JP2010093689A - Image pickup device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To carry out imaging with a right imaging quality, even if no face can be detected in a video movie equipped with a face detecting function. <P>SOLUTION: An image pickup device includes: an image pickup section 1 which images an object; a face detection unit 3 which detects the face in an image; and a controller 4 which adjusts a brightness and a white balance corresponding to face data detected in the face detection unit 3. The controller 4 stops the adjustment of the brightness and the white balance corresponding to the face data when the face being detected by the face detection unit 3 moves out of an imaging region. When no face can be detected within the imaging region, the adjustment is continued with the brightness and the white balance corresponding to the face data detected before. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an imaging apparatus, and more particularly to an imaging apparatus equipped with a face detection means.

Some recent imaging apparatuses are equipped with a face detection function to perform focus and white balance processing so that the detected face is photographed beautifully. For example, face detection means for detecting the face position, face position recording means for recording the face position and the time when the image was taken, white balance adjustment means for changing white balance processing according to the face detection result, and photometry area A face detection result obtained within a predetermined time even when no face is detected, having an exposure control means for setting the position and a focus area setting means for setting the focus area to the face position By controlling the shooting operation on the basis of the image quality, for example, the image quality is prevented from changing even when the subject faces backward (see, for example, Patent Document 1).
JP 2006-157617 A (page 1-5, FIG. 1)

  However, when the above configuration is applied to a video movie or the like having a moving image shooting function, for example, even when a person and the background are recorded and the subject person leaves the screen, the shooting conditions remain unchanged. If, on the other hand, the face becomes smaller due to zooming out and the face cannot be detected, the exposure condition changes after a certain period of time, even though the exposure condition remains the same. There was a problem that it would end up.

  In view of the above-described problems, the present invention is to provide an imaging apparatus that always performs imaging under preferable imaging conditions.

  An imaging apparatus according to the present invention detects an imaging unit that images a subject, an image processing unit that performs image processing including white balance adjustment on the output of the imaging unit, and a specific subject in the image of the image processing unit A feature detection unit, an image processing adjustment unit that adjusts image processing in the image processing unit according to the specific subject detected by the feature detection unit, and a specific subject that is being detected by the feature detection unit are outside the imaging region. When the movement is stopped, the adjustment of the image processing corresponding to the specific subject is stopped, and when the specific subject cannot be detected in the imaging region, the adjustment of the image processing corresponding to the specific subject detected before is adjusted. And a control unit that continues.

  The image pickup apparatus of the present invention can take a video with the same image quality without changing the shooting conditions when the person who becomes the subject zooms out or turns sideways. If you go out of the corner, you can take a picture with a quality that matches the scenery.

(Embodiment 1)
FIG. 1 is a block diagram showing Embodiment 1 of the imaging apparatus of the present invention.

  The imaging unit 1 includes an optical system, an imaging element, and an AD converter. The optical system includes an objective lens, a zoom lens, a diaphragm, an OIS (Optical Image Stabilizer) unit, a focus lens, and the like. The optical system collects light from the subject and forms a subject image. The imaging element captures a subject image formed by the optical system and generates image data. The image sensor is a CMOS image sensor, a CCD image sensor, or the like. The AD converter converts image data generated by the image sensor into a digital signal. The image memory 2 holds the image data output by the imaging unit 1. The face detection unit 3 detects whether or not a face exists in the image data on the image memory 2. When a face is detected, the face detection unit 3 notifies the controller 4 of the position information and size information.

  In the present embodiment, the face in the image data is detected. However, the present invention is not limited to this. For example, an entire person, a specific animal such as a cat, a moving object, or the like may be detected. . In this embodiment, the face detection unit 3 is provided separately from the controller 4, but may be realized by the controller 4.

  The control unit 5 controls the lens, the diaphragm, the OIS unit, the focus, and the like constituting the imaging unit 1 based on the output from the controller 4. The controller 4 is a control unit that controls the entire imaging apparatus. The controller 4 may be configured only by hardware, or may be realized by combining hardware and software. The controller 4 can be realized by a microcomputer. Based on the position and size notified from the face detection unit 3, the controller 4 focuses the position and sends a command to the control unit 5 to control the aperture so that the brightness at the position is optimized. Output. Further, based on the detected position information and size information of the face, a command for correcting the white balance is output to the white balance correction unit 7 so that the portion has the best hue.

  The white balance correction unit 7 performs white balance correction on the image data from the image memory 2 based on an instruction from the controller 4. The image processing unit 8 includes an image compression unit, an image expansion unit, a system encoder, and a system decoder.

  The image compression unit compresses image data by performing DCT (Discrete Cosine Transform), Huffman coding, and the like. The image compression unit is, for example, MPEG-2 or H.264. The image data is compressed by a compression format conforming to the H.264 standard. However, the present invention relates to MPEG-2 or H.264. The image data is not limited to H.264 format. The image decompression unit decodes the image data into an uncompressed state when reproducing compressed image data output from the system decoder.

  The system decoder decodes the AV data recorded in the recording means 9 and generates compressed image data. The system decoder outputs the generated compressed image data to the image expansion unit. The recording unit 9 records and reproduces image data on a recording medium. The recording medium is a memory card, a hard disk, a DVD (Digital Versatile Disc), or the like.

  A case where there is a face in the image data will be described with reference to a flowchart shown in FIG.

  The controller 4 has a unique flag called a face detection flag, and is OFF, which is the initial state when shooting starts. When video shooting is started (step 20) (at this time, the face detection flag is OFF as an initial state), it is determined whether the face is detected by the face detection unit 3 (step 21). If no face is detected, the determination is false and the process proceeds to step 26. Since the face detection flag is still OFF, the routine proceeds to step 31 where normal automatic exposure (AE) processing is performed. The processing in this routine is performed until face detection by the face detection unit 3 is performed.

  When a face is detected by the face detection unit 3, the face is detected and its position information is notified. The determination in step 21 becomes true, and the process proceeds to step 22 where the face detection flag is turned on. Thereafter, the position information where the face is detected is stored (step 23). The controller 4 generates a control signal to correct the white balance and brightness so that the place where the face is present is displayed most clearly, and stores the correction amount (step 25). As long as the face is continuously detected by the face detection unit 3, this routine is performed.

  Next, processing when the face detection unit 3 stops face detection will be described. If face detection is not performed by the face detection unit 3, the determination in step 21 is false and the process proceeds to step 26. Since the face detection flag is still ON, the process proceeds to step 27. In this step, based on the position information of the face detected in the past, it is determined whether the face has gone out of the screen or is suddenly not detected for some reason.

  If it is determined that the image has gone out of the screen, no special correction by the face is required, so the face detection flag that is ON in step 30 is turned off, and the process proceeds to step 31 to perform normal automatic exposure. Processing is performed, and correction by face detection is reset. Therefore, when the determination in step 21 is performed next, naturally no face detection is performed, so the process proceeds to step 26, and the face detection flag is OFF, so the normal automatic exposure in step 31 is performed.

  If it cannot be determined that the subject has gone out of the screen, for example, when the subject turns back and the head should still be in the screen, the process proceeds to steps 28 and 29, and the stored brightness adjustment amount Alternatively, the white balance correction amount is read out, and brightness adjustment and white balance adjustment are performed based on the readout values. That is, the image quality adjustment is performed so that the image quality is the same as the state in which the face was detected immediately before.

  Note that the determination method of step 27 is to check the change in the detected face position information and move toward the outside of the screen. It is possible to determine that “we have gone out of”.

  By performing such processing, for example, even when the subject person faces sideways or the face is not detected because the face sweat is wiped with a handkerchief, the shooting conditions remain unchanged. You can take pictures with high image quality. This is particularly effective when the face size is below the detection limit due to zoom-out or the like, since it is not desired to change the image quality.

  On the other hand, if the subject person goes out of the field of view, for example, if backlight correction is performed, it will be too bright if the adjustment amount is used as it is, and early correction adjustment is necessary. Become. In such a case, the correction by the face detection is reset and the normal automatic exposure processing is performed, so that the correct automatic exposure can always be performed and the image can be shot with the image quality that matches the subject to be shot. is there.

In the present embodiment, audio is not described, but it goes without saying that an audio signal may be compressed and recorded on a recording medium by the recording means 9 together with the compressed moving image, as in a normal video movie. Nor.
(Embodiment 2)
FIG. 3 shows a case where the face detection unit 3 no longer detects a face, and the processing in the case where it cannot be determined that the face detection unit 3 has gone out of the screen is further enhanced. Even if the face is suddenly no longer detected, it can be assumed that the subject goes out of the screen while facing backwards. In such a case, it is better to return to normal automatic exposure. For this reason, in the present embodiment, step 35 is added after step 29.

  That is, the brightness of the entire screen is measured, and if there is not much change, the brightness and white balance are corrected based on the correction amount stored as before. If the brightness changes greatly and the period of the change is short, for example, there may be a case where flash photography is performed at someone else's place. In this case, it is more appropriate to estimate that the subject is still in the screen, and the image quality adjustment with the stored correction amount is performed as usual. Conversely, if the luminance change continues for a certain time or more, it is considered that the subject has moved out of the screen, so it is determined YES in step 36, the face detection flag is turned off in step 31, and normal automatic exposure is performed. It will be.

  In this example, the luminance of the entire screen is measured in step 35, but the amount of pixels that are whiteout (or blackened) may be measured and determined by the degree of change in the amount. . For example, when there is almost no whiteout (or blackout), when the amount of whiteout (or blackout) exceeds a certain ratio (for example, 1/5) of the screen, on the contrary, Alternatively, when there are many black spots), it may be determined that “there is a change” when the increase ratio exceeds a certain amount. Incidentally, the “original” amount may be the last corrected image in which the face is detected by the face detection unit 3.

  The imaging apparatus according to the present invention can be applied not only to moving image shooting but also to a device for shooting still images such as a digital camera.

1 is a block diagram illustrating a configuration of an imaging apparatus according to Embodiment 1 of the present invention. The flowchart which shows the sequence of the process in Embodiment 1 of this invention. The flowchart which shows the sequence of the process in Embodiment 2 of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Imaging part 2 Image memory 3 Face detection part 4 Controller 5 Control part 6 Memory 7 White balance correction part 8 Image processing part 9 Recording means

Claims (4)

An imaging unit for imaging a subject;
An image processing unit that executes image processing including white balance adjustment for the output of the imaging unit;
A feature detection unit for detecting a specific subject in the image of the image processing unit;
An image processing adjustment unit that adjusts image processing in the image processing unit in accordance with the specific subject detected by the feature detection unit;
When the specific subject being detected by the feature detection unit moves outside the imaging area, the adjustment of the image processing according to the specific subject is stopped, and the specific subject cannot be detected within the imaging area. A control unit for continuing adjustment of image processing according to the specific subject detected before;
An imaging apparatus comprising: Provided with a luminance detection means for detecting the average luminance of the video,
The controller is
Even when the specific subject being detected by the feature detection unit cannot be detected within the imaging region, if the average luminance changes by the luminance detection means, image processing adjustment according to the specific subject is performed. The imaging device according to claim 1. The imaging device according to claim 2, wherein when the change in the average luminance does not continue for a certain period of time, the control unit continues adjustment of image processing according to the specific subject detected before. The imaging device according to claim 1, wherein the feature detection unit detects a face in an output image of the image processing unit.

Images