JP2013198013A - Imaging apparatus, imaging control method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect an imaged object region properly even in an environment such as a backlight state and the like.SOLUTION: An imaging apparatus comprises the steps of: detection determination means for determining based on a plurality of discriminators whether a subject region of an image sequentially obtained by an image obtaining means can be detected or not; correction means for conducting correction processing against data of the sequentially obtained image after the detection determination means determines that the subject region cannot be detected; first similarity calculation means for calculating first similarity between an image region for the image whose data has had the data correction processing by the correction means and the discriminators; second similarity calculation means for calculating similarity between an image region for an image whose data has not have correction processing by the correction means and the discriminators; similarity comparison means for comparing the first similarity and the second similarity; and cancellation determination means for determining whether the correction processing by the correction means is to be cancelled or not, based on a comparison result by the similarity comparison means.

Description

  The present invention relates to an imaging device that detects a subject, an imaging control method, and a program.

As a conventional technique, there is a technique for increasing the detection accuracy of a subject area (face area) even when an exposure state such as backlight is not appropriate when detecting a face in an imaging apparatus (see, for example, Patent Document 1). ).
Specifically, face detection is performed after gain adjustment processing is performed on image data captured during backlighting.

JP 2008-108024 A

  However, when the technique disclosed in Patent Document 1 is applied in a backlight state, automatic exposure control (AE) is performed on a live view image (a real-time output image from an image sensor) subjected to a backlight gain adjustment process. Act. As a result, the amount of gain adjustment in the backlight state becomes excessive, and there is a possibility that subsequent detection of the subject area cannot be performed properly.

  The present invention has been made in view of such a situation, and an object thereof is to more appropriately detect a subject area even under an environment in a backlight state or the like.

In order to achieve the above object, an imaging device of one embodiment of the present invention includes:
Image acquisition means for sequentially acquiring images, discriminator storage means for storing a plurality of types of discriminators for detecting a subject area in images sequentially acquired by the image acquisition means, and storage by the discriminator storage means A detection determination unit that determines whether or not a subject region of an image sequentially acquired by the image acquisition unit can be detected based on the plurality of types of discriminators, and the subject region cannot be detected by the detection determination unit. When the determination is made, a correction unit that corrects the data of the sequentially acquired images, and a first similarity that calculates a first similarity between the region of the image whose data is corrected by the correction unit and the classifier Similarity calculation means, second similarity calculation means for calculating the similarity between the image area whose data is not corrected by the correction means and the discriminator, and the first similarity calculation Based on the comparison result by the similarity comparison means for comparing the first similarity calculated by the means and the second similarity calculated by the second similarity calculation means, and the comparison result by the similarity comparison means And cancellation determination means for determining whether or not to cancel the correction processing by the correction means.

  According to the present invention, it is possible to more appropriately detect a subject area even under an environment in a backlight state or the like.

It is a block diagram which shows the structure of the hardware of the imaging device which concerns on one Embodiment of this invention. It is a functional block diagram which shows the functional structure for performing a gamma correction switching process. It is a flowchart which shows the flow of the gamma correction switching process which an imaging device performs. It is a figure which shows the state transition of the control in an imaging device. It is a figure which shows the state transition of the comparative example of control in an imaging device. It is a figure which shows the comparison of operation | movement when control is performed according to the state transition diagram of FIG.4 and FIG.5.

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

FIG. 1 is a block diagram showing a hardware configuration of an imaging apparatus according to an embodiment of the present invention.
The imaging device 1 is configured as a digital camera, for example.

  The imaging apparatus 1 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a bus 14, an input / output interface 15, an imaging unit 16, and an input unit 17. An output unit 18, a storage unit 19, a communication unit 20, and a drive 21.

  The CPU 11 executes various processes according to a program recorded in the ROM 12 such as a gamma correction switching process program (described later) or a program loaded from the storage unit 19 to the RAM 13.

  The RAM 13 appropriately stores data necessary for the CPU 11 to execute various processes.

  The CPU 11, ROM 12, and RAM 13 are connected to each other via a bus 14. An input / output interface 15 is also connected to the bus 14. An imaging unit 16, an input unit 17, an output unit 18, a storage unit 19, a communication unit 20, and a drive 21 are connected to the input / output interface 15.

  Although not shown, the imaging unit 16 includes an optical lens unit and an image sensor.

The optical lens unit is configured by a lens that collects light, for example, a focus lens or a zoom lens, in order to photograph a subject.
The focus lens is a lens that forms a subject image on the light receiving surface of the image sensor. The zoom lens is a lens that freely changes the focal length within a certain range.
The optical lens unit is also provided with a peripheral circuit for adjusting setting parameters such as focus, exposure, and white balance as necessary.

The image sensor includes a photoelectric conversion element, AFE (Analog Front End), and the like.
The photoelectric conversion element is composed of, for example, a CMOS (Complementary Metal Oxide Semiconductor) type photoelectric conversion element or the like. A subject image is incident on the photoelectric conversion element from the optical lens unit. Therefore, the photoelectric conversion element photoelectrically converts (captures) the subject image, accumulates the image signal for a predetermined time, and sequentially supplies the accumulated image signal as an analog signal to the AFE.

The AFE performs various signal processing such as A / D (Analog / Digital) conversion processing on the analog image signal. Through various signal processing, a digital signal is generated and output as an output signal of the imaging unit 16.
Such data represented by the output signal of the imaging unit 16 is hereinafter referred to as “captured image data”. Data of the captured image is appropriately supplied to the CPU 11, RAM 13, and the like.

The input unit 17 includes various buttons and the like, and inputs various types of information according to user instruction operations.
The output unit 18 includes a display, a speaker, and the like, and outputs images and sounds.
The storage unit 19 is composed of a hard disk, DRAM (Dynamic Random Access Memory), or the like, and stores various data such as model data representing a face (hereinafter referred to as “discriminator” as appropriate).
The communication unit 20 controls communication performed with other devices (not shown) via a network including the Internet.

  A removable medium 31 composed of a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is appropriately attached to the drive 21. The program read from the removable medium 31 by the drive 21 is installed in the storage unit 19 as necessary. The removable medium 31 can also store various data such as image data stored in the storage unit 19 in the same manner as the storage unit 19.

FIG. 2 is a functional block diagram illustrating a functional configuration for executing the gamma correction switching process among the functional configurations of the imaging apparatus 1.
The gamma correction switching process is a series of processes for controlling whether to perform gamma correction on image data for face detection.

The imaging unit 16 includes an image acquisition unit 41 that acquires captured image data of a subject.
The image acquisition unit 41 acquires captured image data of the subject, and outputs the acquired captured image data to the image generation unit 42.

  At this time, the image acquisition unit 41 outputs live view image data, focus-locked image data, and captured image data according to the pressed state of the shutter button of the imaging apparatus 1. Specifically, when the shutter button is not pressed, the image acquisition unit 41 continues the imaging operation, and the captured image data sequentially acquired while the imaging operation is continued is used as live view image data. Output. When the shutter button is half-pressed (a state where the shutter button is pressed to a predetermined position that does not reach the lower limit), the image acquisition unit 41 continues the imaging operation while maintaining the focal length when the shutter button is half-pressed. Then, while the imaging operation is continued while the shutter button is half-pressed, sequentially acquired captured image data is output as focus-locked image data. Further, when the shutter button is fully pressed (down to the lower limit), the image acquisition unit 41 determines the focal length at a predetermined position (focal length when focus is locked or autofocus control) at the current angle of view. The focal length), and the captured image data of one frame (or a plurality of frames with different shutter speeds) is output as captured image data.

The CPU 11 includes an image generation unit 42, a face detection unit 43, a face frame setting unit 44, a gamma correction instruction unit 45, and a reaction value determination unit 46.
The image generation unit 42 controls the image acquisition unit 41 to receive the live view image data as image data for face detection.

  In addition, the image generation unit 42 includes a gamma correction processing unit 42 a and is referred to as an instruction signal (hereinafter referred to as “gamma correction instruction signal”) indicating whether or not to perform gamma correction input from the gamma correction instruction unit 45. ), Gamma correction is performed on the data of the live view image input from the image acquisition unit 41. Specifically, when the gamma correction instruction signal indicates that gamma correction is not performed, the live view image data input from the image acquisition unit 41 is output to the face detection unit 43. If the gamma correction instruction signal indicates that gamma correction is to be performed, gamma correction is performed on the live view image data input from the image acquisition unit 41, and the gamma correction live view image data is subjected to face detection. Output to the unit 43 and the reaction value determination unit 46.

The face detection unit 43 performs face detection processing on the live view image data input from the image generation unit 42 using the classifier 47.
Specifically, the face detection unit 43 identifies a plurality of types of data obtained by modeling the face at different angles (the position of the viewpoint with respect to the face) with respect to the live view image data input from the image generation unit 42. A classifier (hereinafter referred to as “sub-classifier”) is read from the classifier 47. Then, the face detection unit 43 calculates an evaluation value (score) indicating the degree of similarity between the face model data of each sub-classifier and the detection target region in the live view image using, for example, an adaboost algorithm. . Furthermore, the face detection unit 43 detects an area having an evaluation value equal to or greater than a threshold value as a face area candidate, and calculates the number of detected face area candidates as a reaction value for each sub-classifier. A sub classifier in which a face area candidate is detected is referred to as a sub classifier that reacts to a detection target area of a live view image.

The face detection unit 43 identifies a sub classifier having the highest response value by detecting a face region in a live view image using a plurality of types of sub classifiers. At this time, the face detection unit 43 determines that the face area cannot be detected when the response value of the identified sub-classifier is equal to or less than the set threshold value, and performs gamma correction of the live view image. An instruction signal (gamma correction activation signal) for output is output to the gamma correction instruction unit 45.
It should be noted that an evaluation value (score calculated by the Adaboost algorithm) of an area detected as a face area candidate in the selected sub-classifier can be used as the response value of the sub-classifier.

Then, when detecting the face area, the face detection unit 43 outputs data for specifying the detected face area to the face frame setting unit 44.
The face frame setting unit 44 determines the size and position of a frame (hereinafter referred to as “face frame”) for identifying the detected face region based on the data specifying the face region input from the face detection unit 43. To decide. Then, the face frame setting unit 44 outputs data representing the determined size and position of the face frame to the face frame display unit 48.

  When the gamma correction activation signal is input from the face detection unit 43, the gamma correction instruction unit 45 outputs a gamma correction instruction signal indicating that gamma correction is performed to the image generation unit 42. Further, when a gamma correction cancellation signal is input from the reaction value determination unit 46, the gamma correction instruction unit 45 outputs a gamma correction instruction signal indicating that gamma correction is not performed to the image generation unit 42.

  The reaction value determination unit 46 acquires a reaction value (hereinafter, appropriately referred to as “correction reaction value”) as a result of applying the discriminator 47 by the face detection unit 43 to the gamma-corrected live view image. Then, the reaction value determination unit 46 calls a reaction value (hereinafter, referred to as “uncorrected reaction value” as appropriate) as a result of applying the discriminator 47 to the captured image data that has not been subjected to gamma correction for the live view image. .) Then, the reaction value determination unit 46 compares the corrected reaction value with the uncorrected reaction value, and when the uncorrected reaction value is larger than the corrected reaction value, an instruction signal for preventing gamma correction (gamma correction). Cancel signal) is output to the gamma correction instruction unit 45.

  In one area of the storage unit 19, a classifier 47, which is model data representing a face, is stored. The discriminator 47 has, for example, a plurality of types of sub discriminators that are data obtained by modeling a face at different angles (viewpoint positions with respect to the face). Each sub-classifier is composed of different types of model data representing facialness, such as face model data viewed from the front, face model data viewed from the left, face model data viewed from the upper right. The Then, the face detection target region in the live view image and the model data of each sub classifier are matched (that is, a score indicating similarity is calculated), thereby calculating the reaction value of each sub classifier. Is done. In the face detection unit 43, face detection is performed using this reaction value.

The output unit 18 includes a face frame display unit 48 for displaying a face frame on the live view image. The face frame display unit 48 displays the face frame superimposed on the live view image based on the data representing the size and position of the face frame input from the face frame setting unit 44.
In the present embodiment, when face detection cannot be performed due to backlight or the like for a live view image for face detection, control is performed to perform gamma correction on the data of the live view image. Then, when face detection by a specific classifier succeeds in a state where gamma correction is performed, face detection is performed by the specific classifier even in a state where gamma correction is not performed, and a state where gamma correction is performed and a state where it is not performed Compare the response values of the specific classifiers. Then, when the reaction value in the state where the gamma correction is not performed is larger than that in the state where the gamma correction is performed, control for turning off the gamma correction is performed.
Here, in the present embodiment, the specific discriminator refers to a sub discriminator having the highest response value in a state where gamma correction is performed.

  As a result, even in a situation where the face area is dark and control is performed to increase the brightness of the captured image by automatic exposure control, face detection cannot be performed due to the gamma correction effect of live view image data for face detection. Can be prevented. Further, since the gamma correction is switched on / off according to the change in brightness of the live view image, it is possible to suppress a situation in which face detection cannot be performed when the backlight state changes to the direct light state thereafter. That is, it is possible to more appropriately perform the face detection process in the backlight state or the follow light state without being affected by the automatic exposure control.

[Operation]
Next, the operation of the imaging apparatus 1 will be described.
[Gamma correction switching processing]
FIG. 3 is a flowchart illustrating a flow of gamma correction switching processing executed by the imaging apparatus 1.

The gamma correction switching process is executed in response to the start of imaging of the live view image in the imaging apparatus 1. In the gamma correction switching process, the process is performed for each frame of the live view image.
When the gamma correction switching process is started, the gamma correction instruction unit 45 outputs a gamma correction instruction signal indicating that gamma correction is not performed to the image generation unit 42 in step S1. As a result, the image generation unit 42 is in a state of outputting a live view image without performing gamma correction.

In step S <b> 2, the face detection unit 43 performs face detection processing on the live view image data input from the image generation unit 42.
In step S3, the face detection unit 43 determines whether or not a face area is detected.
If a face area is detected in the live view image input from the image generation unit 42, it is determined YES in step S3, and the process proceeds to step S4.
On the other hand, when the face area is not detected in the live view image data for face detection input from the image generation unit 42, NO is determined in step S3, and the process proceeds to step S5.

In step S <b> 4, the face frame setting unit 44 sets a face frame in the face area detected by the face detection unit 43, and displays the face frame by the face frame display unit 48.
After such a process of step S4, the gamma correction switching process is repeated.
In step S5, the face detection unit 43 determines whether or not the processing loop in which no face area is detected has been repeated M (M is a predetermined number set in advance). That is, it is determined whether or not the face area has not been detected M times in succession. This makes it possible to switch whether or not to perform gamma correction in order to stably detect the face area. In the present embodiment, all sub-classifiers are applied to one frame of live view images, and all model data are matched in one frame.

If the process loop in which the face area is not detected is not repeated M times, it is determined as NO in Step S5, and the process proceeds to Step S2.
On the other hand, when the processing loop in which no face area is detected is repeated M times, YES is determined in step S5, and the process proceeds to step S6.
In step S <b> 6, the face detection unit 43 outputs a gamma correction activation signal to the gamma correction instruction unit 45. As a result, the image generation unit 42 is in a state of outputting live view image data for face detection to the face detection unit 43 by performing gamma correction.

In step S <b> 7, the face detection unit 43 performs face detection processing on the live view image data for face detection input from the image generation unit 42.
In step S8, the face detection unit 43 determines whether or not a face area is detected.
If no face area is detected in the data of the live view image input from the image generation unit 42, NO is determined in step S8, and the process proceeds to step S9.

On the other hand, if a face area is detected in the live view image data input from the image generation unit 42, YES is determined in step S8, and the process proceeds to step S10.
In step S <b> 9, the face detection unit 43 determines whether or not the processing loop in which no face area is detected is repeated N (N is a predetermined number set in advance). That is, it is determined whether or not the face area has not been detected N times consecutively. This makes it possible to switch whether or not to perform gamma correction in order to stably detect the face area. As in step S5, all the sub-classifiers are applied to one frame of the live view image, and all model data are matched in one frame.

If the processing loop in which the face area is not detected is not repeated N times, NO is determined in step S9, and the process proceeds to step S7.
On the other hand, if the processing loop in which no face area is detected is repeated N times, YES is determined in step S9, and the process proceeds to step S1.
In step S <b> 10, the face frame setting unit 44 sets a face frame in the face area detected by the face detection unit 43, and displays the face frame by the face frame display unit 48.

In step S <b> 11, the reaction value determination unit 46 corrects the corrected response value that is the result of applying the discriminator 47 to the image data that has been subjected to the gamma correction and the data of the live view image that has not been subjected to the gamma correction. Get the reaction value.
In step S12, the reaction value determination unit 46 determines whether or not the uncorrected reaction value is larger than the corrected reaction value. That is, in step S12, which of the detection target areas in the live view image data not subjected to gamma correction and the live view image data subjected to gamma correction is more similar to the model data of the sub classifier. Is determined.

If the uncorrected response value is larger than the corrected response value, YES is determined in step S12, and the process proceeds to step S1.
On the other hand, when the corrected response value is equal to or less than the uncorrected response value, NO is determined in step S12, and the process proceeds to step S7.
As described above, when a face area is detected in a state where gamma correction is performed on live view image data for face detection, the uncorrected response value is larger than the corrected response value (that is, When the gamma correction is not performed when the similarity with the face model data is higher, the gamma correction can be quickly canceled in a situation where the gamma correction is not necessary.
Thereby, even if the shooting environment is switched from the backlit state to the following lighted state, face detection processing corresponding to the shooting environment can be performed quickly.

[State transition diagram]
Next, the control state transition realized in the imaging apparatus 1 as a result of the control based on the gamma correction switching process will be described.
FIG. 4 is a diagram illustrating control state transitions in the imaging apparatus 1. FIG. 5 is a diagram illustrating state transition of a comparative example of control in the imaging apparatus 1. The state transition diagram shown in FIG. 5 is based on control for switching whether or not to perform gamma correction only when face detection cannot be performed, and is different in that the processing of steps S10 to S12 in the present invention is not performed. .

  As illustrated in FIG. 4, the imaging apparatus 1 performs a state A1 in which gamma correction is not performed on live view image data and a state A2 in which gamma correction is performed on live view image data according to the following transition conditions C1 to C7. Transition.

(Conditions for transition from state A1 to state A1)
Transition condition C1: When a face area is detected in state A1 (when YES is determined in step S3)
Transition condition C2: When the processing loop in which no face area is detected is not repeated M times (when NO is determined in step S5)

(Conditions for transition from state A1 to state A2)
Transition condition C3: When a processing loop in which no face area is detected in state A1 is repeated M times (when YES is determined in step S5)
(Conditions for transition from state A2 to state A2)
Transition condition C4: When a face area is detected in state A2 (when YES is determined in step S8)
Transition condition C5: When a processing loop in which no face area is detected is not repeated N times (when NO is determined in step S9)

(Conditions for transition from state A2 to state A1)
Transition condition C6: When a processing loop in which no face area is detected in state A2 is repeated N times (when YES is determined in step S9)
Transition condition C7: When an uncorrected response value obtained by applying live view image data without gamma correction to the sub-classifier having the highest corrected response value in the state A2 is larger than the corrected response value ( (If YES is determined in step S12)
On the other hand, in the comparative example shown in FIG. 5, although it is common in that the state is changed according to the transition point between the state A1 and the state A2 and the transition conditions C1 to C6, the transition from the state A2 to the state A1 according to the transition condition C7. This is different from the control state transition in the imaging apparatus 1 in that there is no such thing.

[Specific operation]
FIG. 6 is a diagram showing a comparison of operations when control is performed according to the state transition diagrams of FIGS. 4 and 5. In FIG. 6, the upper body including the human face portion is the subject, and the density of the oblique lines in the subject indicates the brightness of the image (the darker the density is).
In FIG. 6A, if the imaging condition is backlight, and in this case, no gamma correction is performed, the brightness of the face area is insufficient and the face area is not detected.
In FIG. 6B, gamma correction is performed on the data of the live view image, and as a result, a face area is detected in the live view image.

In FIG. 6C, gamma correction is performed on the data of the live view image, and a face area is detected in the live view image. On the other hand, the image is acquired from the image acquisition unit 41 by automatic exposure control in the imaging unit 16. The live view image input to the generation unit 42 is gradually brightened.
Here, when control is performed according to the state transition diagram of the comparative example shown in FIG. 5, as shown in FIG. 6D, gamma is added to the live view image data for face detection until no face area can be detected. The correction is continuously performed.

On the other hand, when the control is performed according to the state transition diagram of the present embodiment shown in FIG. 4, the face area can be detected without performing gamma correction due to the effect of the automatic exposure control as shown in FIG. At this point, gamma correction is not performed.
Therefore, even when the live view image output from the imaging unit 16 changes more brightly, such as when the imaging condition changes from backlight to forward light, the face area can be detected more reliably.

  As described above, the imaging apparatus 1 according to the present embodiment includes the gamma correction instruction unit 45 and the reaction value determination unit 46, and the face area is detected in a state where the gamma correction is performed on the live view image. In this case, the reaction value determination unit 46 acquires an uncorrected response value in a live view image that is not subjected to gamma correction. Then, the reaction value determination unit 46 compares the corrected reaction value and the uncorrected reaction value in the live view image on which the gamma correction is performed, and when the uncorrected reaction value is larger than the corrected reaction value, the gamma correction instruction unit 45. Output a gamma correction cancel signal.

Then, the gamma correction instruction unit 45 outputs a gamma correction instruction signal indicating that gamma correction is not performed to the image generation unit 42, and the image generation unit 42 stops gamma correction for the live view image.
Therefore, it is possible to prevent the gamma correction effect from becoming excessive in the live view image data for face detection even in a situation where the face area is dark and control is performed to increase the brightness of the captured image by automatic exposure control. In addition, it is possible to suppress a situation in which face detection cannot be performed in a subsequent live view image due to a change in the shooting environment from the backlighting state to the backlighting state.
Therefore, according to the imaging device 1 according to the present embodiment, it is possible to more appropriately perform face detection processing in a backlight state or a follow light state.

[Modification 1]
In step S3 and step S8 of the gamma correction switching process in the above embodiment, all sub-classifiers are applied to one frame of live view images, and all model data are matched in one frame.
On the other hand, in one frame of the live view image, a response value is obtained by applying a part of the sub-classifier (for example, 3 out of all 9), and the processing loop is repeated a plurality of times. Can be applied to live view image data for face detection.

For example, if it is the process of step S3, all the sub discriminators are applied by the process loop of M times determined in step S5, or if it is the process of step S8, N times determined in step S9. All sub-classifiers can be applied before the processing loop.
In this case, since the processing capability required to apply the sub classifier to the live view image can be reduced, the number of operation clocks of the CPU 11 can be reduced.
That is, the face detection process can be executed with lower power consumption.

[Modification 2]
In step S12 of the gamma correction switching process in the above embodiment, when the uncorrected response value is larger than the corrected response value, control is performed so as not to perform gamma correction on the live view image data for face detection. As explained.
On the other hand, in step S12, when the uncorrected response value is greater than the set threshold value Th than the corrected response value, it is determined that gamma correction is not performed on the live view image data for face detection. Can do.

  As a result, the condition for switching to a state where gamma correction is not performed should be set when the response value is definitely larger without gamma correction (the uncorrected response value is definitely greater than the corrected response value). Can do. Further, by appropriately changing the threshold Th, it is possible to control the ease of changing the state to switch to a state in which gamma correction is not performed, so that it is possible to detect the face area more appropriately.

The threshold Th can be determined according to the state of automatic exposure control in the imaging unit 16. For example, when the difference between the actual measurement values with respect to the exposure target value exceeds the set range, the threshold value Th is set larger, and when the difference between the actual measurement values with respect to the exposure target value is within the set range. The threshold value Th can be set smaller.
As a result, when the brightness of the surroundings varies greatly, it is difficult to switch from a state where gamma correction is performed to a state where gamma correction is not performed, and a situation in which switching of gamma correction frequently occurs can be suppressed.

As described above, in the imaging device 1 according to an aspect of the present invention, the image acquisition unit 41 sequentially acquires live view images, and the identifier 47 includes the live view images sequentially acquired by the image acquisition unit 41. A plurality of types of classifiers for detecting a subject area (face area) are configured. The face detection unit 43 determines whether or not the subject region of the live view image sequentially acquired by the image acquisition unit 41 can be detected based on a plurality of types of discriminators, and the gamma correction processing unit 42a If it is determined by 43 that the subject area cannot be detected, the data of the live view images acquired sequentially is gamma corrected. Further, the face detection unit 43 calculates a first similarity between the region of the live view image whose data has been gamma corrected by the gamma correction processing unit 42a and the discriminator 47, and the reaction value determination unit 46 performs gamma correction processing. The second similarity between the region of the live view image whose data is not gamma corrected and the discriminator 47 is calculated by the unit 42a. The reaction value determination unit 46 compares the first similarity and the second similarity, and determines whether or not to cancel the gamma correction by the gamma correction processing unit 42a based on the comparison result.
Therefore, it is possible to more appropriately perform the face detection process in the backlight state or the follow light state.

If the reaction value determination unit 46 determines that the gamma correction is to be canceled, the gamma correction instruction unit 45 causes the face detection unit 43 to detect the subject area for the live view image for which the gamma correction has been canceled.
Thereby, after the situation where gamma correction is unnecessary, the subject area can be detected for the live view image that has not been gamma corrected.

The reaction value determination unit 46 cancels the gamma correction performed by the gamma correction processing unit 42a when the second similarity is larger than the first similarity.
Accordingly, even in a situation where the brightness of the captured image is controlled by automatic exposure control, it is possible to prevent the gamma correction effect from becoming excessive in the live view image data for detecting the subject area.

The reaction value determination unit 46 cancels the gamma correction by the gamma correction processing unit 42a when the second similarity is larger than the first similarity by the threshold Th or more.
As a result, the condition for switching to a state where no gamma correction is performed can be set when the degree of similarity is surely greater without gamma correction. Further, by appropriately changing the threshold Th, it is possible to control the ease of changing the state to switch to a state in which gamma correction is not performed, so that it is possible to detect the subject area more appropriately.
The discriminator 47 is face model data, and the face detection unit 43 determines whether the face area can be detected as a subject area based on the similarity between the face model data and the area of the live view image. .
This makes it possible to more appropriately detect the face area in the subject even in an environment in a backlight state or the like.

  In addition, this invention is not limited to the above-mentioned embodiment, The deformation | transformation in the range which can achieve the objective of this invention, improvement, etc. are included in this invention.

  For example, in the above-described embodiment, the case of switching whether to apply gamma correction to live view image data for face detection has been described as an example. Other correction methods can be used. For example, it may be switched whether or not to perform gain processing (brightness amplification processing) that changes the brightness of the entire live view image.

In the above-described embodiment, the imaging apparatus 1 to which the present invention is applied has been described using a digital camera as an example, but is not particularly limited thereto.
For example, the present invention can be applied to general electronic devices having an imaging function. Specifically, for example, the present invention can be applied to a notebook personal computer, a video camera, a mobile phone, a Web camera, a portable game machine, and the like.

The series of processes described above can be executed by hardware or can be executed by software.
In other words, the functional configuration of FIG. 2 is merely an example and is not particularly limited. That is, it is sufficient that the imaging apparatus 1 has a function capable of executing the above-described series of processing as a whole, and what functional blocks are used to realize this function is not particularly limited to the example of FIG.
In addition, one functional block may be constituted by hardware alone, software alone, or a combination thereof.

When a series of processing is executed by software, a program constituting the software is installed on a computer or the like from a network or a recording medium.
The computer may be a computer incorporated in dedicated hardware. The computer may be a computer capable of executing various functions by installing various programs, for example, a general-purpose personal computer.

  The recording medium including such a program is not only constituted by the removable medium 31 of FIG. 1 distributed separately from the apparatus main body in order to provide the program to the user, but also in a state of being incorporated in the apparatus main body in advance. The recording medium etc. provided in The removable medium 31 is composed of, for example, a magnetic disk (including a floppy disk), an optical disk, a magneto-optical disk, or the like. The optical disk is composed of, for example, a CD-ROM (Compact Disk-Read Only Memory), a DVD (Digital Versatile Disk), or the like. The magneto-optical disk is configured by an MD (Mini-Disk) or the like. In addition, the recording medium provided to the user in a state of being preliminarily incorporated in the apparatus main body includes, for example, the ROM 12 in FIG. 1 in which a program is recorded, the hard disk included in the storage unit 19 in FIG.

In the present specification, the step of describing the program recorded on the recording medium is not limited to the processing performed in chronological order according to the order, but is not necessarily performed in chronological order, either in parallel or individually. The process to be executed is also included.
Further, in the present specification, the term “system” means an overall apparatus constituted by a plurality of devices, a plurality of means, and the like.

  As mentioned above, although several embodiment of this invention was described, these embodiment is only an illustration and does not limit the technical scope of this invention. The present invention can take other various embodiments, and various modifications such as omission and replacement can be made without departing from the gist of the present invention. These embodiments and modifications thereof are included in the scope and gist of the invention described in this specification and the like, and are included in the invention described in the claims and the equivalent scope thereof.

The invention described in the scope of claims at the beginning of the filing of the present application will be appended.
[Appendix 1]
Image acquisition means for sequentially acquiring images;
Discriminator storage means for storing a plurality of types of discriminators for detecting a subject area in an image sequentially acquired by the image acquisition means;
A detection determination unit that determines whether or not a subject region of images sequentially acquired by the image acquisition unit can be detected based on a plurality of types of classifiers stored by the classifier storage unit;
A correction unit that corrects the sequentially acquired image data when the detection determination unit determines that the subject area cannot be detected;
First similarity calculation means for calculating a first similarity between the area of the image whose data has been corrected by the correction means and the discriminator;
Second similarity calculation means for calculating the similarity between the area of the image whose data is not corrected by the correction means and the classifier;
Similarity comparing means for comparing the first similarity calculated by the first similarity calculating means with the second similarity calculated by the second similarity calculating means;
Cancellation determination means for determining whether to cancel the correction processing by the correction means based on the comparison result by the similarity comparison means;
An imaging apparatus comprising:
[Appendix 2]
When the cancellation determination unit determines that the correction process is to be canceled, the detection determination unit further includes a detection control unit that causes the detection determination unit to detect the subject area with respect to the image for which the correction process has been canceled. The imaging apparatus according to appendix 1.
[Appendix 3]
The imaging apparatus according to appendix 1 or 2, wherein the cancellation determination unit cancels the correction processing by the correction unit when the second similarity is greater than the first similarity. .
[Appendix 4]
The supplementary determination means 1 or 2, wherein the cancellation determination unit cancels the correction processing by the correction unit when the second similarity is greater than a set value by a larger value than the first similarity. Imaging device.
[Appendix 5]
The classifier is facial model data,
Any one of appendices 1 to 4, wherein the detection determination unit determines whether or not a face region can be detected as the subject region based on a similarity between a face model data and the image region. The imaging device according to one.
[Appendix 6]
An image acquisition step for sequentially acquiring images;
A detection determination step for determining whether or not a subject region of images sequentially acquired by the image acquisition step can be detected based on a plurality of types of discriminators stored by a predetermined storage unit;
When it is determined that the subject area cannot be detected by the detection determination step, a correction step for correcting the sequentially acquired image data; and
A first similarity calculation step for calculating a first similarity between the region of the image whose data has been corrected in the correction step and the discriminator;
A second similarity calculation step of calculating a similarity between the area of the image in which the data in the correction step is not corrected and the classifier;
A similarity comparison step for comparing the first similarity calculated in the first similarity calculation step with the second similarity calculated in the second similarity calculation step;
A release determination step for determining whether or not to cancel the correction process by the correction step based on the comparison result in the similarity comparison step;
An imaging control method comprising:
[Appendix 7]
Computer
Image acquisition means for sequentially acquiring images;
Detection determination means for determining whether or not a subject region of images sequentially acquired by the image acquisition means can be detected based on a plurality of types of classifiers stored by a predetermined storage means;
A correction unit that corrects the sequentially acquired image data when it is determined by the detection determination unit that a subject area cannot be detected;
First similarity calculating means for calculating a first similarity between the area of the image whose data has been corrected by the correcting means and the discriminator;
Second similarity calculating means for calculating the similarity between the area of the image whose data is not corrected by the correcting means and the discriminator;
Similarity comparing means for comparing the first similarity calculated by the first similarity calculating means with the second similarity calculated by the second similarity calculating means;
Cancellation determination means for determining whether to cancel the correction processing by the correction means based on the comparison result by the similarity comparison means;
A program characterized by functioning as

  DESCRIPTION OF SYMBOLS 1 ... Imaging device, 11 ... CPU, 12 ... ROM, 13 ... RAM, 14 ... Bus, 15 ... Input-output interface, 16 ... Imaging part, 17 ... Input unit 18 ... Output unit 19 ... Storage unit 20 ... Communication unit 21 ... Drive 31 ... Removable media 41 ... Image acquisition unit 42 ... Image Generation unit 42a ... gamma correction processing unit 43 ... face detection unit 44 ... face frame setting unit 45 ... gamma correction instruction unit 46 ... reaction value determination unit 47 .Identifier, 48 ... face frame display

Claims (7)

Image acquisition means for sequentially acquiring images;
Discriminator storage means for storing a plurality of types of discriminators for detecting a subject area in an image sequentially acquired by the image acquisition means;
A detection determination unit that determines whether or not a subject region of images sequentially acquired by the image acquisition unit can be detected based on a plurality of types of classifiers stored by the classifier storage unit;
A correction unit that corrects the sequentially acquired image data when the detection determination unit determines that the subject area cannot be detected;
First similarity calculation means for calculating a first similarity between the area of the image whose data has been corrected by the correction means and the discriminator;
Second similarity calculation means for calculating the similarity between the area of the image whose data is not corrected by the correction means and the classifier;
Similarity comparing means for comparing the first similarity calculated by the first similarity calculating means with the second similarity calculated by the second similarity calculating means;
Cancellation determination means for determining whether to cancel the correction processing by the correction means based on the comparison result by the similarity comparison means;
An imaging apparatus comprising:   When the cancellation determination unit determines that the correction process is to be canceled, the detection determination unit further includes a detection control unit that causes the detection determination unit to detect the subject area with respect to the image for which the correction process has been canceled. The imaging device according to claim 1.   3. The imaging according to claim 1, wherein the cancellation determination unit cancels the correction processing by the correction unit when the second similarity is larger than the first similarity. 4. apparatus.   The cancellation determination unit cancels the correction process by the correction unit when the second similarity is greater than a set value by more than the first similarity. The imaging device described. The classifier is facial model data,
5. The method according to claim 1, wherein the detection determination unit determines whether or not a face region can be detected as the subject region based on a similarity between a face model data and the image region. The imaging apparatus of Claim 1. An image acquisition step for sequentially acquiring images;
A detection determination step for determining whether or not a subject region of images sequentially acquired by the image acquisition step can be detected based on a plurality of types of discriminators stored by a predetermined storage unit;
When it is determined that the subject area cannot be detected by the detection determination step, a correction step for correcting the sequentially acquired image data; and
A first similarity calculation step for calculating a first similarity between the region of the image whose data has been corrected in the correction step and the discriminator;
A second similarity calculation step of calculating a similarity between the area of the image in which the data in the correction step is not corrected and the classifier;
A similarity comparison step for comparing the first similarity calculated in the first similarity calculation step with the second similarity calculated in the second similarity calculation step;
A release determination step for determining whether or not to cancel the correction process by the correction step based on the comparison result in the similarity comparison step;
An imaging control method comprising: Computer
Image acquisition means for sequentially acquiring images;
Detection determination means for determining whether or not a subject region of images sequentially acquired by the image acquisition means can be detected based on a plurality of types of classifiers stored by a predetermined storage means;
A correction unit that corrects the sequentially acquired image data when it is determined by the detection determination unit that a subject area cannot be detected;
First similarity calculating means for calculating a first similarity between the area of the image whose data has been corrected by the correcting means and the discriminator;
Second similarity calculating means for calculating the similarity between the area of the image whose data is not corrected by the correcting means and the discriminator;
Similarity comparing means for comparing the first similarity calculated by the first similarity calculating means with the second similarity calculated by the second similarity calculating means;
Cancellation determination means for determining whether to cancel the correction processing by the correction means based on the comparison result by the similarity comparison means;
A program characterized by functioning as

Images