JP5134116B2 - Imaging apparatus and in-focus position search method - Google Patents

Description

  The present invention relates to a photographing apparatus having a face detection AF function.

  2. Description of the Related Art Conventionally, there are imaging devices that determine a backlight state. For example, according to Patent Document 1, it is determined whether or not the subject is backlit. When the subject is backlit, the integration time of the charge storage type image sensor is lengthened and integration is performed again, and the focus is refocused based on the sensor output at that time. Redo detection.

  According to Patent Document 2, there are two types of sampling means (at least a part of the accumulation time is common), and when it is determined that the light is backlit, focus detection is performed with priority given to the output of the second sampling means.

  According to Patent Document 3, a focus evaluation value is obtained by detecting a signal amount of a high-frequency component of an image signal, a focus position is detected based on the focus evaluation value, and the detected focus position is It is determined whether the focus position is a false focus position.

Japanese Patent Laid-Open No. 05-264887 Japanese Patent Laid-Open No. 11-223759 JP 2006-145666 A

  In the technique of Patent Document 1, in the case of a backlight state, it takes time for focus detection because it performs two integrations and two computations for convenience. In the technique of Patent Document 2, it is necessary to switch the operation of the image sensor in order to change the AF charge accumulation time according to the result of the backlight determination. In Patent Document 3, focusing suitable for a shooting scene is not always performed.

  The present invention has been made in view of such problems, and an object thereof is to increase the focusing rate by performing AF control suitable for a scene.

  The present invention forms a subject image on a photographing element through a photographing lens, extracts a desired frequency component from an image signal obtained from the photographing element by a filter, and integrates the extracted frequency component to evaluate a focus. The present invention relates to an imaging apparatus or an in-focus position search method for calculating a value and searching for an in-focus position of an imaging lens based on the calculated focus evaluation value.

  A photographing apparatus according to the present invention includes a scene discriminating unit that discriminates a photographing scene of a subject image, and a frequency component to be extracted by a search range of a focusing position of a photographing lens or a filter based on the photographing scene discriminated by the scene discriminating unit. And a control unit that moves the photographic lens so as to search for an in-focus position after changing at least one of them.

  According to the present invention, since at least one of the in-focus position search range and the in-focus detection filter is changed according to the shooting scene, and the in-focus position is calculated, it is possible to capture the presence or absence of backlight or the presence or absence of a face. It is possible to search for an accurate in-focus position according to the scene.

  Here, the shooting scene includes a backlit state / front lighted state, shooting of a person or a subject other than a person, brightness of an image, high contrast / low contrast.

  In addition, a photographing apparatus according to the present invention includes a scene discriminating unit that discriminates a photographing scene, and a control unit that executes a focusing position search procedure corresponding to the photographing scene discriminated by the scene discriminating unit.

  According to the present invention, it is possible to perform accurate focusing according to the shooting scene by executing an appropriate focus position calculation procedure according to the shooting scene.

  The imaging apparatus according to the present invention includes a first procedure for calculating a focus evaluation value based on a signal output from the filter, and a second procedure different from the first procedure, respectively. Whether the difference between the temporary in-focus position and the second in-focus position is within a predetermined range, and the difference between the first in-focus position and the second in-focus position. And a control unit that moves the photographic lens to the first provisional focusing position as the final focusing position in response to determining that the difference is within a predetermined range.

  According to the present invention, the temporary in-focus position is calculated in two different procedures, and if those errors are within a predetermined range, the first in-focus position is set as the final in-focus position, The certainty of the temporary focusing position of 1 is ensured as a true focusing position, and accurate focusing can be performed.

  The temporary focus position detection unit detects at least one of the focus evaluation value calculation target region at the first temporary focus position and the frequency component to be extracted by the filter when it is determined that the difference is not within the predetermined range. After changing one, the first temporary focus position is calculated again, and the control unit determines that the difference between the first temporary focus position detected again and the second temporary focus position is within a predetermined range. It may be determined whether the difference is within a predetermined range, and the photographic lens may be moved with the first temporary focusing position as the final focusing position.

  The control unit may repeat recalculation of the first temporary in-focus position until it is determined that the difference is within a predetermined range.

  The control unit may move the photographic lens with the predetermined reference position as the final in-focus position in response to the calculation of the first temporary in-focus position exceeding a predetermined number of 1 or more. Good.

  A scene discriminating unit for discriminating a photographic scene may be further provided, and the control unit may change at least one of a predetermined range or a predetermined reference position based on the photographic scene determined by the scene discriminating unit.

  The scene determination unit detects a backlight state, and the control unit compares two temporary in-focus positions calculated for two different focus evaluation value calculation target areas in accordance with the detection of the backlight state by the scene determination unit. If the difference is within a predetermined range, the two temporary in-focus positions detected for either one of the two focus evaluation value calculation target areas may be set as the final in-focus position.

  A face detection unit that detects a face region based on an image signal, and a region determination that determines at least one of a neck region, a chest region, an abdominal region, and a foot region based on the face region detected by the face detection unit And a section.

  In this case, the two different focus evaluation value calculation target regions may include any two of a face region, a neck region, a chest region, an abdominal region, a foot region, and a region obtained by combining all or part of them.

  The focus evaluation value calculation target region may include any one of a face region, a neck region, a chest region, an abdominal region, a foot region, and a region obtained by combining all or part thereof.

  The frequency component to be extracted by the filter may be switched from a predetermined first frequency component to a predetermined second frequency component lower than the first frequency component.

  The focus position searching method according to the present invention includes a step of determining a shooting scene of a subject image and at least a frequency range to be extracted from a search range of a focus position of a shooting lens or a filter based on the determined shooting scene. And changing either one of them, and moving the photographic lens so as to search for the in-focus position.

  The in-focus position searching method according to the present invention includes a step of determining a photographic scene and a step of executing a focus position search procedure corresponding to the determined photographic scene.

  The in-focus position search method according to the present invention includes a first procedure for calculating a focus evaluation value based on a signal output from a filter and a second procedure different from the first procedure, respectively. Calculating a temporary focusing position and a second temporary focusing position, and determining whether or not a difference between the first temporary focusing position and the second temporary focusing position is within a predetermined range. And, when the difference is determined to be within a predetermined range, the step of moving the photographic lens to the first temporary focusing position as the final focusing position is provided.

  According to the present invention, since at least one of the in-focus position detection range and the in-focus detection filter is changed according to the shooting scene and the in-focus position is calculated, it is possible to capture the presence / absence of backlight or the presence / absence of a face. Accurate focusing according to the scene can be performed.

Front view of digital camera Rear view of digital camera Block diagram of digital camera Block diagram of face detection LSI The figure which shows an example of the face information which a face detection LSI outputs Outline flowchart of shooting operation Outline flowchart of face detection AF Outline flowchart of AF target face area setting Backlight determination processing flowchart Conceptual illustration of backlight judgment The figure which shows an example of the flag which shows the result of backlight determination Flowchart of backlight face AF processing according to the first embodiment The figure which shows the content of AF area expansion processing notionally Flow chart of follow-up AF processing The principal part flowchart of the backlit face AF processing according to the second embodiment

  Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.

  FIG. 1 is a front view of a digital camera (hereinafter abbreviated as a camera) 100 according to a preferred embodiment of the present invention.

  A lens barrel 60 provided in front of the camera 100 includes a photographic lens 101 including a zoom lens 101a and a focus lens 101b, and the focal length is adjusted by moving the zoom lens 101a in the optical axis direction. In addition, focus adjustment is performed by moving the focus lens 101b in the optical axis direction.

  The lens barrel 60 is advanced from the camera body 180 by moving forward and backward between the wide end, which is the shortest focal length position, and the tele end, which is the longest focal length position, from the state where the lens barrel 60 is retracted. Also stored. In this figure, a state in which the lens barrel 60 is retracted into the camera body 180 is shown.

  Further, the camera 100 creates a state in which the photographing lens 101 is protected by covering the front surface of the photographing lens 101 and blocking the photographing lens 101 and the outside when not photographing, and a lens cover 61 that exposes the photographing lens to the outside when photographing. Is provided.

  The lens cover 61 is configured by a mechanism that can be freely opened and closed. The lens cover 61 covers the front surface of the photographing lens 101 in an open state, and exposes the front surface of the photographing lens 101 to the outside in a closed state. The lens cover 61 is opened / closed in conjunction with the power switch 121 being turned on / off. In this figure, the lens cover 61 is open.

  On the upper surface of the camera 100, a mode dial 123 and a power switch 121 provided with a release switch 104 at the center are provided, and a strobe 105a, an AF auxiliary light lamp 105b, a self-timer lamp 105c, and the like are provided on the front. Has been deployed.

  FIG. 2 is a rear view of the camera 100. A zoom switch 127 is provided on the back of the camera 100. When the wide (W) side of the zoom switch 127 is pressed, the lens barrel 60 is extended to the wide end (wide angle) side while being pressed, and when the other side of the tele (T) side is pressed, the lens barrel 60 is being pressed. The lens barrel 60 moves to the telephoto end (telephoto) side.

  On the back of the camera 100, an image display LCD 102, a cross key 124, a face button 125, an information position designation key 126, and the like are also provided. The cross key 124 is an operation system for setting display brightness adjustment / self-timer / macro shooting / flash shooting on the top, bottom, left and right respectively. As will be described later, by pressing the lower key of the cross key 124, the main CPU 20 can set the self-photographing mode in which the CCD 132 performs the shutter operation after the time measurement of the self-timer circuit 83 is completed. When the face button 125 is pressed when the shooting mode is set, face detection described later is started.

  FIG. 3 is a block diagram of the camera 100. The camera 100 is provided with an operation unit 120 for performing various operations when the user uses the camera 100. The operation unit 120 includes a power switch 121 for turning on the power for operating the camera 100, a mode dial 123 for selecting auto shooting, manual shooting, landscape shooting mode, portrait shooting mode, ID photo mode, and the like. A cross key 124 for setting, selecting or zooming various menus, a face button 125, and an information position specifying key 126 for executing or canceling the menu selected with the cross key 124 are provided.

  In addition, the camera 100 is provided with an image display LCD 102 for displaying captured images, reproduced images, and the like, and an operation LCD display 103 for assisting operations.

  The camera 100 is provided with a release switch 104. The release switch 104 transmits a shooting start instruction to the main CPU 20. In the camera 100, switching between photographing and reproduction can be freely performed on a predetermined menu screen. Further, the camera 100 is provided with a flash light emitting device having an AF auxiliary light lamp 105b composed of a light emitting diode (LED) for irradiating a subject with a light projection spot during contrast AF, and a strobe 105a that emits flash light.

  In addition, the camera 100 includes a photographing lens 101, a diaphragm 131, and a CCD sensor 132 (hereinafter referred to as a CCD 132) that is an imaging device that converts a subject image formed through the photographing lens 101 and the diaphragm 131 into an analog image signal. For short). The CCD 132 generates an image signal by accumulating charges generated by subject light irradiated on the CCD 132 for a variable charge accumulation time (exposure period). The CCD 132 sequentially outputs image signals for each frame at a timing synchronized with the vertical synchronization signal VD output from the CG unit 136.

  When the CCD 132 is used as the image sensor, an optical low-pass filter 132a that removes unnecessary high-frequency components in the incident light is disposed in order to prevent the generation of color false signals and moire fringes. In addition, an infrared cut filter 132b that absorbs or reflects infrared light in incident light and corrects a sensitivity characteristic unique to the CCD sensor 132 having high sensitivity in a long wavelength region is provided. The specific arrangement of the optical low-pass filter 132a and the infrared cut filter 132b is not particularly limited.

  The camera 100 also adjusts the white balance of the subject image represented by the analog image signal from the CCD sensor 132 and adjusts the slope (γ) of the straight line in the gradation characteristics of the subject image, and further amplifies the analog image signal. A white balance / γ processing unit 133 including an amplifier with a variable gain is provided.

  Further, the camera 100 includes an A / D converter 134 that A / D converts an analog signal from the white balance / γ processing unit 133 into digital R, G, B image data, and an A / D converter 134. A buffer memory 135 for storing R, G, B image data is provided.

  The R, G, B image data obtained by the A / D conversion unit 134 is also input to the AF detection unit 150.

  The AF detection unit 150 can switch either one of two different high-pass filters having a high or low pass frequency band of image data (a frequency band to be passed without being cut off) as an AF filter. The AF detection unit 150 calculates an AF evaluation value, which will be described later, by using one of these filters preferentially as an AF filter.

  Which filter is preferentially used may be stored in the EEPROM 146 in advance in the default priority order, may be switched according to the result of backlight determination processing described later, or may be switched according to the result of face detection. May be.

  For example, when it is determined by the CPU 20 that the light is forward, when it is determined that the scene is bright, when it is determined that the landscape mode is selected with the mode dial 123, or when it is determined that no face is detected, Even if a face area is detected, if it is determined that the area of the face area is smaller than a predetermined ratio (for example, 1/10) as compared to the area of the entire image (that is, the amount of landscape data is large), multiple areas are divided by division photometry. When the brightness is determined for each area and the difference in brightness between adjacent areas is determined to be greater than or equal to a predetermined level, such as when it is determined to be a high-contrast shooting scene, there are many high-frequency components in the entire image data or in the face area. Therefore, it is preferable to switch to a filter having a higher pass frequency band.

  Alternatively, when it is determined that the light is backlit, when the night view mode is selected, when the area of the face area is smaller than a predetermined ratio, when the scene is dark, or when the scene is low-contrast, other than the face area Since the amount of high frequency components in the region increases, the amount of image data as a whole, or the amount of image data in the face region decreases, the one with the lower pass frequency band to extract the predetermined frequency components as much as possible You can switch to this filter.

  Alternatively, when a face area is not detected, such as a landscape image, the high-pass filter having a higher pass frequency band may be switched.

  Hereinafter, the priority AF filter may be represented by First_Filter, and the non-prioritized AF filter may be represented by Second_Filter. For the sake of simplicity, the higher pass frequency band is referred to as First_Filter and the lower pass frequency band is referred to as Second_Filter. However, the pass frequency bands of these filters may be reversed.

  The AF detection unit 150 uses all or a part (for example, only the G signal) of R, G, B image data corresponding to a desired integration area (AF region) among the image data from the A / D conversion unit 134. Then, a predetermined frequency component is extracted from the image data that is continuous in the one-dimensional horizontal direction or the vertical direction through an AF filter whose pass frequency band can be switched, and a value (AF evaluation value) obtained by integrating these frequency components is CPU 20. Output to.

  The AF detection unit 150 may simultaneously extract different frequency components for both First_Filter and Second_Filter from an image obtained at a single opportunity, and simultaneously output two AF evaluation values corresponding to them. When the AF evaluation value is calculated by sequentially switching First_Filter and Second_Filter, it is necessary to move the focus lens 101b twice when the AF filter is switched to First_Filter and when the AF filter is switched to Second_Filter. It only takes one movement.

  The CPU 20 moves the focus lens 101 b via the focus motor 111 to the lens position where the AF evaluation value output from the AF detection unit 150 is maximized.

  Further, the camera 100 includes a CG (clock generator) unit 136, a photometry / ranging CPU 137, a charge / light emission control unit 138, a communication control unit 139, a YC processing unit 140, and a power supply battery 68. It has been.

  The CG unit 136 includes a vertical synchronization signal VD for driving the CCD sensor 132, a drive signal including a high-speed sweep pulse P, a control signal for controlling the white balance / γ processing unit 133, the A / D conversion unit 134, and communication control. A control signal for controlling the unit 139 is output. Further, a control signal from the photometry / ranging CPU 137 is input to the CG unit 136.

  The photometry / ranging CPU 137 controls the zoom motor 110, the focus motor 111, and the aperture motor 112 for adjusting the aperture to drive the zoom lens 101a, the focus lens 101b, and the aperture 131, respectively, thereby driving the distance to the subject. Is calculated (ranging), and the CG unit 136 and the charge / light emission control unit 138 are controlled. Driving of the zoom motor 110, the focus motor 111, and the aperture motor 112 is controlled by the motor driver 62, and a control command for the motor driver 62 is sent from the photometry / ranging CPU 137 or the main CPU 20.

  The driving source of the zoom lens 101a, the focus lens 101b, the stop 131, and the AF auxiliary light irradiation angle is not necessarily limited to various motors such as the zoom motor 110, the focus motor 111, and the stop motor 112. It may be.

  The CPU 137 for photometry / ranging is based on image data (through image) obtained periodically (every 1/30 seconds to 1/60 seconds) by the CCD 132 when the release switch 104 is half-pressed (S1 is turned on). Measure the brightness of the subject (calculate EV value).

  That is, the AE calculation unit 151 integrates the R, G, and B image signals output from the A / D conversion unit 134 and provides the integrated values to the photometry / ranging CPU 137. The photometry / ranging CPU 137 detects the average brightness (subject brightness) of the subject based on the integrated value input from the AE calculation unit 151, and calculates an exposure value (EV value) suitable for photographing.

  Then, the photometry / ranging CPU 137 determines the exposure value including the aperture value (F value) of the aperture 131 and the electronic shutter (shutter speed) of the CCD 132 based on the obtained EV value according to a predetermined program diagram. (AE operation).

  When the release switch 104 is fully pressed (S2 is turned on), the photometry / ranging CPU 137 drives the aperture 131 based on the determined aperture value, controls the aperture diameter of the aperture 131, and determines the determined shutter speed. Based on the above, the charge accumulation time in the CCD 132 is controlled via the CG 136.

  The AE operation includes an aperture priority AE, a shutter speed priority AE, a program AE, etc. In any case, the subject brightness is measured, and an exposure value determined based on the photometric value of the subject brightness, that is, an aperture value and a shutter. By taking a picture in combination with the speed, control is performed so that an image is taken with an appropriate exposure amount, and it is possible to save troublesome determination of exposure.

  The AF detection unit 150 extracts image data corresponding to a desired AF area selected by the photometry / ranging CPU 137 from the A / D conversion unit 134. The method for searching and detecting the focal position is performed using the feature that the high frequency component of the AF area portion of the image data has the maximum amplitude at the in-focus position. The AF detection unit 150 calculates an amplitude value by integrating the high-frequency component of the extracted image data for one field period. The AF detector 150 is configured such that the photometry / ranging CPU 137 drives and controls the focus motor 111 to move the focus lens 101b within the movable range, that is, between the end point (INF point) on the infinity side and the end point (NEAR point) on the near side. When the maximum amplitude is detected, the detection value is transmitted to the photometry / ranging CPU 137 when the maximum amplitude is detected. However, as in Japanese Patent Application Laid-Open No. 2001-208959 filed by the present applicant, when a landscape mode that is assumed to have a focal position on the infinity side is set, it is necessary to shorten the time required for focusing. Accordingly, the focus lens 101b may be moved from the infinity side to the close side.

  The photometry / ranging CPU 137 obtains this detection value and issues a command to the focus motor 111 to move the focus lens 101b to the corresponding in-focus position. The focus motor 111 moves the focus lens 101b to the in-focus position in accordance with a command from the photometry / ranging CPU 137 (AF operation).

  The photometry / ranging CPU 137 is connected to the release switch 104 through inter-CPU communication with the main CPU 20, and the in-focus position is detected when the release switch 104 is half-pressed by the user. Further, a zoom motor 110 is connected to the photometry / ranging CPU 137, and when the main CPU 20 acquires a zoom command from the user in the TELE direction or WIDE direction, the zoom motor 110 is used. By driving the motor 110, the zoom lens 101a is moved between the WIDE end and the TELE end.

  The charging / light emission control unit 138 is supplied with electric power from the power supply battery 68 in order to cause the strobe 105a to emit light, and charges a flash light emitting capacitor (not shown) or controls light emission of the strobe 105a.

  The charging / light emission control unit 138 receives various signals such as a charging start of the power supply battery 68, a half-press / full-press operation signal of the release switch 104, and a signal indicating the light emission amount and the light emission timing. Is supplied to the self-timer lamp (tally lamp) 105c and the AF auxiliary light lamp 105b, and control is performed so that a desired light emission amount can be obtained at a desired timing.

  Note that the self-timer lamp 105c may be constituted by an LED, or may be made common with the LED constituting the AF auxiliary light lamp 105b.

  A self-timer circuit 83 is connected to the main CPU 20. When the self-photographing mode is set, the main CPU 20 measures time based on the full-press signal of the release switch 104. During this timing, the main CPU 20 causes the self-timer lamp 105c to blink while gradually increasing the blinking speed in accordance with the remaining time via the photometry / ranging CPU 137. The self-timer circuit 83 inputs a timing completion signal to the main CPU 20 after timing is completed. The main CPU 20 causes the CCD 132 to perform a shutter operation based on the timing completion signal.

  The communication control unit 139 includes a communication port 107. The communication control unit 139 outputs an image signal of a subject photographed by the camera 100 to an external device such as a personal computer equipped with a USB terminal. In addition, by inputting an image signal from such an external device to the camera 100, data communication with the external device is performed. The camera 100 also has a function of simulating the function of switching to ISO sensitivity 80, 100, 200, 400, 1600, etc. of a normal camera that takes a photograph on a roll-shaped photographic film, and has an ISO sensitivity of 400 or more. When switched, the high-sensitivity mode is set in which the amplification factor of the amplifier of the white balance / γ processing unit 133 is set to a high amplification factor exceeding a predetermined amplification factor. The communication control unit 139 stops communication with the external device during shooting in the high sensitivity mode.

  The camera 100 also includes a compression / decompression & ID extraction unit 143 and an I / F unit 144. The compression / decompression & ID extraction unit 143 reads and compresses the image data stored in the buffer memory 135 via the bus line 142 and stores the image data in the memory card 200 via the I / F unit 144. In addition, the compression / decompression & ID extraction unit 143 extracts an identification number (ID) unique to the memory card 200 and reads the image data stored in the memory card 200 when reading the image data stored in the memory card 200. Are decompressed and stored in the buffer memory 135.

  The Y / C signal stored in the buffer memory 135 is compressed in accordance with a predetermined format by the compression / decompression & ID extraction unit 143, and then the removable medium such as the memory card 200 or the hard disk (HDD) 75 via the I / F 144. Are recorded in a predetermined format (for example, an Exif (Exchangeable Image File Format) file). Data recording to the hard disk (HDD) 75 or reading of data from the hard disk (HDD) 75 is controlled by the hard disk controller 74 in accordance with a command from the main CPU 20.

  The camera 100 includes a main CPU 20, an EEPROM 146, a YC / RGB conversion unit 147, and a display driver 148. The main CPU 20 controls the entire camera 100. The EEPROM 146 stores solid data and programs unique to the camera 100. The YC / RGB conversion unit 147 converts the color video signal YC generated by the YC processing unit 140 into RGB signals of three colors, and outputs them to the image display LCD 102 via the display driver 148.

  In addition, the camera 100 has a configuration in which an AC adapter 48 for obtaining power from an AC power supply and a power supply battery 68 are detachable. The power supply battery 68 is composed of a rechargeable secondary battery such as a nickel-cadmium battery, a nickel metal hydride battery, or a lithium ion battery. The power supply battery 68 may be a single-use primary battery such as a lithium battery or an alkaline battery. The power supply battery 68 is electrically connected to each circuit of the camera 100 by being loaded into a battery storage chamber (not shown).

  When the AC adapter 48 is loaded in the camera 100 and power is supplied from the AC power source to the camera 100 via the AC adapter 48, the power supply battery 68 is preferential even if it is loaded in the battery storage chamber. The power output from the AC adapter 48 is supplied to each part of the camera 100 as driving power. If the AC adapter 48 is not loaded and the power battery 68 is loaded in the battery storage chamber, the power output from the power battery 68 is supplied to each part of the camera 100 as driving power. Is done.

  Although not shown, the camera 100 is provided with a backup battery separately from the power supply battery 68 housed in the battery housing chamber. For example, a dedicated secondary battery is used as the built-in backup battery and is charged by the power supply battery 68. The backup battery supplies power to the basic functions of the camera 100 when the power battery 68 is not loaded in the battery storage chamber, such as when the power battery 68 is replaced or removed.

  That is, when the power supply from the power supply battery 68 or the AC adapter 48 is stopped, the backup battery is connected to the RTC 15 or the like by a switching circuit (not shown) and supplies power to these circuits. As a result, as long as the backup battery 29 does not reach the end of its life, power supply continues to the basic functions such as the RTC 15 without interruption.

  An RTC (Real Time Clock) 15 is a chip dedicated to timekeeping, and continuously operates by receiving power supply from the backup battery even when power supply from the power supply battery 68 or the AC adapter 48 is turned off.

  The image display LCD 102 is provided with a backlight 70 that illuminates the transmissive or transflective liquid crystal panel 71 from the back side. In the power saving mode, the main CPU 20 determines the brightness of the backlight 70 ( Brightness) is controlled through the backlight driver 72, and the power consumption of the backlight 70 is reduced. In the power saving mode, the information position designation key 126 of the operation unit 120 is pressed to display a menu screen on the image display LCD 102, and a predetermined operation can be performed on the menu screen so that on / off can be set. It has become.

  FIG. 4 shows an example of a block configuration of the face detection LSI 20b.

  FIG. 5 illustrates information output by the face detection LSI 20b as face information. The face detection LSI 20b obtains the number of face areas including the face portion of the person, the position of each face, the size of each face from the CCD RAW image that is sequentially output from the A / D conversion unit 134 and stored in the buffer memory 135. Then, the detection information of each face (probability that the detected face area is an actual face), face information including information such as the orientation of each face and the inclination of each face is detected, and the detection result is buffered. Output to the memory 135.

  FIG. 6 is a flowchart showing an outline of the photographing operation using the face detection process controlled by the CPU 20. As shown in this figure, the timing for performing the face detection process is roughly divided into two cases: a case where the face detection process is performed from a through image (S3) and a case where the face detection process is performed after the main exposure (S12). The purpose of the face detection S1AF (S9) described in this specification is to limit the AF area on the XY plane by using the result of the face detection in the through image, and to focus on surely.

  FIG. 7 is a flowchart showing an outline of the operation of the face detection S1AF.

  An outline of this operation will be described. When face detection is turned on by a predetermined operation such as a pressing operation of the face button 125 and a face is detected as a result of the face detection from the through image, another AF is detected. Operate in preference to the method.

  As a specific method of performing face detection S1AF, for example, the technique described in Japanese Patent Application No. 2006-046123 by the present applicant can be applied.

  FIG. 8 is a flowchart showing the flow of S21 “AF area setting” of the face detection S1AF.

  In S41, an arbitrary face area to be subjected to AF is selected. As a selection method, for example, the size of each face area is obtained, and the face area having the largest area is selected.

  In S42, a backlight determination process is performed. Details of this processing will be described later.

  In S43, the AF area is calculated based on the coordinates of the face area selected in S41 among the coordinates of the face area output by the face detection LSI 20b. The outline of this operation is as follows: the coordinates of the center of the AF area and the four vertices of the rectangle are calculated based on the size, orientation, and inclination of the face included in the face information detected by the face detection LSI 20b. Next, the calculated center coordinates and four vertex coordinates are converted into a coordinate system of a CCD RAW image that is a target for actually acquiring an AF evaluation value. A rectangular closed region formed by connecting the four vertices after the conversion with a straight line is defined as an AF region.

  FIG. 9 is a flowchart showing the flow of the backlight determination process, and FIG. 10 conceptually shows the main points of the backlight determination.

  In the backlight determination processing, pixels inside the periphery of the vertical face area that are opposed to the outer pixels through the face area periphery are determined from the photometric values of the pixels outside the periphery of the face area in the vertical direction detected by the face detection LSI 20b. If the difference value exceeds a predetermined threshold value, the value (Temp_deltaEV) is stored. This is repeated for each pixel in the 16-divided area of AE (S50 to S71 in FIG. 9, FIG. 10).

  Further, the maximum value (max_deltaEV) among the difference values (Temp_deltaEV) stored for each pixel is compared with a value obtained by multiplying a predetermined threshold (FACE_AF_BACKLIGHT_TH) by 10 (S72), and max_deltaEV> FACE_AF_BACKLIGHT_TH × 10. If it is not max_deltaEV> FACE_AF_BACKLIGHT_TH × 10, it is determined that there is no backlight (S74).

  Note that the photometric value of the inner pixel may be subtracted from the photometric value of the outer pixel on the peripheral edge of the face region in the horizontal direction, depending on the configuration of the image sensor and the AF detection unit.

  Further, as shown in FIG. 11, when it is determined that the backlight is backlit in the backlight determination process, the flag “1” indicating that the AF operation specialized for backlight is performed is stored in the EEPROM 146, and it is determined that the backlight is not backlit. If it is, the flag “0” indicating that the AF operation specialized for backlighting is not performed is stored in the EEPROM 146.

  FIG. 12 is a flowchart showing the flow of the AF operation including the specific AF operation corresponding to the case where the flag “1” for performing the AF operation specialized for backlighting is stored in the EEPROM 146. This process represents the detailed contents of S26 to S30 in FIG.

  In this process, first, “backlit face AF area setting”, “interrupt processing OFF setting”, and “global search OFF setting after face narrow range search” are performed.

  Here, “backlit face AF area setting” is a process of setting an AF area based on face information (the AF area calculated in S43 of FIG. 8) as a backlighting face AF area.

In addition, the “interrupt processing OFF setting” means that the AF evaluation value newly acquired by minute driving (step driving) the focus lens 101b by a certain amount (Search_Step) in a certain direction and the previous AF acquired before the minute driving. When the AF evaluation value is reduced by comparing with the evaluation value, it is determined that the AF evaluation value peak point (so-called jaspin position) has been passed (assuming that the AF search value is decreasing) and the AF search is interrupted. The setting is not performed.

  This is because the presence of the peak point of the AF evaluation value is relatively accurately determined in an image with a large amount of signal or a high-contrast image, and after that, it is preferable to interrupt a useless search (interruption process) The face area during backlighting has many noise components, and multiple local maximum points of the AF evaluation value obtained from this appear as Oyama Koyama. This is because there is a possibility that it is not necessarily the true in-focus position. If the AF search is interrupted at the peak of the mountain, accurate focusing may not be completed.

  Further, “all area search OFF setting after face narrow range search” is a setting in which the entire area search is not performed when there is no AF evaluation value peak in the narrow area search using face detection. In other words, the range in which the focus lens 101b is step-driven from the closest side (Near side) to the infinity side (INF side) is limited to a specific range based on a predetermined reference position (Pulse_Target) (toward the closest side) After searching for the peak of the AF evaluation value within a predetermined distance (Pulse Target-Search_Step * Search_Range_Near) and a predetermined distance toward the infinity side (limited to the range of Pulse Target + Search_Step * Search_Range_Inf) Even if there is no AF evaluation value peak, the entire area search is not performed. This is because there are few noise components in the face area at the time of direct light, so even if there is a peak slightly away from the predetermined reference position (Pulse Target), a certain degree of reliability is ensured, but at the time of backlighting This is because the ratio affected by information different from the original face contrast, such as noise components, is high, and the peak position away from the predetermined reference position (Pulse Target) is low in reliability.

 Note that the reference value (Search_Range_Near) for the step drive on the near side and the reference value (Search_Range_Inf) for the step drive on the infinity side can be different values. In this case, the search range on the near side (Pulse Target-Search_Step * Search_Range_Near) and the search range on the infinity side (Pulse Target + Search_Step * Search_Range_Inf) can be set separately.

  Next, an AF search range of the photographing lens 101 is set based on the AF area calculated in S43 of FIG. 8 (S101). Note that the size of the search range can be set to a value different from that during direct light. Next, the AF evaluation value is acquired by driving the photographing lens 101 within the search range (S102), and it is determined whether or not the peak of the AF evaluation value obtained from the frequency component extracted by First_Filter has been obtained. If the peak can be obtained (S103), the process proceeds to S105, and if not, the process proceeds to S104.

  Next, it is determined whether or not the peak of the AF evaluation value obtained from the frequency component extracted by Second_Filter can be obtained (S104). If the peak can be obtained, the process proceeds to S105, and if not, the process proceeds to S115. .

  Then, the position of the photographing lens 101 when the peak of the AF evaluation value is obtained in S103 or S104 is calculated, and this is set as a temporary focusing position (Temp_pint_pulse0) (S105).

  Next, the temporary in-focus position (Temp_pint_pulse0) is a predetermined reference position (Pulse Target, which is a focus position set in advance by an arbitrary method, a so-called placement pin), and a predetermined limit distance (Backlight_Range_Near) on the near side. It is determined whether it is within the range and whether it is within the range of the predetermined limit distance (Backlight_Range_Inf) on the infinity side (S106, S107), and the range of the predetermined limit distance on the near side or infinity side If not, the temporary focus position (Temp_pint_pulse0) is assumed to have low reliability as the focus position, and the process proceeds to S108. If the temporary in-focus position (Temp_pint_pulse0) is within a predetermined range from the predetermined reference position (Pulse Target) on the near side and infinity side, the temporary in-focus position is considered highly reliable as the in-focus position. The process proceeds to S115.

  The method for determining the predetermined reference position (Pulse Target) may be anything as long as it is different from the method for calculating the temporary in-focus position (Temp_pint_pulse0), and various methods are conceivable. For example, when a face area is detected, the calculation is performed based on the ratio of the face in the screen and the zoom position. Alternatively, if no face area is detected, the default position corresponding to the currently set shooting mode such as landscape mode and the zoom position or the default position during normal AF is read from the EEPROM 146, and this is used as a predetermined reference position. Good. Alternatively, the reference position corresponding to the result of the backlight determination may be read from the EEPROM 146.

  Further, the reference value of the limit distance on the near side (Backlight_Range_Near) and the reference value of the limit distance on the infinity side (Backlight_Range_Inf) can be different values. For example, in a long distance shooting scene such as landscape mode, the focus position is on the infinity side, and in a short distance shooting scene such as portrait mode (or face detection) or macro shooting mode, the focus position is on the close side. Depending on the scene, the range in which the in-focus position is assumed widens or narrows, or is biased toward the close side or the infinity side. Therefore, for example, in portrait mode and face detection, the limit value reference value (Backlight_Range_Inf) on the infinity side is made smaller than the reference value (Backlight_Range_Near) on the closest distance side, and the subject is focused behind the intended subject. (So-called rear pin) can be prevented.

  When the reference value of the limit distance on the near side (Backlight_Range_Near) and the reference value of the limit distance on the infinity side (Backlight_Range_Inf) cannot be set separately, the reference position may be changed according to the discrimination result of the shooting scene.

  In S <b> 108, it is determined whether or not a flag “1” indicating that an AF operation specialized for backlighting is performed is stored in the EEPROM 146. If the flag “1” is stored, the process proceeds to S109, and if the flag “0” is stored, the process proceeds to S114.

  S109 to 113 are AF operations specialized for backlighting. Here, the AF area is enlarged in the body direction with reference to the face area by AF area enlargement processing as disclosed in paragraph 0196 and later of Japanese Patent Application No. 2006-046123, and an area (enlarged) that can be obtained by enlargement. The area remaining after subtracting the original face area and the predetermined areas such as the neck and feet from the area is defined as a chest area, which is defined as a new AF area independent of the face area (S109), and the process proceeds to S110. However, the enlarged area itself may be a new AF area.

  FIG. 13 conceptually shows the contents of the AF area expansion processing. In this process, the AF area is sequentially enlarged by a predetermined increment in the direction from the face to the body until reaching a predetermined size in accordance with the inclination of the face included in the face information. FIG. 13 (a) shows a state in which when the tilt of the face is 0 degree, the AF area is vertically enlarged in the same downward increment to the size of the face area up to 5 times the size of the face area. Show. FIG. 13B shows that when the tilt of the face is 90 degrees, the AF area is sequentially enlarged to the left in the same size increment as the face area up to 6 times the size of the face area. Indicates. FIG. 13 (c) shows a state in which when the tilt of the face is 180 degrees, the AF area is vertically enlarged in the same size increment as the face area up to 5 times the size of the face area. Show. FIG. 13D shows that when the tilt of the face is 270 degrees, the AF area is sequentially enlarged to the right in the same size increment as the face area until it reaches 6 times the size of the face area. Indicates.

  The basics of AF using face detection are performed by spot AF that searches for a peak point by obtaining an AF evaluation value from an AF area determined based on the face area. However, in reality, there are cases where the AF evaluation value peak of the face area is small and the in-focus position cannot be obtained. In such a case, the AF area obtained using the AF area enlargement process is used.

  In S110, an AF evaluation value is obtained by setting an AF search range of the photographing lens 101 with reference to the chest region in the image, and a peak of the AF evaluation value calculated from the data extracted by First_Filter is obtained. It is determined whether or not it has been completed, and if a peak can be obtained, the process proceeds to S112, and if it cannot be obtained, the process proceeds to S111.

  In S111, an AF evaluation value is obtained by setting an AF search range of the photographing lens 101 based on the chest region range in the image, and a peak of the AF evaluation value calculated from the data extracted by Second_Filter is obtained. The process proceeds to S112 if a peak can be obtained, and to S120 if a peak cannot be obtained.

  In S112, the position of the photographing lens 101 when the peak of the AF evaluation value is obtained in S110 or S111 is calculated, and this is set as a temporary focusing position (Temp_pint_pulse2).

 In S113, it is determined whether or not the absolute value | Temp_pint_pulse0-Temp_pint_pulse2 | of the difference between the two temporary in-focus positions is less than a predetermined allowable threshold value. If the absolute value is less than the predetermined allowable threshold, the process proceeds to S114. If not, the process proceeds to S119.

  In S114, assuming that the temporary focusing position (Temp_pint_pulse0) has high reliability as the final focusing position, the photographing lens 101 is moved to the temporary focusing position (Temp_pint_pulse0) to focus.

  S115 to S118 are processes in which the processes of S109 to 112 are respectively replaced from the “face area” to the “chest area”. Alternatively, it can be replaced with “neck region”, “abdomen region”, and “foot region”.

  That is, first, in S115, it is determined whether or not a breast region has been determined by the above-described enlargement processing or the like. If the breast region can be determined, the process proceeds to S115, and if it cannot be determined, the process proceeds to S120.

  In S116, an AF evaluation value is calculated based on First_Filter data using the breast region as an AF region, and this peak point is searched. As a result, if there is a peak of the AF evaluation value, the process proceeds to S118, and if there is no peak, the process proceeds to S117.

  In S117, an AF evaluation value is acquired based on the chest region range in the image, and it is determined whether or not the peak of the AF evaluation value based on the Second_Filter data can be obtained. If it cannot be obtained, the process proceeds to S120.

  In S118, the position of the taking lens 101 when the peak of the AF evaluation value is obtained is calculated, and this is set as a temporary in-focus position (Temp_pint_pulse0).

  In S119, assuming that the temporary focusing position (Temp_pint_pulse2) has high reliability as the final focusing position, the photographing lens 101 is moved to the temporary focusing position (Temp_pint_pulse2) to focus. However, a flag indicating whether or not the temporary in-focus position (Temp_pint_pulse2) is the final in-focus position is stored in the EEPROM 146 in advance, and the temporary in-focus position (Temp_pint_pulse2) is set as the final in-focus position. The photographing lens 101 may be moved to the temporary focusing position (Temp_pint_pulse2) only when it is stored.

  In S120, assuming that both the temporary in-focus position (Temp_pint_pulse0) and the temporary in-focus position (Temp_pint_pulse2) have low reliability as the in-focus position, the photographic lens 101 is moved to a predetermined reference position (Pulse Target) and focused. Adjust.

  With the above operation, even in a backlight scene, it is possible to search for the in-focus position as much as possible by changing the AF area from the face to an area related to another human subject, or by reducing the pass frequency band of the filter. .

  FIG. 14 shows the flow of the AF operation including the AF operation corresponding to the case where the flag “0” indicating that the AF operation specialized for backlighting is not performed is stored in the EEPROM 146.

  An outline of this process is as follows: AF evaluation values are acquired by performing minute driving (step driving) of the focus lens 101b in a certain direction (S131 to S132). The drive amount of the step drive is about the focal depth, but is defined by the combination of the focal length (zoom position) and the aperture value at the time of AF search. The actual driving amounts are all the same as the values during normal AF, and fixed value parameters set in the EEPROM 146 are used.

  The determination as to whether or not to interrupt the AF search (S134) and the determination as to whether or not to perform the interruption process only for the evaluation value of the First_filter (S135) are determined by the setting value of the EEPROM 146. Is done.

  In addition, when there is no AF evaluation value peak in the narrow-area search using face detection, whether or not to perform the whole area search (S141) is also determined by the set value of the EEPROM 146. When performing a global search, the determination (S148) of whether to set the search start position to the NEAR side search limit (S149) or the INF side search limit (S150) can also be set by the setting value of the EEPROM 146. .

Second Embodiment
In S117 of the first embodiment, if the chest area is an AF area and a peak cannot be obtained, the process proceeds to S120.

  However, as shown in FIG. 15, if a peak is not obtained in the breast region (“No” in S117), the AF evaluation value is calculated again by changing the AF region, and the process is repeated to obtain the peak. Good. That is, an appropriate region other than the chest region, and a neck region corresponding to the abdomen (or a neck, a leg, or the like) that is sequentially obtained by the above-described enlargement processing is set as an AF region, or is enlarged by the above-described enlargement processing sequentially The obtained enlarged area is set as an AF area (S201), specific frequency components are extracted from the image data of the AF area using First_Filter and Second_Filter, and these are integrated to obtain an AF evaluation value. It is determined whether or not (S202, S203), and if a peak is obtained, a temporary in-focus position (Temp_pint_pulse2) is calculated (S204) and focused here (S119).

  In this way, by repeating the operation of recalculating the AF evaluation value for two different pass frequency bands after changing the AF area, the possibility of obtaining the peak of the AF evaluation value increases, and as much as possible. Accurate focus can be achieved.

  The predetermined number of times to change the AF area is arbitrary. For example, if the AF evaluation value cannot be obtained even if the AF evaluation value is repeatedly calculated again by changing the AF area from the face to the neck, the abdomen, and the foot three times, the photographing lens 101 is moved to a predetermined reference position ( Move to (Pulse Target) and focus.

<Third Embodiment>
In the first and second embodiments, the AF area is changed according to the result of the backlight determination. However, the AF area is not limited to the backlight determination, and may be switched according to the determination of various shooting scenes.

  For example, when the CPU 20 determines that the landscape mode is selected with the mode dial 123 and a face area is detected, the AF area may be enlarged at a predetermined magnification (for example, 1.2 times) centering on the image.

  Alternatively, in the ID photo mode, if the AF area is enlarged, there is a concern about the back pin, so it is better not to enlarge the AF area.

  20: Main CPU, 20b: Face detection LSI, 83: Self-timer circuit, 101: Shooting lens, 105c: Self-timer lamp, 132: CCD, 134: A / D conversion unit, 150: AF detection unit, 151: AE calculation Part

Claims (9)

A subject image is formed on a photographing element through a photographing lens, a desired frequency component is extracted from an image signal obtained from the photographing element by a filter, and a focus evaluation value is calculated by integrating the extracted frequency component. A photographing apparatus for searching for a focus position of the photographing lens based on the calculated focus evaluation value,
In the first procedure for calculating the focus evaluation value based on the signal output from the filter and the second procedure different from the first procedure, the first temporary in-focus position and the second A temporary in-focus position calculating unit for calculating a temporary in-focus position;
It is determined whether the difference between the first temporary focusing position and the second temporary focusing position is within a predetermined range, and it is determined that the difference is within the predetermined range. And a controller that moves the photographic lens to the first temporary focusing position as a final focusing position,
The temporary in-focus position calculation output unit, in response to said difference is determined to the not within a predetermined range, said first focus evaluation value calculation target region and the frequency to be extracted in the filter of the temporary in-focus position After changing at least one of the components, the first temporary focusing position is calculated again,
The control unit determines whether or not a difference between the first temporarily focused position detected again and the second temporarily focused position is within a predetermined range, and the difference is within the predetermined range. response to it is determined that the said first temporary in-focus position final focusing the photographing lens shade apparatus Taking that moves as a position. The imaging apparatus according to claim 1 , wherein the control unit repeats recalculation of the first temporary in-focus position until it is determined that the difference is within the predetermined range. The control unit moves the photographing lens with a predetermined reference position as a final in-focus position when the number of times of calculation of the first temporary in-focus position exceeds a predetermined number of 1 or more. The imaging device according to 2 . A scene discriminating unit for discriminating a shooting scene;
The control unit, based on the photographic scene is determined by the scene discrimination unit, imaging apparatus according to any one of claims 1 to 3, changing at least one of the predetermined range or the predetermined reference position. A face detection unit for detecting a face region based on the image signal;
A region determination unit that determines at least one of a neck region, a chest region, an abdominal region, and a foot region based on the face region detected by the face detection unit;
The imaging apparatus according to claim 1 , wherein the focus evaluation value calculation target region includes any one of a face region, a neck region, a chest region, an abdominal region, a foot region, and a region obtained by combining all or part thereof. . Frequency components to be extracted in the filter, imaging apparatus according to claim 1 or 5 from the predetermined first frequency component is switched to a predetermined second frequency component lower than the first frequency component.   A subject image is formed on a photographing element through a photographing lens, a desired frequency component is extracted from an image signal obtained from the photographing element by a filter, and a focus evaluation value is calculated by integrating the extracted frequency component. A photographing apparatus for searching for a focus position of the photographing lens based on the calculated focus evaluation value,
  In the first procedure for calculating the focus evaluation value based on the signal output from the filter and the second procedure different from the first procedure, the first temporary in-focus position and the second A temporary in-focus position calculating unit for calculating a temporary in-focus position;
  It is determined whether the difference between the first temporary focusing position and the second temporary focusing position is within a predetermined range, and it is determined that the difference is within the predetermined range. And a controller that moves the photographic lens to the first temporary focusing position as a final focusing position,
  The control unit moves an imaging lens with a predetermined reference position as a final in-focus position when the number of times of calculation of the first temporary in-focus position exceeds a predetermined number of 1 or more. . A subject image is formed on a photographing element through a photographing lens, a desired frequency component is extracted from an image signal obtained from the photographing element by a filter, and a focus evaluation value is calculated by integrating the extracted frequency component. A focus position search method for searching for a focus position of the photographing lens based on the calculated focus evaluation value,
In the first procedure for calculating the focus evaluation value based on the signal output from the filter and the second procedure different from the first procedure, the first temporary in-focus position and the second Calculating a temporary in-focus position;
It is determined whether the difference between the first temporary focusing position and the second temporary focusing position is within a predetermined range, and it is determined that the difference is within the predetermined range. Accordingly, the step of moving the photographic lens to the first temporary focusing position as the final focusing position;
When it is determined that the difference is not within the predetermined range, at least one of the focus evaluation value calculation target region at the first temporary focusing position and the frequency component to be extracted by the filter is changed. Calculating the first temporary focusing position again;
It is determined whether or not the difference between the first temporarily focused position detected again and the second temporarily focused position is within a predetermined range, and it is determined that the difference is within the predetermined range. In response, the step of moving the photographic lens with the first temporary focus position as the final focus position;
An in-focus position search method comprising:   A subject image is formed on a photographing element through a photographing lens, a desired frequency component is extracted from an image signal obtained from the photographing element by a filter, and a focus evaluation value is calculated by integrating the extracted frequency component. A focus position search method for searching for a focus position of the photographing lens based on the calculated focus evaluation value,
  In the first procedure for calculating the focus evaluation value based on the signal output from the filter and the second procedure different from the first procedure, the first temporary in-focus position and the second Calculating a temporary in-focus position;
  It is determined whether the difference between the first temporary focusing position and the second temporary focusing position is within a predetermined range, and it is determined that the difference is within the predetermined range. Accordingly, the step of moving the photographic lens to the first temporary focusing position as the final focusing position;
  Moving the photographic lens with a predetermined reference position as a final in-focus position in response to the number of calculations of the first temporary in-focus position exceeding a predetermined number of 1 or more;
  An in-focus position search method comprising:

Images