JP5954535B2 - Image selection apparatus, image selection method, and program - Google Patents

Description

  The present invention relates to a technique for sequentially selecting a predetermined photographed image from a plurality of photographed images supplied in parallel from a plurality of imaging devices.

  Conventionally, as a technique for sequentially selecting a predetermined photographed image from among photographed images supplied in parallel from a plurality of cameras, for example, in the following patent document, a monitoring device composed of a plurality of cameras is attached to a subject of each camera. A technique for receiving a radio signal from a radio tag and selecting a monitoring image in which a subject having the best radio signal reception intensity is captured is described.

JP 2006-101028 A

  However, the above technique only estimates the imaging state of a plurality of monitoring images supplied from a plurality of cameras based on the reception intensity of a wireless signal received from a wireless tag attached to each subject, There has been a problem that an image with high quality is not always selected from a plurality of photographed images.

  The present invention has been made in view of such conventional problems, and can more accurately select a photographed image with better quality among a plurality of photographed images supplied in parallel from a plurality of imaging devices. An object of the present invention is to provide an image selection device, an image selection method, and a program.

In order to solve the above-described problem, in the present invention, an acquisition unit that acquires captured images and image quality information of a plurality of systems sequentially supplied in parallel from a plurality of imaging devices, and the acquisition unit for each system A determination unit that determines whether image quality information can be acquired by the analysis unit, and an analysis of a captured image that is sequentially acquired by the acquisition unit for image quality information of a system that is determined that the image quality information cannot be acquired by the determination unit doing, the image quality information acquisition means for sequentially acquiring, based on the quality indicated respectively by the image quality information sequentially acquisition for each system by the acquisition unit or the image quality information acquisition means is acquired by the acquisition unit that a sequential selection means for selecting a predetermined photographing image from among the plurality of systems of captured images, the discriminating function, the dynamic of the selected function to sequentially select a predetermined photographing image Before starting the, and performs determination of whether or not the image quality information can be acquired by the acquisition unit.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to select more appropriately a picked-up image of better quality among the picked-up images of several systems supplied in parallel from a some imaging device.

It is a figure which shows the switcher illustrated as an image selection apparatus which concerns on this invention common to each embodiment. It is a block diagram which shows the switcher in 1st Embodiment. It is a flowchart which shows the image switching process in 1st Embodiment. It is a block diagram which shows the switcher in 2nd Embodiment. It is a figure which shows a score table. It is a flowchart which shows the image switching process in 2nd Embodiment.

(Embodiment 1)
Hereinafter, a first embodiment of the present invention will be described. FIG. 1 is a block diagram showing a switcher 1 exemplified as an image selection device according to the present invention.

  The switcher 1 has a configuration in which a plurality of video images captured and supplied in parallel by a plurality of cameras A to D are appropriately switched and selectively output to the outside. Here, the video of the plurality of systems is a moving image, and the switcher 1 is sequentially inputted with data of a photographic image of a subject (hereinafter referred to as image data) captured at a predetermined frame rate in each of the cameras A to D. The

  The plurality of cameras A to D are arbitrary imaging devices including CCD (Charge Coupled Device) or CMOS (Complementary Metal Oxide Semiconductor) type imaging sensors. For example, a digital video camera, a digital camera having a video shooting function, It is a camera built in an arbitrary device such as a mobile phone terminal.

  Each of the cameras A to D is equipped with a gyro sensor 101, and the switcher 1 inputs blur amount data indicating the blur amount of each camera sequentially acquired in each of the cameras A to D along with the image data. Is done.

  A connection form between the switcher 1 and each of the cameras A to D is, for example, a wired connection, a wireless connection, or a network connection via an arbitrary communication network. Further, the objects to which the switcher 1 outputs image data are, for example, various image processing apparatuses including a general-purpose personal computer, and image storage apparatuses provided on a network.

  On the other hand, FIG. 2 is a block diagram showing an outline of the electrical configuration of the switcher 1. As shown in FIG. 2, the switcher 1 includes a camera input I / F 11, a video output I / F 12, a control unit 13, a program storage unit 14, and an operation unit 15.

  The camera input I / F 11 is an interface through which image data and blur amount data of each system are sequentially input from the cameras A to D. The image data and blur amount data of each system input to the camera input I / F 11 are sequentially supplied to the control unit 13.

  The video output I / F 12 is an interface that outputs image data of any system sequentially selected by the control unit 13 to the outside.

  The control unit 13 includes a CPU (Central Processing Unit) and its peripheral circuits, and a working internal memory such as a RAM (Random Access Memory), and controls the operation of the switcher 1.

  The program storage unit 14 is composed of, for example, a ROM (Read Only Memory), a non-volatile memory in which stored data can be rewritten as needed. The program storage unit 14 causes the control unit 13 to perform later-described image switching processing for appropriately switching image data output from the video output I / F 12 to the outside based on blur amount data accompanying the image data of each system. Is stored in advance.

  The operation unit 15 includes various switches such as a power switch for the user to turn on / off the power of the switcher 1 and a switch used for instructing start and end of video output.

  On the other hand, in FIG. 3, while the cameras A to D connected to the switcher 1 are operating, the control unit 13 follows the program stored in the program storage unit 14 in response to a video output start instruction from the user. It is the flowchart which showed the content of the image switching process to perform.

  The control unit 13 starts to receive a plurality of lines of image data and blur amount data from each of the cameras A to D along with the start of the operation (step SA1), and a predetermined line of image data determined in the initial state, for example, Output of image data supplied from the camera A is started (step SA2).

  And the control part 13 starts the count of the elapsed time after starting the output of image data by the timer function using a built-in clock (step SA3). Thereafter, while outputting image data, the following processing is repeated.

  First, the control unit 13 checks whether or not the elapsed time during counting has reached a predetermined time (step SA4). The elapsed time confirmed in step SA4 at the beginning of processing is the elapsed time since the start of the output of image data of a predetermined system. The predetermined time is several seconds (for example, 5 seconds).

  Then, the control unit 13 waits for processing until the elapsed time reaches the predetermined time (step SA4: NO), and continuously outputs the image data of the predetermined system.

  Thereafter, when the elapsed time reaches a predetermined time (step SA4: YES), the control unit 13 calculates the shake amount of each camera A to D indicated by the shake amount data supplied accompanying the image data of each system. Confirm that the image data of any system supplied from each of the cameras A to D with the smallest amount of blur is the data of the captured image with the highest image quality, and the image data of the system has the highest priority. It determines as image data (step SA5).

  Next, the control unit 13 checks whether or not the image data of the system determined as the image data having the highest priority is currently being output. If the image data is currently being output (step SA6: YES), it remains as it is. The process returns to step SA4. Thereafter, since the predetermined time has already been reached since the output of the image data of the predetermined system is started (step SA4: YES), the control unit 13 immediately performs the process of step SA5 again.

  On the other hand, if the image data of the system determined as the image data having the highest priority in the process at step SA5 is not currently being output (step SA6: NO), the control unit 13 prioritizes the shooting data to be output. The image data is switched to the new system image data having the highest degree (step SA7). That is, the image data of the system newly determined as the image data having the highest priority is selected as a new output target, which is switched to the image data of the system currently being output and output from the video output I / F 12.

  Accordingly, the photographed image being output from the switcher 1 to an arbitrary device or the like is a new photographed image with the smallest amount of blurring (camera shake) at that time among the plural series of photographed images, that is, the best quality. Switch to the captured image.

  Then, after switching the shooting data to be output, the control unit 13 immediately starts counting the elapsed time from that time point (step SA8).

  Subsequently, the control unit 13 returns to the process of step SA4 until a video output end instruction is received from the user (step SA9: NO), and until a predetermined time elapses after the shooting data to be output is switched (step SA4). : NO), the process waits again.

  That is, the control unit 13 continuously outputs the shooting data of the same system for a certain period of time after newly starting the output of the shooting data of the system different from the previous one.

  Thereafter, the control unit 13 repeats the processing of Step SA5 to Step SA8 again when a predetermined time has elapsed from the previous switching time of the shooting data to be output (Step SA4: YES). Thereafter, if there is a video output end instruction from the user (step SA9: YES), the control unit 13 ends the image switching process.

  When the control unit 13 performs the above processing, the switcher 1 of the present embodiment selectively outputs to the outside a plurality of captured images (videos) that are captured by the cameras A to D and supplied in parallel. While the operation is repeated, it is possible to always select a photographed image with a better quality from a plurality of photographed images and output it to the outside.

  In addition, in the switcher 1 of the present embodiment, the captured image with the highest priority is converted into blur amount data indicating the blur amount of each camera, which is sequentially acquired by each of the cameras A to D and supplied along with the image data. To be determined. That is, the photographed image with the highest quality is information that directly represents the quality of the photographed image of each system and represents the quality of the photographed image determined by the photographing state at the time of actual photographing of the photographed image (image quality information). To be determined.

  Therefore, according to the switcher 1 of the present embodiment, it is possible to more accurately select a photographed image having a better quality from a plurality of photographed images and output the selected image to the outside.

  Further, in the present embodiment, after the start of the output of shooting data of a system different from the previous one, the same system of shooting data is always output for a certain period of time. Therefore, according to the present embodiment, it is possible to prevent a photographed image output to the outside from being instantaneously switched to a photographed image of another system in accordance with a relative quality change between the systems. As a result, the captured images of a plurality of systems can be switched in a more natural state.

  In the present embodiment, since the blur amount data supplied from each camera A to D accompanying the image data of each system is used as information directly representing the quality of the captured image of each system, the highest quality is obtained. It is possible to specify a good captured image, that is, to determine image data having the highest priority from a plurality of sets of image data by a simple process.

  Here, in the present embodiment, a case has been described in which the blur amount data sequentially acquired by the cameras A to D is supplied to the switcher 1 along with the image data of each system. However, it is also assumed that the switcher 1 is connected to a camera that cannot acquire blur amount data.

  Therefore, when implementing the present invention, it is desirable to cause the control unit 13 to perform the following processing. For example, the control unit 13 has been able to acquire the shake amount data associated with the image data for each system after starting reception of the image data of the multiple systems and the shake amount data from each of the cameras A to D (step SA1). If there is a system that cannot acquire blur amount data, a process of storing identification information of the system is performed.

  Thereafter, when there is a system in which the blur amount data cannot be acquired, the control unit 13 performs the above identification prior to the process of step SA5 shown in FIG. 3, that is, the process of determining the image data having the highest priority. A process of directly acquiring blur amount data from the image data of the system indicated by the information is performed.

  An arbitrary method can be adopted as a specific process for directly acquiring the blur amount data from the image data. As a method for acquiring blur amount data, for example, for image data of a system to be processed, image data for a plurality of consecutive frames is temporarily stored, and movement of one or feature points between images of successive frames There is a method in which a vector is detected and the magnitude of a motion vector (movement amount of a feature point) is acquired as a shake amount.

  Thereafter, the control unit 13 is caused to perform processing for determining the image data having the highest priority based on the blur amount data attached to the image data and the blur amount data directly obtained from the image data.

  As described above, in the case where image data of a system not accompanied by blur amount data is included in the image data of a plurality of systems, if the control unit 13 performs a process of acquiring the blur amount data directly from the image data of the system. Even when some or all of the cameras that supply multiple sets of image data cannot acquire blur amount data, a better quality shot image can be selected more accurately from multiple sets of shot images. Can be output.

  Further, in the present embodiment, after switching the shooting data to be output to the outside to shooting data of another system, the time for continuously outputting the shooting data of that system is set to a predetermined fixed time. Although described, the minimum duration may be set freely by the user.

  Further, in the above description, a case has been described in which information (image quality information) sequentially acquired and supplied with image data in each of the cameras A to D is blur amount data indicating the blur amount of each camera. .

  However, in the implementation of the present invention, the information used when determining the captured image with the highest priority among the captured images of a plurality of systems is not limited to the blur amount data indicating the blur amount of each camera. , White balance, number of pixels, and the like. The information (image quality information) used when determining the captured image with the highest priority can be any information as long as it directly represents the quality of the captured image. The information directly representing the quality of the captured image includes the actual blurring amount of each main camera such as the degree of blurring of the image and the amount of blurring of the main subject caused by the fast movement of the main subject such as a person at the time of shooting. Information obtained by analyzing a captured image may be used.

  Furthermore, the information obtained by analyzing the actual captured image includes the presence or absence of a specific subject in the captured image described later in the second embodiment. That is, in the present invention, the quality of the captured image is not limited to the original quality of the image, which is generally referred to as image quality, but varies depending on the use of the image, for example, how much a user's request is satisfied. Quality is also included.

(Embodiment 2)
Next, a second embodiment of the present invention will be described. In the present embodiment, similar to that described in the first embodiment, a plurality of video images that are photographed by a plurality of cameras A to D and supplied in parallel, that is, photographed images are appropriately switched and selectively transmitted to the outside. This is related to the output switcher. However, this embodiment is based on the premise that only captured images are sent from the cameras A to D to the switcher.

  FIG. 4 is a block diagram showing the switcher 51 in the present embodiment. The switcher 51 of the present embodiment further includes a face recognition unit 16 and a face data storage unit 17 in addition to the configuration described in the first embodiment.

  The face recognizing unit 16 performs image recognition processing such as binarization, contour extraction, pattern matching, etc. on the image data sent from each camera A to D to the switcher 51, and simply identifies a specific person's face (hereinafter simply referred to as a specific face). The face recognition process for recognizing the image is performed, and face recognition information including a plurality of items is supplied to the control unit 13.

  In this embodiment, the face recognition information supplied from the face recognition unit 16 to the control unit 13 includes first face recognition information indicating the presence or absence of a specific face in the captured image, and a case where a specific face exists in the captured image. The second face recognition information related to the face orientation and the third face recognition information related to the brightness of the face portion when a specific face is present in the photographed image.

  More specifically, the first face recognition information is “having a specific face” or “not”. The second face recognition information is a corresponding direction when the face orientation is classified into three types according to a predetermined standard, and is “front”, “oblique”, or “horizontal”. The third face recognition information is a state of the corresponding brightness when the brightness of the face portion is evaluated in three stages according to a predetermined standard, and is “normal”, “bright (overexposed)”, “dark” (Unexposed) ".

  The face data storage unit 17 is a specific face to be detected by the face recognition unit 16 and stores face image data registered in advance in the switcher 51 by the user. A method for registering a specific face (storing face image data) in the switcher 51 is arbitrary.

  As an example, while a photographed image in which a specific face is photographed up from the cameras A to D is input to the switcher 51, a face portion of an arbitrary person in the image according to a predetermined switch operation by the user Is automatically recognized by simple image recognition, and image data of the recognized face portion is automatically cut out and recorded.

  In the present embodiment, a program is stored in the program storage unit 14 for causing the control unit 13 to perform an image switching process, which will be described later, which is different from the first embodiment.

  Further, the program storage unit 14 also includes data constituting the first score table 61 to the third score table 63 shown in FIGS. 5A to 5C that are referred to by the control unit 13 in the image switching process. It is remembered.

  The first score table 61 is a table showing evaluation scores according to the contents of the previously described first face recognition information acquired by the face recognition unit 16 (“There is a specific face” or “No”). is there. The second score table 62 shows evaluation scores according to the contents of the previously described second face recognition information acquired by the face recognition unit 16 (any one of “front”, “oblique”, and “horizontal”). It is a table. The third scoring table 63 is the content of the previously described third face recognition information acquired by the face recognition unit 16 (any one of “normal”, “bright (overexposed)”, “dark (underexposed)”). It is a table which shows the evaluation score according to.

  Here, the evaluation scores shown in the first score table 61 to the third score table 63 indicate the image quality of the captured images of each system in the image switching process described later by the control unit 13, and the first face recognition information. A value used for quantification based on the third face recognition information.

  Since the other parts of the switcher 51 are the same as those of the switcher 1 of the first embodiment, the same reference numerals are given to the respective parts and the description thereof is omitted.

  FIG. 6 shows that in the present embodiment, the control unit 13 is stored in the program storage unit 14 in response to a video output start instruction from the user while the cameras A to D connected to the switcher 1 are operating. It is the flowchart which showed the content of the image switching process performed according to a program.

  In the present embodiment, the control unit 13 starts receiving image data of a plurality of systems from each of the cameras A to D along with the start of operation (step SB1), and then the image data of a predetermined system determined in an initial state, For example, output of image data supplied from the camera A is started (step SB2).

  Next, the control unit 13 performs face recognition processing for recognizing a specific face registered in the switcher 1 (image data is stored in the face data storage unit 17) for each type of image data. The recognition unit 16 is made to acquire a plurality of items of face recognition information including the first to third face recognition information described above for each system (step SB3).

  Subsequently, the control unit 13 uses the first score table 61 to the third score table 63 shown in FIGS. 5A to 5C for the image data of each system in the process of step SB3. An evaluation score corresponding to the acquired face recognition information of the plurality of items is acquired for each item (step SB4).

  More specifically, for example, a certain face in a captured image of a certain system exists and the face recognition unit 16 evaluates that the brightness of the specific face is normal. If it is, the control unit 13 acquires “10” as the evaluation score according to the first face recognition information, acquires “3” as the evaluation score according to the second face recognition information, “3” is acquired as an evaluation score corresponding to the face recognition information of 3.

  Next, the control unit 13 sums the evaluation scores of all items of the first face recognition information to the third face recognition information for each image data of each system, and acquires the total score for each system (step SB5). .

  At this time, as described above as an example, the total score acquired by the control unit 13 includes a specific face in which a captured image of a certain system faces the front, and the face recognition unit If the brightness of the specific face is evaluated to be normal at 16, the value is “16”.

  Next, the control part 13 confirms whether the maximum of the total score for every system | strain acquired by step SB5 is more than a predetermined threshold value (step SB6). The threshold used in such processing is a predetermined score corresponding to a predetermined image quality determined in advance.

  When the control unit 13 determines that the maximum total score for each system is not equal to or greater than the threshold and the quality of the captured image of the system having the maximum total score is lower than a predetermined image quality ( Step SB6: NO), the process returns to Step SB3 as it is, and the process up to Step SB5 is repeated.

  Further, the control unit 13 determines that the quality of the captured image of the system in which the maximum total score for each system is equal to or greater than the threshold and the total score is maximum is higher than a predetermined predetermined image quality or higher. When the determination is possible (step SB6: YES), the image data of the system having the maximum total score is determined as the image data having the highest priority among the image data of the plurality of systems (step SB7).

  Subsequently, the control unit 13 confirms whether the image data of the system with the highest priority is currently being output, and if it is currently being output (step SB8: YES), the control unit 13 returns to the process of step SB3 as it is, The process up to step SB7 is repeated.

  On the other hand, if the image data of the system having the highest priority determined in step SB7 is not currently being output (step SB8: NO), the control unit 13 sets the shooting data to be output as a new one having the highest priority. Switching to system image data (step SB9). That is, the image data of the system newly determined as the image data having the highest priority is selected as a new output target, which is switched to the image data of the system currently being output and output from the video output I / F 12.

  Accordingly, the photographed image being output from the switcher 1 to an arbitrary device or the like is switched to a photographed image having the highest quality among a plurality of photographed images at that time and having a certain level of quality.

  Thereafter, the control unit 13 repeats the processing from step SB3 onward until a video output end instruction is received from the user (step SB10: NO), and if there is a video output end instruction from the user (step SB10: YES). ), The image switching process is terminated at that time.

  When the control unit 13 performs the above processing, also in the switcher 51 of the present embodiment, a plurality of systems of captured images (videos) captured by the cameras A to D and supplied in parallel are selectively output to the outside. While repeating this operation, it is always possible to select and output a photographic image with a better quality from among a plurality of photographic images.

  Moreover, in the switcher 51 of the present embodiment, the captured image with the highest priority is obtained from a plurality of items acquired by recognizing a specific face registered in advance for each system of captured images. Determined based on the face recognition information. That is, the photographed image with the highest quality is information that directly represents the quality of the photographed image of each system, and is determined based on information (image quality information) obtained by analyzing the actual photographed image.

  Therefore, according to the switcher 51 of the present embodiment, it is possible to more accurately select a photographed image having a better quality among a plurality of photographed images and output the selected image to the outside.

  Further, in the present embodiment, the captured image that is newly output during the output of the captured image of an arbitrary system has the highest quality among the captured images of a plurality of systems, and has an image quality of a certain level or more. Limited to things. In other words, on the condition that the image quality exceeds a certain level, a photographed image with the highest quality among the photographed images of each system is newly output. Therefore, according to the present embodiment, it is possible to prevent a captured image output to the outside from being frequently switched to a captured image of another system in accordance with a relative quality change between the systems.

  That is, when the image quality of the captured images of a plurality of systems is generally low, the order of quality of the captured images between the systems is frequently switched because the range of quality of the captured images between the systems is narrow. It is expected that the photographed image having the highest quality among the photographed images of the system is frequently replaced.

  On the other hand, in the present embodiment, on the condition that the image quality exceeds a certain level, a photographed image with the highest quality among the photographed images of each system at that time is newly output. Therefore, as described above, when the image quality of the captured images of a plurality of systems is generally low, even if the captured image with the highest quality among the captured images of each system is frequently replaced, the image is output to the outside. It is possible to prevent the captured image from being frequently switched to a captured image of another system. As a result, the captured images of a plurality of systems can be switched in a more natural state.

  In the present embodiment, the description has been given of the case where the image quality that is a reference of a captured image to be newly output at an arbitrary time point, that is, the threshold value used in the process of step SB6 described above is fixedly determined in advance. However, the threshold value may be configured so that the user can set in advance. Furthermore, it may be configured such that the user can change the threshold value in real time by a predetermined switch operation while the switcher 51 outputs a captured image of any system to the outside.

  In the present embodiment, the case has been described where the information obtained by analyzing the captured images of each system is face recognition information including a plurality of items of first face recognition information to third face recognition information. The number of items of face recognition information can be increased or decreased as necessary.

  Moreover, in this embodiment, when quantifying the image quality of the captured image of each system based on the face recognition information, each of the first face recognition information to the third face recognition information for each image data of each system. Although the evaluation scores of the items are simply summed, a predetermined weight may be set for each item, and the evaluation scores of each item may be weighted and summed.

  When weighting and summing the evaluation scores of each item, it is conceivable that the weight set for each item is set according to the object to which the switcher 51 outputs a captured image. An example is as follows.

  Here, in addition to the first face recognition information to the third face recognition information, the face recognition information includes the fourth face recognition information indicating the size of a specific face in the photographed image, and the content is the face recognition information. It is assumed that any one of “Large”, “Medium”, and “Small” is a corresponding size when the size of is divided into three levels. In addition, a fourth score table showing the evaluation score according to the fourth face recognition information is prepared, and the evaluation score shown in the fourth score table corresponds to, for example, the face size “large”. It is assumed that the thing corresponding to the thing “3” and “medium” is “2”, and the thing corresponding to “small” is “1”. Furthermore, it is assumed that the target for outputting the captured image is an external device having an arbitrary display screen for displaying the captured image (video).

  With the above assumptions, it is conceivable that each item of face recognition information is weighted according to the resolution of the display screen. More specifically, the control unit 13 is caused to acquire the resolution of the display screen in the external device by an arbitrary method. Then, in the first to third face recognition information, a weight of 1 is set for each of the first to third face recognition information regardless of the resolution of the display screen. If the resolution is higher than a predetermined reference, The face recognition information (face size) is set to 3 times as a weight, and conversely if the resolution is lower than a predetermined standard, the fourth face recognition information is set to 1 time as a weight. Make it.

  In this case, when the resolution of the display screen that displays the photographed image (video) is lower than the standard, the larger the specific face in the photographed image, the higher the quality of the photographed image selected as compared with the one in the present embodiment. It becomes easy to be done. In other words, a captured image whose face is easy to confirm on the display screen can be output with higher frequency. As a result, it is possible to select a photographed image with a better quality among a plurality of photographed images and output it to the outside more accurately.

  Here, the case where information (image quality information) obtained mainly by analyzing captured images of each system is face recognition information has been described, but information obtained by analyzing captured images of each system is a face. Even if it is information other than the recognition information, if it is information of multiple items, when quantifying the image quality of the captured images of each system, a predetermined weight is set for each item. The evaluation scores may be weighted and summed.

  Further, in the present embodiment, the case has been described where the information obtained by analyzing the captured images of each system is the face recognition information representing the quality of a specific face part, but the captured images of each system are analyzed. The obtained information may be information representing the quality of only a specific subject area other than the face. Also, the information obtained by analyzing the captured images of each system can naturally be information representing the overall quality of the captured images.

  Further, in the present embodiment, the description has been given of the case where the switcher 51 (the control unit 13) acquires the face recognition information indicating the quality of a specific face portion from the captured images of each system. However, the face recognition information and information representing the quality of only a specific subject area other than the face may be acquired from the cameras A to D in whole or in part.

  Further, in the present embodiment, the control unit 13 causes the captured image to be newly output during the output of a captured image of an arbitrary system to have the highest quality among a plurality of captured images, and to exceed a certain level. A case has been described in which processing that limits image quality (hereinafter referred to as quality limitation processing) is performed.

  However, in parallel with the quality limiting process or in place of the quality limiting process, the control unit 13 newly outputs imaging data of a system different from the previous one as described in the first embodiment. After the start, processing for continuously outputting the same system of photographing data for a certain period of time (hereinafter referred to as output holding processing) may be performed.

  The same applies to the first embodiment described above. That is, the control unit 13 according to the first embodiment may perform the quality limiting process described in the second embodiment in parallel with the output holding process or instead of the output holding process. .

  Further, in the first embodiment and the second embodiment, the switcher 1 or 51 has the highest priority from among a plurality of lines of photographing data sequentially supplied in parallel from the plurality of cameras A to D. A case has been described in which only image data (photographed images) of a system determined to be high is sequentially selected and output.

  However, when carrying out the present invention, a configuration in which two or more pieces of image data are selected from a plurality of sets of image data and output may be employed. For example, a configuration may be adopted in which image data of the first system and image data of the second system having the priority order corresponding to the priority are sequentially selected from a plurality of systems of image data and output.

  Furthermore, in the configuration for outputting image data of two or more systems, the image size (number of pixels) of each image data is converted into an image size corresponding to the priority and output, or each image data is prioritized. Alternatively, priority information indicating the degree may be added and output. In that case, when displaying a plurality of captured images based on image data of two or more systems, a captured image having a relatively high priority can be displayed in a larger size.

  As mentioned above, although several embodiment of this invention and its modification were demonstrated, if these are in the range in which the effect of this invention is acquired, it can change suitably, and embodiment after change is also a claim. It is included in the scope of the invention described in the scope of the invention and the invention equivalent to the invention.

The invention described in the scope of the claims of the present application will be appended below.
[Claim 1]
Acquisition means for acquiring a plurality of systems of captured images sequentially supplied in parallel from a plurality of imaging devices, and the quality of the captured image obtained by analyzing the actual captured image, or capturing at the time of actual capturing of the captured image Image quality information acquisition means for acquiring the image quality information representing the quality of the captured image determined by the state for each of the plurality of systems, and based on the quality respectively indicated by the image quality information acquired for each system by the image quality information acquisition means An image selection apparatus comprising: a selection unit that sequentially selects a predetermined captured image from a plurality of captured images acquired by the acquisition unit.
[Claim 2]
The image selection apparatus according to claim 1, further comprising an output unit that outputs a captured image selected by the selection unit.
[Claim 3]
Based on the quality indicated by each of the image quality information, it further comprises priority determining means for sequentially determining a photographed image having the highest priority among the plurality of photographed images, and the selecting means is configured to determine the priority. 3. The image selection apparatus according to claim 1, wherein the picked-up images determined to have the highest priority by the means are sequentially selected.
[Claim 4]
4. The image selection apparatus according to claim 1, wherein the image quality information is information indicating a blur amount.
[Claim 5]
The image quality information acquisition means is determined as the image quality information based on a quality of a captured image obtained by analyzing an actual captured image in the plurality of imaging devices or a captured state at the time of actual capturing of the captured image. 5. The image selection apparatus according to claim 1, wherein predetermined information is acquired as quality of a captured image and sequentially supplied together with the captured images of the plurality of systems.
[Claim 6]
6. The image selection apparatus according to claim 5, wherein the image quality information acquisition unit acquires image quality information that cannot be acquired from a captured image of the system when any of the image quality information of the plurality of systems cannot be acquired.
[Claim 7]
The image selection apparatus according to claim 1, wherein the image quality information acquisition unit acquires the image quality information from each of the plurality of captured images.
[Claim 8]
2. The image quality information of each system acquired by the image quality information acquisition unit includes image quality information that directly represents the quality of only a specific subject area in a captured image of each system. 7. The image selection device according to any one of 7.
[Claim 9]
The image quality information acquiring unit acquires image quality information including a plurality of items as the image quality information for each system, and assigns a weight corresponding to a predetermined condition to the image quality information of each item for each item. The priority determining means evaluates the quality indicated by the image quality information of each item for each item for the captured images of each system, and the evaluation result for each item is assigned to each item by the weighting means. And adding the weights set in the above, and sequentially determining the photographed image having the highest priority among the photographed images of the plurality of systems based on the aggregated results for each system with weighting. The image selection device according to claim 3.
[Claim 10]
The selection unit is provided on the condition that the quality indicated by the image quality information acquired by the image quality information acquisition unit relating to the photographed image determined to have the highest priority by the priority determination unit is equal to or higher than a predetermined threshold. 10. The image selection apparatus according to claim 3, wherein the picked-up images determined to have the highest priority by the priority determination means are sequentially selected.
[Claim 11]
The selection operation control means for stopping the selection operation of a predetermined photographed image in the selection means until a specified time elapses after the previous selection operation is performed. The image selection apparatus described.
[Claim 12]
In the method of sequentially selecting a predetermined captured image from a plurality of captured images sequentially supplied in parallel from a plurality of imaging devices, the quality of the captured image obtained by analyzing the actual captured image, or the captured image Based on the step of sequentially acquiring image quality information representing the quality of the captured image determined by the shooting state at the time of actual shooting for each of the plurality of systems, and the quality indicated by the image quality information acquired for each system, And a step of sequentially selecting a predetermined photographed image from the plurality of photographed images.
[Claim 13]
A computer of an image selection device that selects and sequentially selects a predetermined captured image from a plurality of captured images sequentially supplied in parallel from a plurality of imaging devices, and a captured image obtained by analyzing an actual captured image Image quality information acquisition means for acquiring the image quality information representing the quality or the quality of the captured image determined by the shooting state at the time of actual shooting of the captured image for each of the plurality of systems, and for each system by the image quality information acquiring means A program that functions as a selection unit that sequentially selects a predetermined captured image from among a plurality of captured images acquired by the acquisition unit based on the quality indicated by the image quality information.

1 Switcher 9 Operation unit 11 Camera input I / F
12 Video output I / F
DESCRIPTION OF SYMBOLS 13 Control part 14 Program memory | storage part 15 Operation part 16 Face recognition part 17 Face data memory | storage part 51 Switcher 61 1st score table 62 2nd score table 63 3rd score table 101 Gyro sensor AD Camera

Claims (12)

An acquisition means for acquiring a plurality of captured images and image quality information sequentially supplied in parallel from a plurality of imaging devices;
For each of the systems, determination means for determining whether image quality information can be acquired by the acquisition means,
Image quality information acquisition means for sequentially acquiring image quality information of a system determined to be unable to acquire image quality information by the determination means, by analyzing captured images sequentially acquired by the acquisition means;
Based on the quality indicated respectively by the image quality information sequentially acquisition for each system by the acquisition unit or the image quality information obtaining means, a predetermined photographing image from among the captured images of a plurality of systems to be acquired by the acquisition unit and a sequential selection selecting means,
The image selection apparatus according to claim 1, wherein the determination unit determines whether or not image quality information can be acquired by the acquisition unit before the selection unit starts an operation of sequentially selecting a predetermined captured image .   The image selection apparatus according to claim 1, further comprising an output unit that outputs a captured image selected by the selection unit. Based on the quality indicated by each of the image quality information, further comprising priority determination means for sequentially determining a captured image having the highest priority among the captured images of the plurality of systems,
The image selection apparatus according to claim 1, wherein the selection unit sequentially selects a captured image determined to have the highest priority by the priority determination unit. The picture image Quality information image selection apparatus according to claim 1 or 2, wherein the information indicating the blur amount. The selection operation of a predetermined photographing image in the selection means, to any of claims 1 to 4, characterized in that Ru further comprising a selection operation control means for stopping until the lapse of predetermined time since the previous selection operation performed The image selection apparatus described. Acquisition means for acquiring a plurality of captured images sequentially supplied in parallel from a plurality of imaging devices;
A quality information obtaining unit that the image quality information indicating the quality of the resulting photographed image by analyzing the captured images sequentially obtained for each line, sequentially acquired by the acquisition unit,
A discriminating unit that sequentially discriminates whether or not the highest quality is equal to or higher than a predetermined threshold among the quality indicated by the image quality information sequentially acquired for each system by the image quality information acquiring unit;
A selection unit that sequentially selects a photographed image of the highest quality among a plurality of photographed images sequentially acquired by the acquiring unit when the determining unit determines that the highest quality is equal to or higher than a predetermined threshold. and,
When the discriminating unit determines that the highest quality is equal to or higher than a predetermined threshold, the captured image sequentially selected by the selecting unit is output. And an output means for maintaining the image selection apparatus. 6. the image quality information of the image quality information obtaining means each are Ru-based integration of acquired by the, characterized in that image quality information indicating the quality of only the specific object region in the captured image for each said system directly is included The image selection device described in 1. The image quality information acquisition means sequentially acquires image quality information consisting of a plurality of items as the image quality information for each system,
For each item before outs image quality information, further comprising a weighting means for setting a weight according to a predetermined condition,
Said determining means, for capturing images of each of the strains, the each item of the image quality information Ataishi commentary quality, with a weighting that is set for each item the evaluation result for each said item by said weighting means with aggregate, based on the accumulation result for each line accompanied by only the with weighting, that most high quality of the captured images of the plurality of channels is sequentially determined whether it is greater than the predetermined threshold value The image selection device according to claim 6 or 7, characterized in that A process of acquiring a plurality of captured images and image quality information sequentially supplied in parallel from a plurality of imaging devices ;
For each system, a process for determining whether or not the image quality information can be acquired;
A process of sequentially acquiring image quality information of a system determined to be unable to acquire image quality information by analyzing captured images that are sequentially acquired;
Based on the quality indicated respectively by the image quality information obtained for each line, see contains a process for sequentially selecting the predetermined photographing image from among the photographic images of the plurality of systems,
An image selection method comprising: determining whether or not the image quality information can be acquired before starting an operation of sequentially selecting the predetermined photographed image . Processing for acquiring captured images of a plurality of systems sequentially supplied in parallel from a plurality of imaging devices;
Image quality information indicating the quality of the resulting photographed image by analyzing the captured images sequentially obtained for each of the system, a process of sequentially obtaining,
Among the quality indicated by the image quality information sequentially acquired for each system, a process of sequentially determining whether the highest quality is a predetermined threshold or more,
When it is determined that the highest quality is equal to or higher than a predetermined threshold , among the plurality of captured images, a process of sequentially selecting the highest quality captured image ;
A process of outputting the sequentially selected captured images when it is determined that the highest quality is equal to or higher than a predetermined threshold, and maintaining the output of the output captured image when not determined. An image selection method as a feature. On your computer,
An acquisition function for acquiring a plurality of captured images and image quality information sequentially supplied in parallel from a plurality of imaging devices,
A discriminating function for discriminating whether image quality information can be acquired by the acquisition function for each system.
Image quality information acquisition function for sequentially acquiring image quality information of a system determined to be unable to acquire image quality information by the determination function, by analyzing captured images sequentially acquired by the acquisition function ,
Based on the quality indicated respectively by the image quality information sequentially obtained for each line by the acquisition function or the quality information acquiring function, sequentially a predetermined shot image from among the captured images of a plurality of systems acquired by the acquisition function Selection function to select
Realized,
The determination function determines whether or not image quality information can be acquired by the acquisition function before the selection function starts an operation of sequentially selecting a predetermined captured image . On your computer,
An acquisition function for acquiring captured images of a plurality of systems sequentially supplied in parallel from a plurality of imaging devices;
The acquisition function by the image quality information indicating the quality of the captured image obtained by analyzing the captured images sequentially obtained for each line, the image quality information obtaining function for sequentially acquired,
Wherein among the quality of Ru indicated by the image quality information acquired image quality information sequentially obtained for each system by the function, determination function the highest quality is sequentially determined whether it is greater than the predetermined threshold value,
A selection function that sequentially selects a photographed image of the highest quality among a plurality of photographed images sequentially acquired by the acquisition function when the highest quality is determined to be equal to or higher than a predetermined threshold by the determination function. ,
When the discriminating function determines that the highest quality is equal to or higher than a predetermined threshold, the photographic image sequentially selected by the selection function is output. Output function to maintain
A program to realize

Images