JP2012124769A - Imaging apparatus having subject detection means - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a system which can maintain responsibility for detecting a subject without limiting an angle of view to be detected and can detect the subject with high accuracy even when the subject becomes big.SOLUTION: An imaging apparatus has means for recording an image input from imaging means such as a digital camera and a digital video camera, or speech information input from speech input means, in a recording medium provided in the digital camera and the digital video camera, and for retrieving and playing the image and the speech information recorded in the recording medium, and for detecting the subject. The imaging apparatus further has means for pre-reading a desired pixel data from an imaging device; and means for dividing a pixel for constituting the pixel data into a first group and a second group, firstly retrieving the first group from the imaging means, and then retrieving the second group after the first group is retrieved.

Description

  The present invention records video information input from an imaging means (sensor), such as a digital camera or a digital video camera, or audio information input from an audio input means on a recording medium of the digital camera or digital video camera, The present invention relates to a method for controlling subject detection means in an imaging apparatus capable of reading and reproducing video and audio information recorded on the recording medium from the recording medium.

  In the subject detection means in the imaging apparatus, in order to improve the responsiveness of the subject detection process and to detect the subject in the photographed image at high speed, the region for subject detection in the photographed image is limited. (For example, refer to Patent Document 1).

JP 2010-010737 A

  However, in a system in which the responsiveness of subject detection is improved by limiting the region for subject detection in the captured image and reading out the pixel data in the region in advance from the imaging means (sensor). Since the angle of view is limited, there is a possibility that the subject cannot be detected when the subject covers a wide range (for example, the subject is large).

Fig. 6 (1)
As shown in FIG. 4, the region to be subject to detection is limited in the captured image, and the pixel data of the region is read out from the imaging means (sensor) in advance, so that the subject can be selected based on the pixel data read out in advance. This is an example of area division in a system in which the responsiveness of subject detection is enhanced by detection.

  In the figure, an area A is an area for subject detection, and pixel data of the area is read out in advance from the imaging means (sensor). The area B is an area other than the area A in the captured image, and after reading out the pixel data of the area A from the imaging means (sensor), the pixel data of the area is read out.

  The reading process of the pixel data from the imaging means (sensor) is completed within one cycle of the vertical synchronization signal (V).

  As shown in FIG. 6B, when the subject covers a wide range during shooting (for example, the subject becomes large), the subject cannot be detected from the image data existing in the region A (for example, the subject exists in the region A). It cannot be recognized as a face from pixel data). If the area A is enlarged to avoid this, the number of pixels constituting the area A increases, and the area for subject detection becomes larger. As a result, as shown in FIG. 6 (3), the subject detection process takes time. For example, when pixel data is read out from the imaging means (sensor) in synchronization with the vertical synchronization signal (V), the subject detection process is performed as follows. Processing time is required until the next V. That is, the subject detection process is not completed within the 1V period.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a system with high subject detection accuracy even when the subject becomes large, while maintaining the responsiveness of subject detection, without the restriction of the detection angle of view.

  In order to achieve the above object, according to the present invention, in an imaging system capable of reading out desired image data from an imaging means (sensor) in advance, the pixels constituting the image data are classified into a first group and a second group. In this case, the first group is read first from the image pickup means, and the subject is detected based on the image data composed of the first group of pixels read first.

The image data recorded on the recording medium is generated by combining the outputs of the first and second groups.
The pixels included in the first group are discretely arranged, and the second group is arranged at other positions.

  Further, it is assumed that the first group includes only pixels of a specific color, or the first group includes a higher percentage of pixels of a specific color than the second group.

  According to the present invention, even when the subject covers a wide range (the subject is large), the detection angle of view is not limited, and detection can be reliably performed even when the subject becomes large while maintaining detection response (detection accuracy is improved). Can be increased).

Configuration diagram of an imaging apparatus for explaining the first and second embodiments Timing diagram for explaining the first and second embodiments Timing diagram for explaining the first and second embodiments Processing sequence explanatory diagram of the first embodiment Timing chart for explaining the second embodiment Processing sequence explanatory diagram of the second embodiment Illustration of conventional example Illustration of conventional example Illustration of conventional example

[Example 1]
FIG. 1 shows a configuration diagram of an imaging apparatus such as a digital camera or a digital video camera according to the first embodiment of the present invention.

  In the figure, 101 is a focus lens, 102 is a solid-state image sensor such as a CMOS, 103 is a TG (Timing Generate) unit that performs read control of the solid-state image sensor 102 such as a CMOS, and 104 is read from the solid-state image sensor 102. A / D converter (sensor interface) that converts the received signal into a digital signal, 105 is a camera signal processor that processes the signal from the solid-state imaging device converted into a digital signal and generates image data, and 106 is image data. Is an IP conversion unit that converts an interlace signal to a progressive signal for generating a high-definition still image, 107 is a moving image resizing unit for converting the image data into a moving image recording angle, and 108 is a moving image resizing unit. NR (Noise Reduction) processing for the resized image at 107 Reference numeral 109 denotes a moving image encoding unit that encodes image data resized to the angle of view of a moving image, and 110 records a video signal and an audio information signal from a microphone (not shown) as audio information input means. A recording signal processing unit for converting into a format, further decoding audio data and sending it to a speaker (not shown) as an audio signal. Reference numeral 111 denotes a recording medium for recording video information and audio information.

  112 is a still image encoding unit that performs encoding processing such as JPEG, 113 is a still image recording unit such as an SD card that stores still image data encoded by still image encoding means such as JPEG, 114 Is a main memory such as an SDRAM that temporarily stores data processed by various signal processing units and functional blocks such as an encoding unit, and the functional blocks share and use the main memory 114. The signal processing data and the image data are transferred via the main memory 114. 115 is an arbiter unit that controls access requests from each functional block to the main memory 114 and arbitrates access. 116 divides a field of view angle made up of pixel data read from the solid-state image sensor 102 such as a CMOS. An area control unit 117 detects a subject being photographed and detects a movement of the subject, a motion / subject detection unit 118, and a system control unit 118 controls the imaging apparatus system.

  FIG. 2A is an explanatory diagram of the timing for reading out pixel data from the image sensor 102 in the imaging apparatus having such a processing system. In this configuration, pixel data is read from the image sensor 102 in synchronization with the vertical synchronization signal (V), and all pixel data is read from the image sensor 102 such as a CMOS over a period of 1V.

  In this configuration, the image sensor 102 is a primary color single-plate Bayer-type CMOS sensor, and the pixels (all pixel data) for configuring image data are divided into a first group and a second group, and the first group first. The second group is read after reading from the image sensor 102 and reading the first group.

  The pixels included in the first group are discretely arranged, and the second group is arranged at other positions.

  Further, it is assumed that the first group includes only pixels of a specific color, or the first group includes a higher percentage of pixels of a specific color than the second group.

  In this configuration, the specific color is G1 data.

  FIG. 2B shows pixel data read from the image sensor 102.

  Reading out the pixel data from the image sensor 102 reads out only the G1 data in advance in the first group, that is, all the pixel data. Thereafter, R, G2, and B data other than G1 are read out from the image sensor 102 in the second group, that is, all pixel data.

  Both the first group of pixel data (G1) and the second group of pixel data (R, G2, B) are stored in the DRAM 114 via the arbiter 115.

  The pixel data (G1) of the first group and the pixel data (R, G2, B) of the second group are combined by the camera signal processing unit 105, and the combined image data is used as image data to be recorded on the recording medium. .

The first group of pixel data (G1) read out in advance is also input to the subject detection unit 117, and the subject detection unit 117 detects the subject based on the G1 data.
A flowchart of this processing sequence is shown in FIG.

After the image pickup apparatus is turned on (S301), when a recording instruction is issued from the photographer (S302), the first group is read first from the image sensor 102 (S303), the first group is read, and then the second group is read. Read (S305). The first group of pixel data is input to the subject detection unit 117, and the subject is detected (S308). This sequence continues during recording.
Since the subject is detected based on the pixel data read out from the image sensor 102 in advance as described above, the subject detection is performed in comparison with the conventional configuration (FIG. 6) in which the area is divided and the pixel data is sequentially read from the image sensor. This makes it possible to detect the subject for the entire captured image (all the angle of view) while maintaining the responsiveness of the image, eliminates the restriction of the detected angle of view, and reliably detects the subject even if the subject becomes large.

[Example 2]
Another embodiment in this configuration will be described below.

  The configuration of the imaging apparatus in the present embodiment is also the same as that shown in FIG.

  The image sensor 102 is a primary color single-plate Bayer array CMOS sensor, and pixels (all pixel data) for configuring image data are divided into a first group and a second group, and read from the image sensor 102 first from the first group. After reading the first group, the second group is read.

  The pixels included in the first group are discretely arranged, and the second group is arranged at other positions.

  Further, it is assumed that the first group includes only pixels of a specific color, or the first group includes a higher percentage of pixels of a specific color than the second group.

  In this configuration, the specific color is G1 data.

  FIG. 2B shows pixel data read from the image sensor 102.

  Reading out the pixel data from the image sensor 102 reads out only the G1 data in advance in the first group, that is, all the pixel data. Thereafter, R, G2, and B data other than G1 are read out from the image sensor 102 in the second group, that is, all pixel data.

  Both the first group of pixel data (G1) and the second group of pixel data (R, G2, B) are stored in the DRAM 114 via the arbiter 115.

  The pixel data (G1) of the first group and the pixel data (R, G2, B) of the second group are combined by the camera signal processing unit 105 and used as image data to be recorded on the recording medium.

  The first group of pixel data (G1) read out in advance is also input to the subject detection unit 117, and the subject detection unit 117 detects the subject based on the G1 data.

  As described above, since the subject is detected based on the pixel data read out from the image sensor 102 in advance, the subject detection is performed in comparison with the conventional configuration in which the area shown in FIG. 6 is divided and the pixel data is sequentially read from the image sensor. , And the subject can be detected for the entire captured image (all angles of view), the detection angle of view is not limited, and the subject can be reliably detected even if the subject becomes large.

However, the first group of pixel data (G1) is discretely arranged in all the pixel data.

  That is, it is equivalent to reading out pixels from the image sensor 102 and the resolution of the image data used to detect the subject is lowered. When the image sensor 102 is a primary color single-plate Bayer CMOS sensor as in this configuration, the first group of pixel data (G1) read out from the image sensor 102 in advance is thinned to ¼.

  When the subject is small, the resolution of the subject detection image used for subject detection processing decreases as the resolution of the pixel data read in advance decreases, and as a result, the accuracy of subject detection may decrease. There is sex.

  Therefore, in order to suppress a decrease in the accuracy of subject detection, the image sensor 102 is adaptively controlled as necessary, and pixel data read from the image sensor 102 is switched.

  FIG. 4 shows a timing chart of the sequence in this configuration. After the image pickup apparatus is powered on, the area control unit 116 instructs the TG 103 to select pixel data to be read from the image sensor 102.

  Immediately after turning on the power of the imaging device, it is not known where the subject to be photographed is within the shooting angle of view, so the first group of pixel data (G1) is read out from the imaging device 102 in advance, and the first group of pixel data ( After reading G1) in advance, the second group of pixel data (R, G2, B) is read. By detecting a subject based on the first group of pixel data (G1) read out in advance, the subject detection is performed over the entire captured image (full angle of view). At this time, the position in the captured image with respect to the detected subject is stored in the subject detection unit 117. Here, the information to be stored is stored as the coordinates of the area including the position in the captured image with respect to the subject, and is stored as the area A as coordinate information.

  Here, based on the coordinate information of the area A, the system control unit 118 determines whether the area A is an area of 1/4 or less of the total angle of view or an area larger than 1/4.

  When it is determined that the area A is an area equal to or less than ¼ of the total angle of view, the object position information (coordinate information of the area A) stored in the object detection unit 117 is transmitted through the system control unit 118 to the area control unit. The area control unit 116 controls the TB 103 based on the subject position information, and reads the pixel data of the area A based on the subject position information from the image sensor 102 in advance. After reading the pixel data of the area A, the pixel data of the area B which is an area other than the area A is read.

  If it is determined that the area A is an area larger than ¼ of the total angle of view, the first group of pixel data (G1) is advanced from the image sensor 102 to the area control unit 116 through the system control unit 118. And give instructions to read. Upon receiving the instruction, the area control unit 116 controls the TB 103 to read out the first group of pixel data (G1) in advance.

  After the pixel data (G1) of the first group is read out in advance, the pixel data (R, G2, B) of the second group is read out. By detecting a subject based on the first group of pixel data (G1) read out in advance, the subject detection is performed over the entire captured image (full angle of view).

  A flowchart of this processing sequence is shown in FIG.

  After the image pickup apparatus is powered on (S501), when a recording instruction is issued from the photographer (S502), the first group is read first from the image sensor 102 (S503), the first group is read, and then the second group is read. Read (S505). The pixel data of the first group is input to the subject detection unit 117, the subject is detected (S508), and the result is stored (S509). The result is stored as the coordinates of the area including the subject. It is determined whether the area is an area of 1/4 or less of the total angle of view or an area larger than 1/4 (S510).

  When it is determined that the region is a region having a quarter or more of the total angle of view, the pixel of the first group is also used when reading pixel data from the image sensor 102 in synchronization with the next vertical synchronization signal (V). Read data in advance.

  When it is determined that the region is a region equal to or less than ¼ of the total angle of view, when reading out pixel data from the image sensor 102 in synchronization with the next vertical synchronization signal (V), the region is read from the image sensor 102. Read in advance (S511). After reading out the area, an area other than the area is read out (S513).

  This sequence continues during recording.

  As described above, the size of the subject is determined based on the subject position information, and when the subject size is smaller than a quarter of the total angle of view, the region is divided and the pixel data of the divided region is captured. If the size of the subject is larger than ¼ of the total angle of view, the pixel data resolution is less affected when the subject size is larger than an area of ¼, so the specific color data (G1) is preceded by the image sensor. By reading and detecting the subject based on the pixel data of the specific color read in advance, the responsiveness of subject detection can be maintained regardless of the size of the subject, and depending on the size of the subject By adaptively switching control over the image sensor and changing pixel data read from the image sensor, the subject detection angle of view is not restricted, and the subject can be detected reliably.

DESCRIPTION OF SYMBOLS 101: Focus lens 102: Image pick-up element 103: TG part 104: A / D conversion part 105: Camera signal processing part 106: IP conversion part 107: Moving image resizing part 108: NR processing part 109: Movie encoding part 110: Recording Signal processing unit 111: recording medium 112: still image encoding unit 113: still image recording medium 114: main memory 115: memory control unit 116: area control unit 117: motion / subject detection unit 118: system control unit

Claims (8)

The video information input from the imaging unit or the audio information input from the audio input unit, such as a digital camera or digital video camera, is recorded on the recording medium of the digital camera or digital video camera, and recorded on the recording medium. Pixels for configuring image data having means for pre-reading pixel data from the image sensor in an imaging apparatus capable of reading and reproducing video and audio information from a recording medium and further having means for detecting a subject Is divided into a first group and a second group, the first group is read from the imaging unit first, the first group is read, and then the second group is read. apparatus. 2. The imaging apparatus having subject detection means according to claim 1, wherein the pixels included in the first group are discretely arranged, and the second group is arranged at other positions. The first group includes only specific color pixels, or the first group includes a higher percentage of specific color pixels than the second group. An imaging apparatus comprising the subject detection means according to Item 2. 4. The subject detection unit according to claim 1, wherein the image data recorded on the recording medium is generated by combining the pixel data of the first and second groups. 5. Imaging device. 5. The apparatus according to claim 1, wherein after the imaging apparatus is powered on, the subject is read from the imaging unit in advance from the first group, and the subject is detected based on an image generated from the pixel data of the first group. An imaging apparatus comprising the subject detection unit according to any one of the preceding claims. Means for adaptively switching pixel data read from the imaging means in accordance with the size of the detected subject, and the size of the detected subject is based on an area including 1 / N pixels of the total number of pixels. 6. An imaging apparatus having subject detection means according to claim 1, further comprising means for comparing the sizes of detected subjects. 2. The apparatus according to claim 1, further comprising a unit that reads out the first group of pixel data from the imaging unit in advance when the size of the detected subject is larger than an area including 1 / N of the total number of pixels. An imaging apparatus comprising the subject detection unit according to claim 6. A means for dividing the photographed image into a plurality of areas, and when the size of the detected subject is smaller than an area including the number of pixels of 1 / N of the total number of pixels, a specific area among the divided areas 8. An imaging apparatus having subject detection means according to claim 1, further comprising means for reading out the pixel data from the imaging means.