JP4502992B2 - Imaging device - Google Patents

Description

The present invention relates to a photographing apparatus that focuses on a person such as a digital camera. More specifically, the present invention relates to a photographing apparatus that recognizes a person in the photographing range and focuses on the person when detecting sound.

  When photographing using a photographing device such as a camera provided in a digital camera or a portable terminal (portable communication terminal such as a mobile phone, PHS, or PDA) (hereinafter collectively referred to as “photographing device”), The act of focusing is important for taking pictures clearly.

  Whether to focus manually or using the autofocus function of the camera. However, since it is generally difficult to perform manually, an autofocus function is often used in a normal case in order to reduce the burden on the photographer.

  In general, when the autofocus function is used, it is possible to focus on the subject by capturing the subject in the center of the viewfinder. In other words, the subject in the center of the viewfinder is in focus (see FIG. 4).

  In order to reduce the burden on the photographer, photographing apparatuses that perform photographing control such as a shutter by voice are disclosed in Patent Document 1 and Patent Document 2 below. Further, Japanese Patent Application Laid-Open No. 2003-228561 discloses a photographing system that detects a sound level collected by sound collecting means such as a microphone and controls the recording timing of a photographed image based on fluctuations in the sound level.

JP 2005-24792 A JP 58-137828 A JP 2005-159558 A

  Although it is possible to reduce the burden on the photographer by using the invention disclosed in each of the above-mentioned patent documents, even if these inventions are used, if the autofocus function is used, it is reflected in the center portion of the viewfinder. The subject will be in focus.

  However, depending on the composition at the time of shooting, for example, there are cases where it is desired to shoot a person and a building together. In such a case, the human is not necessarily located at the center of the viewfinder. For this reason, if an image is captured using the autofocus function as it is, the focus is adjusted to the subject located at the center of the viewfinder, so that the person may not be in focus and the person may be blurred.

  In order to prevent this, it is necessary to use a function called focus lock. The focus lock is a function for fixing the focus that has been adjusted once. When shooting using this focus lock, first move the camera so that the person who is the subject is in the center of the viewfinder, and then perform the specified operation (typically half-press the shutter). By doing this, the person is focused, and then the camera is moved again to the composition to be photographed, and then the shutter is pressed to focus on the person who is not located at the center and shoot. By using such a focus lock, it is possible to perform photographing while focusing on a person who is not located at the center.

  By using such a focus lock, it is possible to focus on a subject not located at the center, which is a very convenient function. However, for each subject to be focused, it is necessary to perform a predetermined operation such as half-pressing the shutter by moving the camera so that the subject is located at the center of the viewfinder, which is troublesome. It is difficult for children and elderly people to perform such complicated operations.

  In view of this, the present inventor noticed that when a person is photographed, the photographer often makes a call to the subject person while holding the photographing apparatus, and uses the focus lock function as described above. We have invented an imaging device that can automatically focus on a person and take a picture.

The invention of claim 1 is an imaging apparatus having a focus adjustment function by an autofocus method and a sound collection device, wherein the imaging apparatus is specified by a photographer at the time of shooting issued to a person who is a subject . A voice detection unit that detects a speech by using the sound collecting device and detects whether the photographer has pressed the shutter, and a human detection. When the image is captured by the image capturing apparatus, a human recognition unit that recognizes a human being who is a subject other than the photographer, and a focus that is used to focus a portion recognized as a human in the image by the focus adjustment function. And a processing unit.

By using a photographing apparatus such as the present invention, anyone can easily focus on a person and take a picture without using a conventional focus lock function. In addition, since it is not necessary to move the camera over and over like focus lock, shooting can be performed quickly. In general, when a person is photographed, the photographer often calls out to the person who is the subject. For this reason, the present invention is implemented by using the shout as a trigger.

In the invention according to claim 2 , the voice detection unit is a photographing apparatus that detects a voice of the photographer when a voiceprint registered in advance matches a voiceprint in the voice.

  In addition to simply detecting sound with the photographing apparatus, it may be possible to recognize the voice of a specific photographer.

The invention of claim 3 is a photographing processing method in a photographing apparatus having an autofocus type focus adjustment function and a sound collecting device, and the photographing apparatus is issued by a photographer to a person who is a subject. Detecting a voice that is a specific shout at the time of shooting using the sound collecting device, and detecting whether or not the photographer has pressed the shutter to determine whether or not it is a human shot and is a human shot Is determined , the human being who is the subject other than the photographer is recognized from the image photographed by the photographing device, and the portion recognized as the human in the image is focused by the focus adjustment function, This is a photographing processing method for photographing with a photographing device.

The invention described in claim 1 can also be configured as in the present invention. In other words, anyone can easily focus on a person and take a picture without using a conventional focus lock function. In addition, it is not necessary to move the camera over and over like focus lock, so shooting can be performed quickly. In general, when a person is photographed, the photographer often calls out to the person who is the subject. For this reason, the process of the present invention is started using the shout as a trigger.

By using the photographing apparatus of the present invention, even if a person is not located at the center portion of the viewfinder, it is possible to easily take a picture while focusing on the person. This makes it easy for a child or an elderly person to take a clear picture in focus on a human being. Also, since the operation of moving the camera such as focus lock is not required, the time required for shooting is shortened.

  A conceptual diagram of a system configuration of the photographing apparatus 1 of the present invention is shown in FIG. The photographing apparatus 1 includes at least an arithmetic device 2 such as a CPU and a sound collecting device 3 such as a microphone. The photographing apparatus 1 also has a focus adjustment function using an autofocus method. This focus adjustment function may be a commonly used one.

  The photographing apparatus 1 of the present invention may further include a storage device such as a RAM or a hard disk for storing information. Each function (each means) of the present invention is executed when a means (program, module, etc.) for executing the process is read into the arithmetic unit 2. When each function uses the information stored in the storage device in the process, each function reads the corresponding information from the storage device, and uses the read information in the processing in the arithmetic device 2 as appropriate.

  Each means in the present invention is only logically distinguished in function, and may be physically or practically the same area.

  The sound collecting device 3 is a device that collects sound outside the photographing device 1 such as a microphone, and is preferably a device that can detect the voice of the photographer.

  The arithmetic device 2 is loaded with a program, a module, or the like that realizes the voice detection unit 4, the human recognition unit 5, and the focus processing unit 6, and performs the processing of the present invention.

  The voice detection unit 4 detects the voice of the photographer with the sound collecting device 3. Although the photographer's voice may be simply detected, the photographer's specific utterance voice, for example, “high, cheese!” May be detected. At this time, the voice print of the photographer can be stored in advance, and the voice of only the photographer registered in advance can be detected by comparing with the data. In this way, it is determined that a person is to be photographed by detecting sound and detecting that the shutter is pressed.

  When the voice recognition unit 4 determines that the human photographing is performed, the human recognizing unit 5 recognizes the person to be focused from the image photographed by the camera lens. Various methods can be used to recognize a person from an image. For example, by extracting a “human face” from an image or by extracting a “human outline” There are methods of recognizing and recognizing by “skin color”. For example, in order to extract a human face from an image, by determining the positional relationship, distance relationship, color, etc. of characteristic points of the human face such as nose, eyes, mouth, eyebrows called feature points, A human face can be extracted from the image. As for the contour, the contour line of the object shown in the image is extracted, and the contour is close to a pre-stored contour or a contour having human contour characteristics (for example, face, arm, torso, foot). It may be recognized as a human being by extracting a contour). There may be a plurality of people in the image. In such a case, multiple humans may be recognized, or the person closest to the center, the person who appears most magnified, or the person who can be determined to be facing the front is recognized as the subject person. May be.

  The focus processing unit 6 performs a process of focusing on a part recognized as a human in the image recognized by the human recognition unit 5 using a focus adjustment function using an autofocus method. There are two types of auto focus methods: an active method and a passive method, and a passive method includes a phase difference detection method and a contrast detection method. Of these, the contrast detection method is most frequently used in digital cameras. This is a method of focusing from the state of contrast in an image taken by a camera. Therefore, for example, when the imaging apparatus 1 uses a contrast detection method, the human focus is recognized from the human contrast state recognized by the human recognition unit 5. Even when a method of focusing other than this is used or when used in combination, a process of focusing on a portion recognized as a person may be performed. When the active method is adopted, infrared rays or ultrasonic waves are irradiated. Therefore, an instruction to irradiate infrared rays, ultrasonic waves, or the like is given to these irradiation mechanisms for a person who has recognized the image. The irradiation direction and the irradiation angle can be calculated by calculating how much the person is displaced from the center point of the image.

  When a plurality of persons can be recognized as described above, each person may be focused.

  Next, an example of a processing process of photographing using the photographing apparatus 1 of the present invention will be described with reference to the flowchart of FIG. 2 and the conceptual diagram of FIG.

  When photographing a person using the photographing apparatus 1 of the present invention, the photographer first determines the composition. Then, when shooting, a call such as “high, cheese!” Is made to a human subject (S100). This is a natural operation that does not cause any sense of incongruity from the surroundings because it is generally performed even when performing normal human photography. In addition, when the photographer points the lens of the photographing apparatus 1 toward the subject, an image photographed by the lens, such as the subject, can be acquired through the lens of the camera. This image can be confirmed with a display device such as a camera finder or a liquid crystal display device.

  Then, the voice of the photographer is detected by the sound collecting device 3 of the photographing device 1. The sound detected by the sound collection device 3 is detected by the sound detection unit 4 (S110) and contains a specific keyword (such as “high, cheese”) stored in advance, or the voiceprint of the detected sound data is stored in advance. It is determined whether a photographer registered in advance is taking a picture by performing a process such as whether or not the voice print matches. Instead of making such a determination, it is also possible to simply collect the uttered voice with the sound collecting device 3 and detect it with the voice detecting unit 4.

  The camera recognizes that shooting is performed, for example, by pressing the shutter of the camera simultaneously with or following the call. In other words, the camera determines that the person is to be photographed by detecting the voice of the photographer and the pressing of the shutter by the voice detection unit 4, and the subsequent processing in the human recognition unit 5 is executed.

  If the voice detection unit 4 detects that the photographer is photographing, a person is recognized from the image photographed by the camera (S120). Various methods can be used for human recognition, but it can be recognized by extracting “human face”, by extracting “human contour”, or by “skin color” A recognition method can be used.

  When a person is recognized from the image photographed by the camera in this way, the focus processing unit 6 performs a process of focusing on a portion recognized as the person in the image using a focus adjustment function by an autofocus method. Perform (S130).

  Shooting is then performed with the person in focus. That is, there is a shutter and an image is taken. FIGS. 3A and 3B are diagrams schematically showing photographs taken in this manner. In FIG. 3A, the human recognition unit 5 recognizes a human being next to the motorcycle located at the center, and the focus processing unit 6 focuses on the person to take a picture. In FIG. 3B, the human recognition unit 5 recognizes a person who is away from the landscape at the center and located on the left side of the photograph, and the focus processing unit 6 focuses on the person to take a picture. That is, the photographer performs the photographing in a state where the person is in focus only by photographing with the camera held in the state of FIG. 3A or FIG. 3B.

  On the other hand, FIGS. 4 (a) and 4 (b) schematically show photographs taken when the conventional autofocus method is used. In this case, the motorcycle and the landscape located in the center are in focus. As a result, humans are unclear if they are left as they are. Therefore, the focus lock function must be used. By using the present invention, it is not necessary to use the focus lock function, and even a child or an elderly person can easily take a picture in which a person is in focus. In combination with the conventional focus lock function, the function of the present invention may be used for humans, and other subjects may be handled by the focus lock function.

  In the above-described processing, it is determined that the person is to be photographed by detecting the voice in the voice detection unit 4, but instead of the voice detection unit 4, for example, a predetermined button provided in the photographing apparatus 1 is photographed. The photographing apparatus 1 may be provided with a photographing mode accepting unit (not shown) that accepts an input of an instruction to photograph a human when the person presses it. In addition to the buttons, the shooting mode reception unit (not shown) may receive an input of a mode for shooting a person from the screen.

By using the photographing apparatus 1 of the present invention, even if a person is not located at the center portion of the viewfinder, it is possible to easily take a picture while focusing on the person. This makes it easy for a child or an elderly person to take a clear picture in focus on a human being. Also, since the operation of moving the camera such as focus lock is not required, the time required for shooting is shortened.

It is a conceptual diagram which shows an example of the system configuration | structure of this invention. It is a flowchart which shows an example of the processing process of this invention. It is a figure which shows typically the photograph at the time of image | photographing using the imaging device of this invention. It is a figure which shows typically the photograph at the time of image | photographing using the conventional imaging device.

Explanation of symbols

1: Shooting device 2: Arithmetic device 3: Sound collecting device 4: Voice detection unit 5: Human recognition unit 6: Focus processing unit

Claims (3)

An imaging device having a focus adjustment function by an autofocus method and a sound collecting device,
The imaging device
By detecting a speech voice, which is a specific utterance sound when shooting by a photographer with respect to a human being, using the sound collecting device, and detecting a press of the shutter by the photographer, A voice detection unit that determines whether the image is shot;
When it is determined that the image is taken by a human, a human recognition unit that recognizes a human being who is a subject other than the photographer from an image taken by the photographing device
A focus processing unit that focuses on a portion recognized as a human in the image by the focus adjustment function;
A photographing apparatus comprising: The voice detection unit further includes:
When the voiceprint registered in advance matches the voiceprint in the voice, it is detected as the voice of the photographer.
The imaging apparatus according to claim 1, wherein: A photographing processing method in a photographing device having a focus adjustment function by an autofocus method and a sound collecting device,
The imaging device
By detecting a speech voice, which is a specific shout when shooting by a photographer with respect to a human being, using the sound collecting device, and detecting a shutter press by the photographer, Determine if it ’s shooting,
When it is determined that the image is taken by a human , the human being who is the subject other than the photographer is recognized from the image taken by the photographing device,
After the portion recognized as a human in the image is focused by the focus adjustment function, the image is taken by the photographing device.
An imaging processing method characterized by the above.

Images