the process proceeds from step S9 to step S11 to control the driver 28a to move the zoom lens 12 in the optical axis direction. When the process of step S11 is completed, the process returns to step S3.

When the shutter button 54 is half-depressed, YES is determined in the step S7, and the integrated value fetching process 2 is performed in a step S13. As a result, 256 integrated values y (i, j) are set in the register rgst2. In step S15, the possibility that the object scene is a sports scene is determined based on the integrated value y (i, j) of two consecutive frames set in the registers rgst1 and 2. Possibility is expressed as a percentage.

When the process of step S15 is completed, focus adjustment is performed in step S17. Specifically, the driver 28b is controlled to move the focus lens 14 in the optical axis direction, and the focus lens 14 is set at the in-focus position detected by this. When the focus adjustment is completed, the process proceeds from step S19 to step S21 after waiting for the generation of the vertical synchronizing signal, and the signal processing circuit 26 causes the integrated value r
(I, j), g (i, j), b (i, j) and y
Take in (i, j). The integrated value r (i,
j), g (i, j), b (i, j) and y (i, j)
Is set in the register rgst3.

In step S23, the distance from the zoom lens 12 to the main subject, the distance between the focus lens and the image sensor 20, and the integral values r (i, j), g (i, j) and g (i, j) set in the register rgst3 and b
Based on (i, j), the possibility that the scene is a portrait scene is determined. In step S25, the integrated values r (i, j), g (i, j), b set in the register 3
Based on (i, j) and y (i, j), the possibility that the object scene is an evening scene is determined. In step S27, the integrated value y (i, i, i, i) set in the register rgst3 is set.
Judge the possibility that the scene is a night scene based on j). Here again, the possibility is expressed as a percentage.

In step S29, steps S15, S
The possibility with the highest percentage is specified from the possibilities calculated in S23, S25, and S27, and the scene corresponding to the specified possibility is determined as the scene. In step S29, the camera setting (shooting mode setting) corresponding to the confirmed scene is further performed. When the scene is determined to be a sports scene, the program diagram is corrected so that a moving subject is clearly captured (sport mode setting). When the scene is determined to be a portrait scene, the program diagram is corrected so that the background is blurred, and the white balance adjustment gain is corrected so that changes in the skin color of the person are suppressed (portrait mode setting). When the scene is confirmed as a sunset scene, the program diagram is corrected so that the distant view is captured clearly, and the white balance adjustment gain is corrected so that the change in the colors of the sunset can be suppressed (setting of the sunset mode). . When the scene is confirmed as a night scene,
Correct the program diagram so that the illumination is prominent (night view mode setting).

In step S31, a character generator (not shown) is driven to display an OSD character corresponding to the confirmed scene on the monitor 44. CPU3
Since there is a possibility of erroneous discrimination of the object scene in the automatic discrimination according to No. 2, in this embodiment, the operator is notified by a visible message which scene has been decided. This improves operability. Although detailed description is omitted, the settings of the sports scene, portrait scene, evening scene, or night scene can be changed by an operator's manual operation.

In step S33, the register rgst2
The optimum aperture amount and the optimum exposure time are specified based on the integral value y (i, j) set in step S29 and the program diagram corrected by the camera setting in step S29, and the optimum aperture amount is determined by the driver 28c. Set to. The shutter mechanism 18, in step S39 described later,
The drive is performed when the optimum exposure time has elapsed from the start of main exposure.

In step S35, the shutter button 54
It is determined whether or not is fully pressed, and in step S37, it is determined whether or not the operation of the shutter button 54 is released. When the shutter button 54 is fully pressed, the shooting process / recording process in step S37 is performed and then step S37 is performed.
Return to 1. The subject image is recorded on the recording medium 50 by the photographing process and the recording process. When the operation of the shutter button 54 is released, the process returns to step S1 without performing the photographing process / recording process.

The integrated value acquisition process 1 in step S5 is as shown in FIG.
And according to the subroutine shown in FIG. First step S
In step 41, the vertical position number i is set to "4", and step S4
At 3, the horizontal position number j is set to "4". Step S4
In 5, the integrated value y (i, j) is read from the register rgst1 and the integrated value y (i, j) is set to the specific integrated value Ysp.
It is set to rgst4 as 1 (i, j). Step S
In 47, the horizontal position number j is incremented, and in step S49, the incremented vertical value number j is set to "1".
2 ". If j <12, step S
Returning to 45, if j = 12, the vertical position number i is incremented in step S51, and the vertical position number i incremented in step S53 is compared with "12". Here, if i <12, the process returns to step S43,
When i = 12, the process proceeds to step S55.

In step S55, the vertical position number i is set to "0", and in the following step S57, the horizontal position number j is set to "0". In step S59, the vertical position number i and the horizontal position number j are 0 <i <15 and 0 <
It is determined whether or not the condition of j <15 is satisfied. If both conditions are satisfied, the process directly proceeds to step S63. On the other hand, if any one of the above conditions is not satisfied, the integrated value y of the register rgst1 in step S61.
Rg with (i, j) as the specific integrated value Ysp1 (i, j)
It is set to st4, and then the process proceeds to step S63. In step S63, the horizontal position number j is incremented, and in the subsequent step S65, the incremented horizontal position number j
Is compared with "16". Then, if j <16, the process returns to step S59, but if j = 16, step S67.
Proceed to. In step S67, the vertical position number i is incremented, and in step S69, the incremented vertical position number i is compared with "16". And i <16
If so, the process returns to step S57, but if i = 16, the process returns to the upper hierarchy routine.

By such processing, a total of 124 specific integrated values Ysp1 (i, j) for 64 blocks forming the central region CTR1 and 60 blocks forming the peripheral region ARD1 shown in FIG. 13 are obtained.

The integrated value fetching process 2 of step S13 shown in FIG. 5 follows the subroutine shown in FIGS.
However, this subroutine includes steps S75 and S
7 and 8 except that the integrated value y (i, j) stored in the register rgst2 is set in the register rgst2 as the specific integrated value Ysp2 (i, j) in 91.
Since it is the same as the subroutine shown in, duplicate description will be omitted. By this processing, the central area C shown in FIG.
The 124 specific integrated values Ysp2 (i, j) forming TR1 and the peripheral region ARD1 are obtained.

The possibility determination process of step S15 shown in FIG. 5 follows the subroutine shown in FIGS.
First, the variables Cctr and Card are set to "0" in step S101, and the vertical position number i and the horizontal position number j are set to "4" in steps S103 and S105. In step S107, the specific integrated value Ys
The difference absolute value ΔYsp (i, j) between p1 (i, j) and Ysp2 (i, j) is calculated, and in the subsequent step S109, the calculated difference absolute value ΔYsp (i, j) is set to the threshold value Yt.
Compare with h1.

[0052]

## EQU1 ## ΔYsp (i, j) = | Ysp1 (i, j)-
If Ysp2 (i, j) | ΔYsp (i, j) ≧ Yth1, step S1
In step S111, the variable Cct is determined as YES.
After incrementing r, the process proceeds to step S113.
On the other hand, if ΔYsp (i, j) <Yth, NO is determined in step S109, and the process directly proceeds to step S113.

In step S113, the horizontal position number j is incremented, and in step S115, the incremented horizontal position number j is compared with "12". And
If j <12, the process returns to step S107, but j = 12
If so, the process proceeds to step S117. In step S117, the vertical position number i is incremented, and step S11
In 9, the incremented vertical position number i is “12”.
Compare with. If i <12, step S10
Returning to 5, if i = 12, the process proceeds to step S121.

The absolute difference value ΔYsp (i, j) calculated in step S107 corresponds to the amount of movement of the subject in each block (i, j) forming the central area CTR1. If the amount of movement is large, YES is determined in step S109 and the variable Cctr is incremented. Therefore, the larger the proportion of the subject having a large amount of movement in the central region CTR1, the larger the value of the variable Cctr.

In step S121, the vertical position number i is set to "0", and in the following step S123, the horizontal position number j is set to "0". In step S125, the vertical position number i and the horizontal position number j are 0 <i <15 and 0.
Whether or not the condition of <j <15 is satisfied is determined, and if both conditions are satisfied, the process directly proceeds to step S131.
If either one is not satisfied, the process proceeds to step S126. In step S126, the absolute difference value ΔYsp (i, j) between the specific integrated values Ysp1 (i, j) and Ysp2 (i, j) is calculated in accordance with the above equation 1, and the subsequent step S126
At 127, the calculated absolute difference value ΔYsp (i, j) is compared with the threshold value Yth1. And ΔYsp (i, j)
If ≧ Yth, the variable Card is incremented in step S129 and then the process proceeds to step S131.
If sp (i, j) <Yth, step S1 as it is.
Proceed to 31.

The horizontal position number j is incremented in step S131, and the incremented horizontal position number j is compared with "16" in step S133. And
If j <16, the process returns to step S125, but j = 16
If so, the process proceeds to step S135. In step S135, the vertical position number i is incremented, and step S13
In 7, the incremented vertical position number i is "16".
Compare with. If i <16, step S12
Returning to 3, if i = 16, the process proceeds to step S139.

The absolute difference value ΔYsp (i, j) calculated in step S126 corresponds to the amount of movement of the subject in each block (i, j) forming the peripheral area ARD1. If the amount of movement is large, YES is determined in step S127 and the variable Card is incremented. Therefore, the larger the ratio of the subject having a large amount of movement in the peripheral area ARD1, the larger the value of the variable Card.

In step S139, the possibility Psprt that the object scene is a sports scene is calculated according to equations (2) to (4). When the possibility Psprt is calculated, the process returns to the upper hierarchy routine.

[0059]

(2) Rcrt = Cctr / 64 * 100

[0060]

## EQU00003 ## Rard = Card / 60 * 100

[0061]

Psprt = Rctr-a * Rard a: The number of blocks forming the constant central region CTR1 is “64”, and the number of blocks forming the peripheral region ARD1 is “6”.
Therefore, by dividing the variable Cctr by “64” and multiplying the divided value by “100”, the ratio R of the subject having a large amount of motion to the central region CTR1 is R.
ctr is calculated. Further, by dividing the variable Card by “60” and multiplying the divided value by “100”, the ratio Rard of the subject having a large amount of motion in the peripheral area ARD1 is obtained. In order to determine the movement of only the subject existing in the central area CRT1, it is necessary to eliminate the movement of the entire object scene, so the ratio Rard is multiplied by a constant a, and the ratio Rctr is subtracted by the multiplication value a * Rard. ing. This gives the probability Psprt as a percentage. Note that the possibility Psprt shows a high numerical value when shooting a scene in which a baseball pitcher throws a ball as shown in FIG.

The possibility determination process of step S23 shown in FIG. 6 follows the subroutine shown in FIGS.
First, the positions of the zoom lens 12 and the focus lens 14 are detected in step S141, and the distance L1 from the zoom lens 12 to the main subject and the focus lens 14 and the image sensor 2 are detected in steps S143 and S145.
The distance L2 from 0 is detected (see FIG. 18).

The ROM 58 stores the graph shown in FIG. According to FIG. 17, the horizontal axis and the vertical axis are the positions of the zoom lens 12 and the focus lens 14, respectively. The position of the zoom lens 12 is the driver 28a.
Is represented by the number of steps of a stepping motor (not shown) provided in the driver, and the position of the focus lens 14 is represented by the number of steps of a stepping motor (not shown) provided in the driver 28b. A plurality of curves A to I corresponding to the distance to the main subject are drawn on the plane formed by the vertical axis and the horizontal axis. Curve A ~
I is the distance to the subject is 0.4 m and 0.5, respectively.
m, 0.6m, 0.8m, 1.0m, 1.5m, 2.0
The lens positional relationship at m, 3.0 m and infinity (∞) is shown.

Therefore, in step S143, the distance L1 is detected based on the positions of the zoom lens 12 and the focus lens 14 obtained in step S141 and the graph shown in FIG. Further, in step S145, the focus lens 1 obtained in step S141
The distance L2 is detected from the position of 4.

In step S147, the vertical block number FC of the area (face area) to be occupied by the person's face on the screen, that is, the object scene, is calculated according to equations (5) to (6).

[0066]

## EQU00005 ## face2 = face1 * L2 / L1 face1: length of person's face (constant: 30 cm) face2: the length of the face image projected on the light receiving surface

[0067]

FC = 16 * face2 / hh: Vertical size of light receiving surface formed on image sensor Referring to FIG. 18, a person (main subject) having a face length face1 is at a distance L1 from the zoom lens 12. Assuming that the face image is present at a position distant from the image sensor 20, the length face2 of the face image projected on the light receiving surface of the image sensor 20 is
This corresponds to a value obtained by multiplying 1 by L2 / L1. In addition, the number of vertical blocks FC of the face area projected on the light receiving surface is fac
This corresponds to a value obtained by dividing e2 by the vertical size h of the light receiving surface and multiplying the divided value by "16".

Returning to FIG. 15, in steps S149 and S151, the calculated vertical block number FC is determined. When the number of vertical blocks FC is less than "2", the number of vertical blocks FC is set to "2" in step S153, and then the process proceeds to step S156. When the number FC of vertical blocks exceeds "8", the number FC of vertical blocks is set to "8" in step S155, and then the process proceeds to step S156. On the other hand, when the condition of 2 ≦ FC ≦ 8 is satisfied, the process directly proceeds to step S156. In this way, the area of the face area is set within the range of 2 blocks × 2 blocks to 8 blocks × 8 blocks.

At step S156, variables Cin and Cout are set to "0", and at steps S157 and S157.
At 159, the vertical position number i and the horizontal position number j are set to "0". In step S161, the register r
The integrated values r (i, j), g (i,
j) and b (i, j) are read out, and the color evaluation values R and G of the block (i, j) are calculated according to Equation 7.

[0070]

## EQU00007 ## R = r (i, j) / (r (i, j) + g (i,
j) + b (i, j)) G = g (i, j) / (r (i, j) + g (i, j) + b
(I, j)) In step S163, it is determined whether the calculated color evaluation values R and B belong to the skin color area SKN shown in FIG.
If NO, the process proceeds directly to step S171. on the other hand,
If YES in step S163, the flow proceeds to step S165, and block (i, j) is processed in steps S147 to S15.
It is determined whether or not it belongs to the face area defined by the size obtained in 5. Specifically, it is determined whether both the condition shown in Formula 8 and the condition shown in Formula 9 are satisfied. Then, if both of these conditions are satisfied, the block (i, j) is considered to belong to the face region, and if either of these conditions is not satisfied, the block (i, j) is satisfied.
j) is considered not to belong to the face area.

[0071]

[Equation 8] 8-FC / 2 ≦ i ≦ 7 + FC / 2

[0072]

## EQU00009 ## 8-FC / 2.ltoreq.j.ltoreq.7 + FC / 2 According to Eqs. 8 and 9, the face region is formed at the center of the screen. For example, if FC = 6, the area FACE indicated by hatching in FIG. 20 is set as the face area. When the block (i, j) belongs to such a face area, the variable Cin is incremented in step S167 and then step S171.
If the block (i, j) does not belong to the face area, the variable Cout is incremented in step S169 and then the process proceeds to step S171.

In step S171, the horizontal position number j is incremented, and in step S173, the incremented horizontal position number j is compared with "16". And
If j <16, the process returns to step S161, but j = 16
If so, the process proceeds to step S175. In step S175, the vertical position number i is incremented, and step S17
In 7, the incremented vertical position number i is "16".
Compare with. If i <16, step S15
9, but if i = 16, the process proceeds to step S179. By performing such processing, the variable Cin indicates the number of skin color blocks belonging to the face area, and the variable Cout.
Indicates the number of skin color blocks that do not belong to the face area.

In step S179, the probability Pptrt that the object scene is a portrait scene is calculated according to equation 10. When the possibility Pptrt is calculated, the process returns to the upper hierarchy routine.

[0075]

## EQU10 ## Pptrt = (Cin-Cout * n) / F
C ² * 100 n: The larger the number of flesh-colored blocks belonging to the constant face area, the higher the possibility that the subject has a human face, and the smaller the number of flesh-colored blocks belonging to the face area, the higher the possibility that a subject has a human face. Is low. However, the more skin color blocks that exist in areas other than the face area, the less likely the subject is a human face. Therefore, the variable Cin is subtracted by the multiplication value obtained by multiplying the variable Cout by the constant n. On the other hand, FC ²
Is the total number of blocks belonging to the face area, and by dividing the subtraction value by the total number of blocks and multiplying the division value by “100”, the possibility Pptrt that the object scene is a portrait scene is obtained as a percentage. In addition,
When a person such as shown in FIG. 21 shoots a scene in the center of the screen, the possibility Pptrt shows a high numerical value.

The possibility determination process of step S25 shown in FIG. 6 follows the subroutine shown in FIGS.
First, in step S181, the variable Css is set to "0",
In step S183, the vertical position number i is set to "0",
Then, in step S185, the horizontal position number j is set to "0". In step S187, it is determined whether the vertical position number i and the horizontal position number j satisfy the conditions of 6 ≦ i ≦ 9 and 6 ≦ j ≦ 9. Then, if none of the conditions is satisfied, the process proceeds to step S197, but if at least one of the conditions is satisfied, the process proceeds to step S189.

In step S189, the register rgst
The integrated values r (i, j), g (i, j) and b (i, j) set to 3 are read out, and the color evaluation values R and G of the block (i, j) are calculated according to the above-mentioned equation 7. To do. In step S191, the calculated color evaluation values R and G are shown in FIG.
7, it is determined whether or not it belongs to the evening scene area EVEN indicated by diagonal lines. If NO, the process proceeds to step S195, but YES
If so, the process proceeds to step S193. In step S193, the integrated value y (i, i, i) set in the register rgst3 is set.
j) is read, and it is determined whether the integrated value y (i, j) has high brightness. Specifically, it is determined whether y (i, j) exceeds the threshold value Yth2. And y
If (i, j) ≦ Yth2, then step S19 is performed.
7, if y (i, j)> Yth2, the variable Css is incremented in step S195 and then the process proceeds to step S197.

In step S197, the horizontal position number j is incremented, and in step S201, the incremented horizontal position number j is compared with "16". And
If j <16, the process returns to step S187, but j = 16
If so, the process proceeds to step S201. In step S201, the vertical position number i is incremented, and in step S20
In 3, the incremented vertical position number i is "16".
Compare with. If i <16, step S18
5, the process proceeds to step S205 if i = 16.

By executing such processing, as shown in FIG.
For each block forming the cross-shaped region CRS1 indicated by hatching in FIG. 5, a determination process as to whether it is an evening scene and a determination process as to whether it is a high luminance are performed. Variable Cs
s will indicate the number of high-luminance blocks in the evening scene.

In step S205 shown in FIG. 23, the peripheral luminance values Yupper, Ylower, Yleft and Yright are set to "0". Also, step S2
In 07 and S209, the vertical position number i and the horizontal position number j are set to "0". In step S211, 0 ≦
It is determined whether the condition of i ≦ 1 is satisfied, it is determined in step S215 whether the condition of 14 ≦ i ≦ 15 is satisfied, and in step S219 it is determined whether the condition of 0 ≦ j ≦ 1 is satisfied. And then step S225
Then, it is determined whether or not the condition of 14 ≦ j ≦ 15 is satisfied.

If YES is determined in the step S211, the process proceeds to a step S213, and the integrated value y (i, j) stored in the register rgst3 is added to the peripheral brightness value Yupper. If YES is determined in the step S215, the process proceeds to a step S217, and the integrated value y (i, j) stored in the register rgst3 is added to the peripheral luminance value Ylower. If YES is determined in the step S219, the process proceeds to a step S221, and the integrated value y (i, j) stored in the register rgst3 is added to the peripheral luminance value Yleft. If YES is determined in step S225, step S22
7, the integrated value y stored in the register rgst3
(I, j) is added to the peripheral luminance value Yright.

The horizontal position number j is incremented in step S229, and the incremented horizontal position number j is compared with "16" in step S231. And
If j <16, the process returns to step S211, but j = 16
If so, the process proceeds to step S233. In step S233, the vertical position number i is incremented, and step S23
In 5, the incremented vertical position number i is "16".
Compare with. If i <16, step S20
9, but if i = 16, the process proceeds to step S237.

The peripheral luminance values Yupper, Ylower, Yleft and Y obtained by the above processing
light indicates the luminance of the peripheral areas ARD2a, ARD2b, ARD2c, and ARD2d indicated by the slanted lines and the thick line in FIG. 26, respectively.

In step S237, the peripheral luminance value Yup
The difference between per and Ylower is detected, and based on the difference, it is determined whether or not the difference in brightness between the top and bottom of the screen is large. In step S239, the difference between the peripheral luminance values Yleft and Yright is detected, and it is determined based on the difference whether the luminance difference between the left and right sides of the screen is large. Specifically, in step S237, it is determined whether the condition of expression 11 is satisfied, and in step S239, it is determined whether the condition of expression 12 is satisfied.

[0085]

[Equation 11] | Yupper-Ylower |> Yth3

[0086]

[Expression 12] | Yleft-Yright |> Yth4 Then, if neither the condition of the expression 11 nor the condition of the expression 12 is satisfied, the process proceeds to step S241, and the possibility Peven that the scene is an evening scene is set to "0". Set. On the other hand, if at least one of the condition of expression 11 and the condition of expression 12 is satisfied, the possibility Peven is calculated according to expression 13 in step S243. Step S241 or S
When the processing of step 243 is completed, the process returns to the upper layer routine.

[0087]

[Equation 13] Peven = Css / 112 * 100 The number of blocks forming the cross region CRS1 shown in FIG. 25 is “112”, and the variable Css is a block having high brightness in the evening scene among the blocks forming the cross region CRS1. Is the number of. Therefore, by multiplying the divided value obtained by dividing the variable Css by “112” by “100”, the possibility Peven that the object scene is an evening scene can be obtained as a percentage. Note that the possibility Peven shows a high numerical value when shooting a scene in which the setting sun sets in the mountains as shown in FIG. 28.

The possibility determination process of step S27 shown in FIG. 6 follows the subroutine shown in FIGS. 29 and 30.
First, the total luminance value Ysum is set to "0" in step S251, the vertical position number i is set to "0" in step S253, and the horizontal position number j is set to "0" in step S255. In step S257, the register rgst3
The integrated value y (i, j) is read from the
(I, j) is added to the brightness sum value Ysum.

In step S259, the horizontal position number j is incremented, and in step S261, the incremented horizontal position number j is compared with "16". And
If j <16, the process returns to step S257, but j = 16
If so, the process proceeds to step S263. In step S263, the vertical position number i is incremented, and step S26
In 5, the incremented vertical position number i is "16".
Compare with. If i <16, step S25
Returning to step 5, if i = 16, the number 1 is calculated in step S267.
The average luminance value Yavr is calculated according to 4.

[0090]

[Equation 14] Yavr = Ysum / 256 By repeating steps S255 to S265, the luminance sum total value Ysum indicates the sum total of the integrated values y (i, j) set in the register rgst3. Since the total number of blocks formed on the screen is “256”, the average brightness value obtained by Expression 14 indicates the average brightness of the screen.

In step S269, the variable Cni is set to "0", and in step S271, the vertical position number i is set.
Is set to "0", and the horizontal position number j is set to "0" in step S273. In step S275, the integrated value y (i, j) is read from the register rgst3, and it is determined whether the integrated value y (i, j) satisfies the condition shown in Expression 15.

[0092]

[Mathematical formula-see original document] y (i, j)> Yavr * m m: constant If the condition shown in Formula 15 is not satisfied, the process directly proceeds to step S279, but if the condition shown in Formula 15 is satisfied, the variable Cni is set in step S277. After incrementing, the process proceeds to step S279.

In step S279, the horizontal position number j is incremented, and in step S281 the incremented horizontal position number j is compared with "16". And
If j <16, the process returns to step S275, but j = 16
If so, the process proceeds to step S283. In step S283, the vertical position number i is incremented, and step S28
In 5, the incremented vertical position number i is "16".
Compare with. If i <16, step S27
Returning to step 3, if i = 16, the process proceeds to step S287.

By repeating steps S273 to S285, the integrated value y of all blocks forming the screen is obtained.
(I, j) is compared with a multiplication value obtained by multiplying the average luminance value Yavr by a constant m. The variable Cni corresponds to the number of high-luminance blocks in which the integral value y (i, j) exceeds the multiplication value among the 256 blocks forming the screen.

In step S287, the average luminance value Yavr
Is compared with a threshold value Ynight, and Yavr ≧ Ynight
If so, in step S289, the possibility Pnight of the scene being a night scene is set to "0". On the other hand, Ya
If vr <Ynight, the variable Cni is compared with the threshold value Cmax in step S291. And Cni ≦ C
If it is max, the process directly proceeds to step S295, but C
If ni> Cmax, the variable Cni is calculated in step S293.
Is set to the threshold value Cmax and then the process proceeds to step S295. In step S295, the probability Pn
Calculate light. Step S289 or S295
When the process of is finished, the process returns to the upper hierarchy routine.

[0096]

[Expression 16] Pnight = Cni / Cmax * 100 If the luminance average value Yavr is low, the object scene may be a night scene, and therefore the operation of Expression 16 is performed when the condition of Yavr <Ynight is satisfied. Variable Cni noticed when the average luminance value Vavr is sufficiently small
Can be regarded as the number of high-brightness regions scattered in the night view. Further, as the number of high-luminance regions increases, the luminance average value Yavr
However, the process proceeds to step S295 when the brightness of the area other than the high brightness area is considerably low. Therefore, in the calculation executed when the average luminance value Yavr is sufficiently low, by placing importance on the variable Cni related to the size of the high luminance region, the possibility Pnight of the scene being a night scene can be obtained. Note that the possibility Pnight shows a high numerical value when a high-rise group of buildings is photographed at night in which window lights are scattered as shown in FIG.

As can be seen from the above description, in determining the possibility of a portrait scene, the zoom lens 12
And the distance L1 to the main subject existing in the object scene is measured based on the position of the focus lens 14 (S14).
1, S143), the distance L2 between the focus lens 14 and the image sensor 20 is measured based on the position of the focus lens 14 (S145). The face area that the face of the person should occupy on the light receiving surface of the image sensor 14 is specified based on the measured distance L1 and the interval L2 (S
147-S155). On the other hand, the skin color region of the object scene is detected based on the integrated values r (i, j), g (i, j) and b (i, j) (S163). The shooting mode is determined based on the size of the skin color area included in the face area and the size of the skin color area outside the face area (S179, S2).
9). That is, if the possibility ptrt calculated based on each size is the largest, the shooting mode is determined to be the portrait mode. The object scene is photographed in the decided photographing mode (S39).

As described above, since the photographing mode is determined based on the relationship between the face area specified based on the distance L1 to the main subject and the skin color area of the field, the operability of the camera is improved. be able to.

In this embodiment, a digital camera is used for explanation, but the present invention can of course be applied to an analog video camera and a silver salt camera.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention.

FIG. 2 is a block diagram showing an example of a configuration of a signal processing circuit applied to the embodiment in FIG.

FIG. 3 is an illustrative view showing one example of a block formed on a screen.

FIG. 4 is an illustrative view showing one example of a mapping state of an SDRAM applied to the embodiment in FIG.

FIG. 5 is a flowchart showing a part of the operation of FIG. 1 embodiment.

FIG. 6 is a flowchart showing another portion of the operation of FIG. 1 embodiment.

FIG. 7 is a flowchart showing another portion of the operation of the embodiment in FIG. 1;

FIG. 8 is a flowchart showing yet another portion of the operation of the embodiment in FIG. 1;

FIG. 9 is a flowchart showing another portion of the operation of the embodiment in FIG. 1;

FIG. 10 is a flowchart showing another portion of the operation of the embodiment in FIG. 1;

FIG. 11 is a flowchart showing yet another portion of the operation of the embodiment in FIG. 1;

FIG. 12 is a flowchart showing another portion of the operation of the embodiment in FIG. 1;

FIG. 13 is an illustrative view showing a part of the operation of the embodiment in FIG. 1;

FIG. 14 is an illustrative view showing one example of a sports scene.

FIG. 15 is a flowchart showing one portion of the operation of the embodiment in FIG.

FIG. 16 is a flowchart showing another portion of the operation of the embodiment in FIG. 1;

FIG. 17 is a graph showing a relationship between a zoom lens position and a focus lens position and a subject distance.

FIG. 18 is an illustrative view showing a part of the operation of the embodiment in FIG. 1;

FIG. 19 is a plan view showing a distribution state of color evaluation values.

FIG. 20 is an illustrative view showing a part of the operation of the embodiment in FIG. 1;

FIG. 21 is an illustrative view showing one example of a portrait scene.

22 is a flowchart showing one portion of behavior of FIG. 1 embodiment. FIG.

FIG. 23 is a flowchart showing another portion of the operation of the embodiment in FIG. 1;

FIG. 24 is a flowchart showing another portion of the operation of the embodiment in FIG. 1;

FIG. 25 is an illustrative view showing a part of the operation of the embodiment in FIG. 1;

FIG. 26 is an illustrative view showing another portion of the operation of the embodiment in FIG. 1;

FIG. 27 is a plan view showing a distribution state of color evaluation values.

FIG. 28 is an illustrative view showing one example of a sunset scene.

FIG. 29 is a flowchart showing one portion of behavior of the embodiment in FIG. 1;

FIG. 30 is a flowchart showing another portion of the operation of the embodiment in FIG. 1;

FIG. 31 is an illustrative view showing one example of a night scene.

[Explanation of symbols]

10 ... Digital camera 12 ... Zoom lens 14 ... Focus lens 20 ... Image sensor 26 ... Signal processing circuit 32 ... CPU 36 ... Memory control circuit 38 ... SDRAM 44 ... Monitor 46 ... JPEG codec 48 ... Recording medium 54 ... Shutter button 56 ... Zoom button

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) H04N 5/238 H04N 101: 00 5C065 9/04 G02B 7/11 D // H04N 101: 00 G03B 3/00 AF term (reference) 2H002 AB03 FB56 2H011 BA31 DA00 2H051 BA47 EA28 2H102 AA66 5C022 AA13 AB66 AC11 AC32 AC42 AC54 AC69 5C065 AA03 BB02 BB16 BB30 DD01 GG24

Claims (5)

[Claims] 1. A camera for shooting in a shooting mode according to an object scene, said first measuring means for measuring a distance to a main subject existing in said object scene, and said face area to be occupied by a human face. Based on the distance measured by the first measuring means, specifying means for specifying from the object scene, detecting means for detecting a skin color area of the object scene, and based on a relationship between the face area and the skin color area, A camera, comprising: a determination unit that determines a shooting mode. 2. The camera according to claim 1, further comprising a zoom lens, wherein the first measuring unit measures the distance based on a position of the zoom lens. 3. A focus lens, an image sensor in which an optical image of the field of view that has passed through the focus lens is incident on a light receiving surface, and second measuring means for measuring a distance between the focus lens and the light receiving surface. Wherein the specifying unit specifies a face area of the object scene image projected on the light receiving surface based on the distance measured by the first measuring unit and the distance measured by the second measuring unit. The camera according to Item 1 or 2. 4. The first skin color area discriminating means for discriminating the size of the first skin color area included in the face area,
Second skin color area determination means for determining the size of the second skin color area deviating from the face area, and the photographing mode is set to the portrait mode based on the size of the first skin color area and the size of the second skin color area. The camera according to claim 1, further comprising a mode setting unit. 5. The camera according to claim 1, further comprising output means for outputting a message corresponding to the photographing mode determined by the determination means.