JP7257242B2 - Video monitoring device and video monitoring program - Google Patents

Description

The present invention relates to a video monitoring apparatus and a video monitoring program for monitoring each resource video displayed on a multi-screen.

Conventionally, in television broadcasting stations, etc., multiple resource images such as weather camera images, relays, news materials, etc. It is sent to the broadcasting station via a transmission line such as. External transmission always requires stable operation (stable line and stable video), and usually these multiple resource videos are displayed on a multi-screen, and the operator visually monitors the status of each resource video. there is

On the other hand, Patent Literature 1 discloses a technique of comparing a plurality of resource videos with a video that is currently being broadcast, and determining the consistency between the resource video and the video that is used in the broadcast. According to the technology described in Patent Document 1, in a multi-screen displaying a plurality of resource videos, if any resource video matches the video being on-air during broadcasting (broadcast use video), the resource video will be on-air. It is possible to tally display the resource video (display a frame surrounding the resource video).

JP 2018-156344 A

However, there is room for improvement in the currently used method of displaying a plurality of resource videos on a multi-screen and viewing them by the operator, as described below.
In transmission over an IP line, operation (transmission) starts without the operator noticing. Also, with IP lines, even if a line failure occurs and freezes, an alarm is not raised, so it is difficult to notice an abnormality. In weather camera images and the like, even if the image is blown out or the position is shifted from the default position, it is often not noticed unless the operator carefully looks at it. In addition, it is difficult to constantly monitor the main line and the backup FPU line during sports broadcasting.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a video monitoring apparatus and a video monitoring program capable of detecting a change in the state of a multi-screen and notifying the operator of the change. .

In order to solve the above-described problems, a video monitoring apparatus according to the present invention is a video monitoring apparatus for monitoring each resource video on a multi-screen displaying a plurality of resource videos on one screen, comprising storage means and video reading means. , analysis pattern selection means, analysis processing means, and monitoring result output means.

According to this configuration, the video monitoring apparatus reads the input image in which the resource name of the resource video is displayed for each resource video in the multi-screen by the video reading means, and the analysis pattern selection means reads the input image for each resource video. The resource name displayed on the child screen is extracted by character recognition, and the storage means storing the analysis pattern indicating the content of the analysis process corresponding to the type of resource video is referenced and extracted. You can select a parsing pattern that corresponds to the resource name you specified.
Thereby, the video monitoring apparatus can identify the resource video displayed on the multi-screen by the resource name, and select an analysis pattern corresponding to the type of the resource video of the identified resource name. Therefore, the video monitoring apparatus can execute the analysis processing of each resource video with the content of the optimum analysis pattern corresponding to the type of resource video.

In addition, the video monitoring apparatus executes analysis processing indicated by the selected analysis pattern for each resource video by the analysis processing means, and the monitoring result output means is configured to detect the state change of each resource video detected by the analysis processing. Result information can be output.
As a result, the video monitoring apparatus can output the result information regarding the state change of the resource video detected in the analysis processing for each resource video, and inform the operator or the like.

As described above, the video monitoring apparatus of the present invention automatically outputs changes in the state of resource video for each resource video in the multi-screen. be able to. In addition, even if the state of the resource video changes due to an abnormality in the video such as freezing or overexposure, the change in state (abnormality) of the resource video can be reliably notified to the operator or the like.

According to the present invention, it is possible to provide a video monitoring apparatus and a video monitoring program that detect a change in the state of a multi-screen and notify the operator.

1 is a functional block diagram showing the configuration of a video monitoring device according to an embodiment; FIG. FIG. 4 is a diagram for explaining character information added to an input image (multi-screen) according to the embodiment; It is a functional block diagram which shows the structure of the analysis processing means which concerns on this embodiment. It is a functional block diagram which shows the structure of the monitoring result output means which concerns on this embodiment. It is a figure which illustrates the monitoring screen which concerns on this embodiment. It is a figure which illustrates the monitoring screen which concerns on this embodiment. 4 is a diagram exemplifying the data configuration of initial setting information saved in a setting file according to the embodiment; FIG. 4 is a diagram exemplifying the data configuration of analysis pattern setting information saved in a setting file according to the embodiment; FIG. 4 is a flow chart showing the overall processing flow of the video monitoring apparatus according to the embodiment; 6 is a flow chart showing the flow of processing of analysis pattern 1 according to the present embodiment. 9 is a flowchart showing the flow of processing of analysis pattern 2 according to the present embodiment; 9 is a flowchart showing the flow of processing of analysis pattern 3 according to the present embodiment; 9 is a flowchart for explaining a specific example of freeze detection in analysis pattern 3; 10 is a flowchart for explaining a specific example of whiteout detection in analysis pattern 3; 10 is a flowchart for explaining a specific example of positional deviation detection in analysis pattern 3; 10 is a flow chart showing the flow of processing of analysis pattern 4 according to the present embodiment. 10 is a flowchart for explaining a specific example of freeze detection in analysis pattern 4; 10 is a flowchart for explaining a specific example of silence detection in analysis pattern 4;

EMBODIMENT OF THE INVENTION Hereafter, the form (it is called "embodiment" hereafter) for implementing this invention is demonstrated with reference to drawings.
<<Configuration of Video Monitoring Device>>
First, the configuration of the video monitoring apparatus 1 will be described with reference to FIG.
FIG. 1 is a functional block diagram showing the configuration of a video monitoring device 1 according to this embodiment.
The video monitoring apparatus 1 captures, as an input image 1000, a multi-screen that displays a plurality of resource videos such as weather camera videos, live broadcasts, and news materials. Then, the video monitoring apparatus 1 divides the input image 1000 for each resource, and performs character recognition of the resource name (character information) displayed on the child screen. The video monitoring apparatus 1 selects an analysis pattern from the character recognition result, and executes analysis processing suitable for each resource. When the video monitoring apparatus 1 detects a state change or an abnormality in each resource video, the video monitoring device 1 performs tally display on the output image 2000, outputs an audio alarm, or the like in order to notify the operator. Execute output processing of monitoring results.

This video monitoring apparatus 1 captures a plurality of resource videos displayed on a multi-screen by, for example, an HD-SDI (High Definition-Serial Digital Interface) input/output board. In addition, the video monitoring apparatus 1 outputs an output image 2000 that tally-displays screens of resources whose status has changed and resources in which an abnormality has been detected, from the HD-SDI input/output board as a result of monitoring. A speaker 6 is connected to the video monitoring apparatus 1, and audio information indicating a state change or abnormality is output to the speaker 6. FIG.

A monitor 5 is connected to the video monitoring apparatus 1, and a monitoring result indicating state changes and abnormalities is displayed as a monitoring screen 3000 (see FIGS. 5 and 6, which will be described later).
The video monitoring apparatus 1 is installed, for example, in addition to equipment for monitoring transmission lines in a broadcasting station.
This video monitoring apparatus 1 includes control means 10, storage means 20, and input/output means 30, as shown in FIG.

The input/output means 30 consists of a communication interface for transmitting and receiving information, an input/output interface for transmitting and receiving information between an input device such as a keyboard and an output device such as a monitor 5 and a speaker 6. .
Specifically, the input/output unit 30 includes, for example, the above-described HD-SDI input/output board, and captures the multi-screen displaying the resource video as the input image 1000 . Also, a multi-screen tally display of the resource video in which the state change or abnormality is detected is output as an output image 2000 via the HD-SDI input/output board.

An input image 1000 (and an output image 2000) shown in FIG. 1 is an image in which a plurality of resource images such as weather camera images, broadcasts, and news materials are displayed on one screen. is exemplified as a multi-screen that displays 25 resource images obtained by dividing the image into 5×5. The number of divisions of this multi-screen is not limited to 25 (5×5) screens, and may be, for example, 9 (3×3) screens or 16 (4×4) screens.

The multi-screen input image 1000 will be described in detail.
FIG. 2 is a diagram for explaining character information added to the input image 1000 according to this embodiment. Character information is attached to the input image 1000 according to the present embodiment as a resource name indicating the contents of the resource of each image of each resource video. The positions on the 25-divided multi-screen are described as "Pic1" to "pic25" as shown in FIG.

In the case of the 25-screen multi-screen shown in FIG. The resource name is attached as text information to each resource video, such as "N41", the text information of the resource at the screen position of "Pic9", "Report IP1", etc. at the screen position of "Pic16". .
Also, when audio information is attached to the resource video, an audio level meter is displayed, for example, at the lower left of the screen of each resource video (not shown). The sound level meter displays the volume of sound as a level meter on the screen.

Returning to FIG. 1, the storage unit 20 is composed of a flash memory, a hard disk, a RAM (Random Access Memory), or the like. The storage means 20 also stores resource names of resource videos, character recognition information for identifying the resource names, analysis pattern information associated with each character recognition information, etc., in association with the screen positions of the multi-screen. is stored as the setting file 200 (see FIGS. 7 and 8 for details). Further, in this setting file 200, various parameters for executing analysis processing are set in advance in association with each resource image.
In addition, the storage means 20 temporarily stores a program (video monitoring program) for executing each function of the control means 10 and information necessary for processing of the control means 10 .

The control means 10 controls overall control of the image monitoring apparatus 1, and includes image reading means 11, analysis pattern selection means 12, analysis processing means 13, monitoring result output means 14, and processing setting means 15, as shown in FIG. consists of
The control means 10 is implemented, for example, by a CPU (Central Processing Unit) (not shown) expanding a program (video monitoring program) stored in the storage means 20 into a RAM and executing the program.

The video reading means 11 reads, as an input image 1000, a plurality of resource videos displayed on a multi-screen via an HD-SDI input/output board provided in the input/output means 30, for example. Then, the video reading means 11 outputs the read input image 1000 to the analysis pattern selecting means 12 .

The analysis pattern selection means 12 divides the input image 1000 into child screens for each resource, and character-recognizes character information (resource names) displayed on the child screens. The analysis pattern selection means 12 selects character information (including partial match) that matches the character string recognized from the child screen, and character recognition information ( (See FIG. 7) to identify the character string. It should be noted that the character string recognized from the child screen and the character recognition information entered in the setting file 200 are matched using NCC (Normalized Cross Correlation), which is one of the existing character recognition techniques. It may be performed by evaluating the degree of similarity.
The analysis pattern selection means 12 selects an analysis pattern corresponding to the identified character string (character recognition information in FIG. 7) using the setting file 200 (initial setting information 210).
The analysis pattern information is information indicating specific analysis contents of the analysis processing means 13, which will be described later, and the analysis processing means 13 executes each analysis process according to this analysis pattern.

The analysis processing means 13 performs analysis processing using one or more of image switching detection, freeze detection, whiteout detection, silence detection, and positional deviation detection for each analysis pattern selected by the analysis pattern selection means 12. do.
Here, the configuration of the analysis processing means 13 will be described in more detail with reference to FIG.
As shown in FIG. 3 , the analysis processing means 13 includes image switching detection means 131 , freeze detection means 132 , whiteout detection means 133 , silence detection means 134 , and displacement detection means 135 .

Image switching detection means 131 detects that the default image has been switched according to the type of default image. Note that what the default image is is stored in advance in the setting file 200 (specifically, the initial setting information 210 in FIG. 7) in the storage means 20 .
(1) When the default image is no signal (freeze image), the image switching detection unit 131 performs arithmetic processing of the luminance level of each pixel of the previous frame and the current frame, and detects that the video has been switched if there is a change. do.

(2) When the default image is a weather camera image, the image switching detection unit 131 detects image switching based on the feature amount of the image. Specifically, as feature points in the image, positions such as edges where brightness changes are large are extracted, and a histogram (feature amount) in the direction of the gradient is calculated from information on the brightness gradient of the area around the feature points. Then, matching processing is performed on the feature amounts of the two screens, and when the value of the matching cost obtained by matching is less than a predetermined threshold, the image switching detection means 131 determines that the two screens are different. Thereby, the image switching detection unit 131 can detect that the screen has been switched.
The image switching detection unit 131 may detect image switching using NCC (Normalized Cross Correlation), which is a template matching technique. In the NCC, image switching can be detected by evaluating the degree of similarity between a pre-registered template image and a target input image.

(3) The image switching detection unit 131 acquires a color histogram when the default image is an image having a characteristic color such as a color bar, a green image, or a blue stripe image, and based on the color information, Detect image switching.

The freeze detection means 132 performs arithmetic processing of the luminance level of each pixel of the previous frame and the current frame, and detects freeze when the previous and subsequent images match at least a predetermined number of frames.
In addition, the blown-out highlight detection unit 133 calculates the luminance level of the image, and detects blown-out highlights when it exceeds a set threshold value.

Silence detection means 134 compares the audio level meter displayed in the divided child screen resource video with the audio level meter during silence, and detects that there is silence when these images match for a predetermined number of frames or more. do.

The positional deviation detection unit 135 determines whether or not there is a positional deviation between the image at the default position and the current image by detecting a feature amount. The positional deviation detection means 135 detects that the positional deviation has occurred when the positional deviation exceeds a predetermined threshold value set in advance.

Returning to FIG. 1 , the monitoring result output unit 14 outputs result information regarding the state change and abnormality of the resource video detected by the analysis processing unit 13 .
Here, with reference to FIG. 4, the configuration of the monitoring result output means 14 will be described in more detail.
As shown in FIG. 4 , the monitoring result output means 14 includes tally output means 141 , audio output means 142 , monitoring screen display means 143 and log output means 144 .

When the image switching detection means 131 of the analysis processing means 13 detects that a certain resource image is switched from the color bar to the image being transmitted, the tally output means 141 performs , for example, outputs a tally image with a green frame. Further, when the image switching detection means 131 detects the end of transmission, it outputs a tally image with a yellow frame, for example, for a predetermined time.
Further, the tally output means 141 detects an abnormality in the multi-screen when the freeze detection means 132, whiteout detection means 133, silence detection means 134, and position deviation detection means 135 of the analysis processing means 13 detect an abnormality. For the detected resource image, for example, a tally image with a red frame is output.

An output image 2000 in FIG. 1 shows an example in which a tally image with a red frame is output because an abnormality has been detected in the resource images whose screen positions are “Pic17” and “Pic25”. In this way, when a change in the state of an image or an abnormality is detected simultaneously on a plurality of resource screens, the tally output means 141 simultaneously outputs a plurality of tally images.

Returning to FIG. 4, the audio output unit 142 outputs preset audio information to the speaker 6 as an audio alarm when the analysis processing unit 13 detects an image change or an abnormality.
For example, when the image switching detection means 131 of the analysis processing means 13 detects the start of transmission as a change in the state of the resource video, it outputs an audio alarm such as "Transmission has started."
In addition, when the freeze detection means 132, the whiteout detection means 133, the silence detection means 134, and the position deviation detection means 135 detect an abnormality, the audio output means 142 outputs an alarm sound to alert the operator as audio information. Output to the speaker 6.
Audio information for outputting these audio alarms is stored in the storage means 20 in advance.

The monitoring screen display means 143 displays the status change or abnormality detection on the monitoring screen displayed on the monitor 5 connected to the video monitoring apparatus 1 when the analysis processing means 13 detects the status change or abnormality of the resource video. do.
FIG. 5 is a diagram exemplifying a monitoring screen 3000 when a state change is detected according to this embodiment.
The monitoring screen 3000 includes an area for displaying a multi-screen including the tally image output by the tally output means 141 (hereinafter referred to as a "multi-screen area 3100") and an area for displaying status changes and detection of abnormalities (hereinafter referred to as a "multi-screen area 3100"). The screen is configured to include an "image change display area 3200") and an area for displaying the latest log (hereinafter referred to as a "log display area 3300").

When the image switching detection means 131 of the analysis processing means 13 detects the start of transmission due to a change in the state of the resource video, the monitoring screen display means 143 displays the resource image in the image change display area 3200 with a green background, for example. Displays the name and status (eg, “HD-CS1 Transmitting”).
Further, when the monitor screen display means 143 detects the freeze detection alarm of "Shimonoseki IP" after the monitor screen 3000 ("HD-CS1 transmitting") in FIG. As shown, the image change display area 3200 displays the resource name and status in a red background, for example, "Shimonoseki IP Freeze Detected". At this time, "Transmitting HD-CS1" may be displayed as a tab 3201 in FIG.

Returning to FIG. 4, the log output unit 144 stores information on the state change and abnormality of the resource video detected by the analysis processing unit 13 in the storage unit 20 as log information.
Further, the log output unit 144 displays the log information stored in the storage unit 20 in the log display area 3300 of the monitor screen 3000 via the monitor screen display unit 143 described above.

Note that the tally output means 141, the voice output means 142, the monitoring screen display means 143, and the log output means 144 included in the monitoring result output means 14 detect image switching and abnormalities by the analysis processing of the analysis processing means 13. Then, the output processing of the analysis result is immediately executed. In addition, by selecting and presetting one or more arbitrary functions from the tally output means 141, the voice output means 142, the monitoring screen display means 143, and the log output means 144, a desired function can be selected. and output the analysis result.

Returning to FIG. 1, the process setting means 15 stores initial setting information for analyzing each resource video, analysis pattern information, parameters for executing each analysis process, etc. in the storage means 20 as a setting file 200. Remember.

FIG. 7 is a diagram exemplifying the data configuration of the initial setting information 210 saved in the setting file 200 according to this embodiment.
The initial setting information 210 stores resource names 212, character recognition information 213, analysis patterns 214, and default images 215 in association with screen positions 211 of the multi-screen.

The resource name 212 is the name of the resource video displayed at that screen position 211 .
The character recognition information 213 is information for character recognition by the analysis pattern selection means 12 by comparing with character information (resource names) displayed on child screens obtained by dividing the multi-screen.
For example, "outside 1" is registered as the character recognition information 213 for the resource name 212 of "outside 1". “HD-CS” is registered as the character recognition information 213 for the resource names 212 of “HD-CS1” and “HD-CS2”. "Kintaikyo camera" is registered as the character recognition information 213 for the resource name 212 of "Kintaikyo camera". For example, when the resource name 212 has the same analysis pattern 214 and default image 215 as in "news IP1", "news IP2", ... "news IP8", numbers are used as the character recognition information 213. The removed common character portion “News IP” can be registered as the character recognition information 213 .

In the analysis pattern 214, one or more of various analysis processes (image switching detection, freeze detection, whiteout detection, silence detection, misregistration detection, etc.) executed by the analysis processing means 13 in association with the character recognition information 213. The identification information of the analysis pattern defined using is stored.

The default image 215 stores an initial image before analysis processing of each resource video.
For example, if the resource name 212 is “outside station 1” or “HD-CS1”, then “green image” is registered as the default image 215 . If the resource name 212 is “Kintaikyo camera”, then “weather camera image” is registered as the default image 215 . If the resource name 212 is “News IP1”, a “freeze image” is registered as the default image 215 . A plurality of images may be registered as the default image 215 . In that case, for example, the positional deviation detection unit 135 (FIG. 3) can determine that the image is in the default position (no positional deviation) if the degree of similarity with any one of the plurality of default images is high. .

FIG. 8 is a diagram exemplifying the data configuration of the analysis pattern setting information 220 saved in the setting file 200 according to this embodiment.
The analysis pattern setting information 220 determines which of the image switching detection, freeze detection, whiteout detection, silence detection, positional deviation detection, etc., is to be executed by the analysis processing means 13 according to the type of resource video. The content of analysis processing (processing scenario) is set (patterned).

For example, "analysis pattern 1" performs image switching detection of the "green image" as the default image, performs character recognition of the characters displayed (blinking) in the color bar that is the switched image, and determines whether the character is being transmitted or not. It defines the processing content of detecting a change in the end state based on image switching. "Analysis pattern 2" is the processing content of analyzing the state change during transmission and transmission completion by detecting the start of transmission from the default image by detecting image switching and detecting the end of transmission by detecting image freezing. stipulates. The “analysis pattern 3” defines the processing contents of performing freeze detection, blown-out highlight detection, and misregistration detection of default video (for example, weather camera image). "Analysis pattern 4" defines the processing contents of performing image switching detection from the default image, and performing freeze detection and silence detection of the image being transmitted.
Hereinafter, a combination of analysis processes, for example, "analysis pattern 5" to "analysis pattern 9", is made to correspond to the type of resource video, and the processing contents are defined.

Also, in this setting file 200, various parameters required for analysis processing indicated by analysis patterns are set for each resource video.
For example, in freeze detection, a threshold value for determining that an image is frozen, in whiteout detection, a predetermined brightness value for determining whiteout in an image, in positioning detection, a predetermined value for positional deviation from the default position, etc. are set as various parameters. Thresholds for predetermined elapsed times (predetermined times 'k', 'm', 'p', etc., which will be described later) for confirming various types of detection are also registered in the setting file 200 in advance.

The process setting means 15 displays the initial setting information 210 and various parameter information input screens on the monitor 5 via the input/output means 30, and stores them in the storage means 20 by having the operator or the like input them. . For example, by selecting (clicking) the "Settings" tag 3001 on the monitoring screen 3000 in FIG. 5, an input screen for the initial setting information 210 and various parameters may be displayed and the operator or the like may input. In addition, the processing setting means 15 may take in the initial setting information 210 and various parameter information from an external device or the like and store them in the storage means 20 .
In addition, the processing setting means 15 causes the monitor 5 to display a screen in which a mode (maintenance mode) for changing to a setting that does not execute the analysis processing can be selected for each resource video, and the operator or the like selects the maintenance mode. Analysis processing may not be executed for a specific resource video that does not require analysis processing.

≪Operation of Video Monitoring Device≫
Next, operations of the video monitoring apparatus 1 will be described with reference to FIGS. 9 to 18. FIG.
First, the overall processing flow of the video monitoring apparatus 1 will be described. Here, as the information of the setting file 200, initial setting information 210 (FIG. 7), analysis pattern setting information 220 (FIG. 8), and various parameter information are registered in advance in the storage means 20, and can be processed by the control means 10. It will be explained assuming that the

FIG. 9 is a flow chart showing the overall processing flow of the video monitoring apparatus 1 according to this embodiment.
First, the video reading means 11 of the video monitoring apparatus 1 reads a plurality of resource videos shown in a multi-screen as an input image 1000 via the input/output means 30 (eg, HD-SDI input/output board) (step S1).

Subsequently, the analysis pattern selection means 12 divides the input image 1000 into child screens for each resource, and extracts character information displayed on the child screens using an existing character recognition method (step S2). Then, the analysis pattern selection means 12 refers to the initial setting information 210 (FIG. 7) of the setting file 200 stored in the storage means 20, and compares the extracted character information with the character recognition information 213 of the initial setting information 210. Then, a matching character string is identified, and an analysis pattern corresponding to the identified character string is selected for each resource video (child screen) (step S3). The process from steps S1 to S3 is the process executed by the video monitoring apparatus 1 by default.

Subsequently, the analysis processing means 13 executes analysis processing with the selected analysis pattern for each resource video (step S4). Note that the analysis processing means 13 performs analysis processing on the divided child screens (resource video) in a predetermined order (for example, the order of "Pic1", "Pic2", . . . , "Pic25").
Then, the monitoring result output unit 14 outputs result information regarding the state change and abnormality of the resource video detected by the analysis processing unit 13 for each resource video (step S5).

Next, the video monitoring apparatus 1 (analysis processing means 13 and monitoring result output means 14) determines whether or not instruction information for system termination has been received (step S6). Then, if the instruction information for system termination has not been received (step S6→No), the process returns to step S4, and the processes of steps S4 to S5 are repeated. In other words, the analysis processing for the divided child screens (resource video) is repeated. On the other hand, when the system termination instruction information is received (step S6→Yes), the video monitoring apparatus 1 terminates the process.

Note that in the multi-screen of the input image 1000, if there is a change in the arrangement or name of the resource video, the video monitoring apparatus 1 stops processing, initializes, and reads the setting file 200 again. Then, the process from step S1 is executed.

Next, analysis patterns 1 to 4 will be described as specific examples of each analysis pattern. The other analysis patterns 5 to 9 shown in FIG. 8 are combinations of the same processing as the individual analysis processing executed in the analysis patterns 1 to 4, so the explanation is omitted.
In the following description, it is assumed that the processing based on the analysis pattern in the resource video of the split child screen is executed from step S4 to step S5 in FIG.

<Analysis pattern 1>
FIG. 10 is a flow chart showing the flow of processing of the analysis pattern 1. As shown in FIG. An example “analysis pattern 1” in which a resource video whose screen position is “Pic4” and whose resource name is “HD-CS1” is analyzed will be described.

Here, in the initial setting of steps S1 to S3 in FIG. 9, the analysis pattern selection means 12 of the video monitoring device 1 extracts "HD-CS1" as character information displayed on the child screen of "Pic4", Assume that analysis pattern 1 is selected by referring to initial setting information 210 in setting file 200 and specifying “HD-CS” in character recognition information 213 .

When the analysis processing is started, the analysis processing means 13 of the video monitoring apparatus 1 executes the processing content (scenario) indicated by "analysis pattern 1" in the analysis pattern setting information 220 (FIG. 8) of the setting file 200. FIG.

Specifically, first, the image switching detection unit 131 monitors image switching from the "green image", which is the default image. Then, the image switching detection means 131 detects that the default image "green image" of the HD-CS1 has been switched to the color bar image (step S101). Subsequently, the image switching detection unit 131 performs character recognition on the character string blinking in the color bar (step S102), and uses a pre-registered character (here, "FH-46 color bar"). ). Then, the tally output means 141 of the monitoring result output means 14 outputs a green-framed tally image for the resource image (step S103-1), and the audio output means 142 outputs "HD-CS1 receives FH-46 color bar. It has started." is output from the speaker 6 (step S103-2). Further, the monitor screen display means 143 displays "HD-CS1 transmitting" in the image change display area 3200 (FIG. 5) of the monitor screen 3000 displayed on the monitor 5 (step S103-3).
Note that the processes of steps S103-1, S103-2, and S103-3 are executed in parallel. The processes with subnumbers attached to the step numbers are processes that are executed in parallel also in the flowcharts shown below.

Next, the image switching detection means 131 detects that the image of "HD-CS1" has been switched to the default image (green image) after the material transmission (step S104). Then, the tally output unit 141 of the monitoring result output unit 14 stops outputting the tally image in the green frame (turns off the tally in the green frame) (step S105-1), and the audio output unit 142 outputs "End of transmission with HD-CS1." done.” is output from the speaker 6 (step S105-2). In addition, the monitor screen display means 143 displays "HD-CS1 transmission end" in the image change display area 3200 of the monitor screen 3000 displayed on the monitor 5 (step S105-3).

Although not shown, the log output means 144 of the monitoring result output means 14 outputs the information of "HD-CS1 transmission start" and "HD-CS1 transmission end" at the respective dates and times in steps S103 and S105. At the same time, it is stored in the storage means 20 as log information and displayed in the log display area 3300 of the monitoring screen 3000 (FIG. 5).

By doing so, the image monitoring apparatus 1 can make the operator recognize the start and end of the transmission of the HD-CS1 with certainty by the processing of the analysis pattern 1. FIG.

<Analysis pattern 2>
FIG. 11 is a flow chart showing the processing flow of analysis pattern 2. As shown in FIG. An example “analysis pattern 2” in which a resource video whose screen position is “Pic16” and whose resource name is “report IP1” is analyzed will be described.

Here, in the initial setting of steps S1 to S3 in FIG. 9, the analysis pattern selection means 12 of the video monitoring device 1 extracts "news IP1" as the character information displayed on the child screen of "Pic16" and sets it. Assume that the analysis pattern 2 is selected by referring to the initial setting information 210 in the file 200 and specifying the “news IP” in the character recognition information 213 .

When the analysis processing is started, the analysis processing means 13 of the video monitoring apparatus 1 executes the processing content (scenario) indicated by “analysis pattern 2” of the analysis pattern setting information 220 (FIG. 8) of the setting file 200. FIG.

Specifically, first, the image switching detection means 131 captures the image displayed on the screen of "Report IP1" and compares the images before and after to determine whether or not the image has been switched (step S201). Note that the default image of "News IP1" is a "freeze image". When the image switching detection unit 131 determines that the image has not been switched (step S201→No), it ends the processing and proceeds to the analysis processing of the child screen in the next order. On the other hand, when it is determined (detected) that the image has been switched (step S201→Yes), the tally output means 141 of the monitoring result output means 14 outputs a tally image framed in green for the resource image (step S202- 1) The voice output means 142 outputs a voice saying "transmission has started with news report IP1" from the speaker 6 (step S202-2). In addition, the monitor screen display means 143 displays "News IP1 is being transmitted" in the image change display area 3200 of the monitor screen 3000 (FIG. 5) displayed on the monitor 5 (step S202-3).

Subsequently, the freeze detection means 132 captures the images displayed on the screen of "News IP1" and compares them before and after to determine whether the images before and after are the same (step S203). Then, when the freeze detection unit 132 determines that the images before and after are not the same (step S203→No), the process is terminated (the elapsed time count (n) in the next step is set to 0).

On the other hand, in step S203, if the freeze detection unit 132 determines that the images before and after are the same (step S203→Yes), that is, if it detects that it is a "freeze image", The elapsed time is counted (n) (step S204).

Then, the tally output unit 141 continues outputting the green-framed tally image (step S205) when "count (n) ≤ k" (for example, k = 10 seconds), and ends the process. If "count (n)>k", the tally output unit 141 outputs a tally image with a yellow frame (step S206), and ends the process. Then, when "count (n)>m" (m>k, for example, m=60 seconds), the tally output means 141 stops outputting the yellow-framed tally image (turns off) (step S207), End the process.

Although illustration is omitted, the log output unit 144 of the monitoring result output unit 14 outputs the information of "report IP1 transmission start" and "report IP1 transmission end" together with the date and time in steps S202 and S207. It is stored in the storage means 20 as log information and displayed in the log display area 3300 of the monitoring screen 3000 (FIG. 5).

By doing so, the operator can reliably recognize the start and end of the transmission of the report IP1 by the processing of the analysis pattern 2. FIG.

<Analysis pattern 3>
FIG. 12 is a flow chart showing the flow of the analysis pattern 3 process. An example “analysis pattern 3” in which the screen position is “Pic13” and the resource name is “Kintai Bridge camera” is analyzed will be described.

Here, in the initial setting of steps S1 to S3 in FIG. 9, the analysis pattern selection means 12 of the image monitoring device 1 extracts "Kintaikyo camera" as the character information displayed on the child screen of "Pic13". , the initial setting information 210 in the setting file 200 is referred to, and the analysis pattern 3 is selected by specifying “Kintaikyo camera” in the character recognition information 213 .

The analysis processing means 13 of the video monitoring apparatus 1 detects freeze as the processing content (scenario) indicated by "analysis pattern 3" in the analysis pattern setting information 220 (FIG. 8) of the setting file 200 (step S310: see FIG. 13). , whiteout detection (step S320: see FIG. 14), and positional deviation detection (step S330: see FIG. 15) are executed. Note that the default image of the "Kintaikyo camera" is a weather camera image.

(freeze detection)
FIG. 13 is a flowchart for explaining a specific example of freeze detection in analysis pattern 3. FIG.
First, the freeze detection unit 132 captures the images displayed in the resource video of the "Kintaikyo camera", compares them before and after, and determines whether the before and after images are the same (step S311). Then, when the freeze detection unit 132 determines that the images before and after are not the same (step S311→No), the process is terminated (the elapsed time count (n) in the next step is set to 0). At this time, the freeze has been detected before, and if the freeze detection is canceled after that, the tally in the red frame is extinguished (step S312), and the process ends. On the other hand, when the freeze detection means 132 determines that the images before and after are the same (step S311→Yes), that is, when it detects that it is a "freeze image", the elapsed time from the time of detection is counted. (n) (step S313).

Then, the freeze detection means 132 determines whether "count (n) > k" (for example, k = 30 minutes), that is, whether the elapsed time count (n) exceeds a predetermined time k ( step S314). If the count (n) does not exceed the predetermined time k (step S314→No), the freeze detecting means 132 terminates the process. On the other hand, when the freeze detection means 132 determines that the count (n) exceeds the predetermined time k (step S314→Yes), the monitoring result output means 14 executes the following process.

The tally output means 141 outputs a tally image framed in red for the resource image (step S315-1), and the audio output means 142 outputs an alarm sound (an alarm sound indicating an abnormality) from the speaker 6 (step S315-1). 2). In addition, the monitor screen display means 143 displays "Kintai Bridge Camera Freeze Detection" in the image change display area 3200 of the monitor screen 3000 (FIG. 5) displayed on the monitor 5 (step S315-3).

(Brightness detection)
FIG. 14 is a flowchart for explaining a specific example of whiteout detection in analysis pattern 3. In FIG.
First, the overexposure detecting means 133 calculates the luminance level of the image displayed on the “Kintaikyo camera” and determines whether or not it exceeds a set predetermined threshold value, that is, whether overexposure occurs. (step S321). Then, when the overexposure detection unit 133 determines that the brightness of the image does not exceed the predetermined threshold value, that is, the overexposure is not detected (step S321→No), the process is terminated (proceeding to the next step). time count (n) = 0). At this time, whiteout was previously detected, and when the whiteout detection is canceled after that, the red-framed tally is extinguished (step S322), and the process ends. On the other hand, if the whiteout detection unit 133 determines (detects) that the brightness of the image exceeds the predetermined threshold value, that is, that whiteout has occurred (step S321→Yes), the elapsed time from the time of detection The time is counted (n) (step S323).

Then, the whiteout detection means 133 determines whether "count (n)>p" (for example, p=15 minutes), that is, whether the elapsed time count (n) exceeds the predetermined time p. (Step S324). If the count (n) does not exceed the predetermined time p (step S324→No), the whiteout detecting means 133 ends the process. On the other hand, when the whiteout detection means 133 determines that the count (n) exceeds the predetermined time p (step S324→Yes), the monitoring result output means 14 executes the following process.

The tally output means 141 outputs a red-framed tally image for the resource image (step S325-1), and the audio output means 142 outputs an alarm sound (an alarm sound indicating an abnormality) from the speaker 6 (step S325-1). 2). In addition, the monitor screen display means 143 displays "Kintaikyo camera overexposure detected" in the image change display area 3200 of the monitor screen 3000 (FIG. 5) displayed on the monitor 5 (step S325-3).

(Position deviation detection)
FIG. 15 is a flowchart for explaining a specific example of detection of positional deviation in the analysis pattern 3. As shown in FIG.
First, the positional deviation detection means 135 determines whether or not there is a positional deviation between the image displayed in the "Kintai Bridge camera" and the image at the default position and the current image by detecting a feature amount. (Step S331). It should be noted that the positional deviation detection means 135 registers in advance a plurality of images such as a daytime image, a nighttime image, and other images at a predetermined time (for example, a specific default image during an election period, etc.) as default images. Keep Then, if the degree of similarity with any one of the plurality of images is high, the positional deviation detection unit 135 may determine that the position is the default position.
If it is determined in step S331 that the positional deviation detection means 135 has not deviated from the default position by more than a predetermined amount, that is, if it is determined that there is no positional deviation (step S331→No), the process ends (the Elapsed time count (n) = 0). At this time, if the positional deviation has been detected before and the positional deviation is no longer detected, the red-framed tally is extinguished (step S332), and the process ends. On the other hand, if the positional deviation detection means 135 determines (detects) that the positional deviation exceeds the predetermined amount from the default position (step S331→Yes), the elapsed time from the time of detection is counted (n). (step S333).

Then, the positional deviation detection means 135 determines whether "count (n)>m" (for example, m=30 minutes), that is, whether the elapsed time count (n) exceeds the predetermined time m. (Step S334). If the count (n) does not exceed the predetermined time m (step S334→No), the positional deviation detection means 135 ends the process. On the other hand, when the positional deviation detection means 135 determines that the count (n) exceeds the predetermined time m (step S334→Yes), the monitoring result output means 14 executes the following processing.

The tally output means 141 outputs a red-framed tally image for the resource image (step S335-1), and the audio output means 142 outputs an alarm sound (an alarm sound indicating an abnormality) from the speaker 6 (step S335-1). 2). In addition, the monitoring screen display means 143 displays "Kintaikyo camera position deviation detection" in the image change display area 3200 of the monitoring screen 3000 (FIG. 5) displayed on the monitor 5 (step S335-3).

Although illustration is omitted, the log output means 144 of the monitoring result output means 14 outputs "Kintaikyo camera Freeze detection", "Kintaikyo camera overexposure detection", and "Kintaikyo camera misalignment detection" along with the date and time are stored as log information in the storage means 20, and the log of the monitoring screen 3000 (Fig. 5) is stored. It is displayed in the display area 3300 .

By doing so, the processing of the analysis pattern 3 allows the operator to reliably recognize freezes, overexposure, and misalignment of the image of the “Kintaikyo camera”.

<Analysis pattern 4>
FIG. 16 is a flow chart showing the flow of the analysis pattern 4 processing. An example “analysis pattern 4” in which a resource video whose screen position is “Pic1” and whose resource name is “outside station 1” is analyzed will be described.

Here, in the initial setting of steps S1 to S3 in FIG. 9, the analysis pattern selection means 12 of the video monitoring device 1 extracts "outside station 1" as character information displayed on the child screen of "Pic1", It is assumed that analysis pattern 4 is selected by referring to initial setting information 210 in setting file 200 and detecting "outside station 1" in character recognition information 213 .

When the analysis processing is started, the analysis processing means 13 of the video monitoring apparatus 1 executes the processing content (scenario) indicated by "analysis pattern 4" in the analysis pattern setting information 220 (FIG. 8) of the setting file 200. FIG.

Specifically, first, the image switching detection unit 131 captures an image displayed on the screen of "outside station 1" and compares the image before and after to determine whether or not the image has been switched (step S401). The default image of "outside station 1" is the "green image". When the image switching detection unit 131 determines that the image has not been switched (step S401→No), it ends the processing and shifts to the analysis processing of the child screen in the next order. On the other hand, if it is determined that the image has switched (step S401→Yes), the analysis processing means 13 clicks the "process start" button (for example, the 5, 3002) is blinked. This prompts the operator to select (click) the "start processing" button. Then, the analysis processing unit 13 acquires input information indicating that the "start processing" button has been selected (step S402). At this time, the tally output means 141 outputs a green-framed tally image for the resource image (step S403). As a result, it is shown on the multi-screen that the resource video of "outside station 1" is being analyzed.

It should be noted that the analysis processing means 13 is operated when, for example, a "processing stop" button (eg, reference numeral 3003 in FIG. 5) arranged on the monitoring screen 3000 is selected during analysis processing, or an abnormality is detected during subsequent analysis processing. First, when it is detected that the resource image has been switched to the "green image" which is the default image of "outside station 1", the tally output means 141 stops outputting the tally image of the green frame and turns off the light.

Then, the analysis processing means 13 of the image monitoring apparatus 1 executes freeze detection (step S410: see FIG. 17) and silence detection (step S420: see FIG. 18).

(freeze detection)
FIG. 17 is a flowchart for explaining a specific example of freeze detection in analysis pattern 4. FIG.
The freeze detection means 132 captures the image displayed on the screen of "outside station 1" and compares the images before and after to determine whether the images before and after are the same (step S411). Then, when the freeze detection unit 132 determines that the images before and after are not the same (step S411→No), the process is terminated (the elapsed time count (n) in the next step is set to 0). At this time, if the freeze has been detected before and the freeze detection is canceled after that, instead of the red-framed tally image, the green-framed tally image is output (step S412), and the process is terminated. On the other hand, when the freeze detection means 132 determines that the images before and after are the same (step S411→Yes), that is, when it detects that it is a “freeze image”, it counts the elapsed time from the time of detection. (n) (step S413).

Then, the freeze detection means 132 determines whether "count (n) > k" (for example, k = 5 seconds), that is, whether the elapsed time count (n) exceeds a predetermined time k ( step S414). If the count (n) does not exceed the predetermined time k (step S414→No), the freeze detection means 132 terminates the process. On the other hand, when the freeze detection means 132 determines that the count (n) exceeds the predetermined time k (step S414→Yes), the monitoring result output means 14 executes the following process.

The tally output means 141 outputs a red-framed tally image for the resource image (step S415-1), and the audio output means 142 outputs an alarm sound (an alarm sound indicating an abnormality) from the speaker 6 (step S415-1). 2). In addition, the monitor screen display means 143 displays "outside station 1 freeze detection" in the image change display area 3200 of the monitor screen 3000 (FIG. 5) displayed on the monitor 5 (step S415-3).

(Silence detection)
FIG. 18 is a flowchart for explaining a specific example of silence detection in analysis pattern 4. In FIG.
The silence detection unit 134 monitors the audio level meter displayed in the resource video of "outside station 1" and determines whether the audio meter image is during silence (step S421). Then, when the silence detection unit 134 determines that the image is not the sound meter image during silence (step S421→No), the processing is terminated (the elapsed time count (n) in the next step is set to 0). At this time, silence has been detected before, and when the silence detection is canceled after that, instead of the red-framed tally image, the green-framed tally image is output (step S422), and the process ends. On the other hand, when the silence detecting means 134 determines (detects) that the audio meter image is silent (step S421→Yes), the elapsed time from the time of detection is counted (n) (step S423).

Then, the silence detection means 134 determines whether "count (n)>m" (for example, m=5 seconds), that is, whether the elapsed time count (n) exceeds the predetermined time m ( step S424). If the count (n) does not exceed the predetermined time m (step S424→No), the silence detection means 134 ends the process. On the other hand, when the silence detection means 134 determines that the count (n) exceeds the predetermined time m (step S424→Yes), the monitoring result output means 14 executes the following process.

The tally output means 141 outputs a tally image framed in red for the resource image (step S425-1), and the audio output means 142 outputs an alarm sound (an alarm sound indicating an abnormality) from the speaker 6 (step S425-1). 2). In addition, the monitor screen display means 143 displays "outside station 1 silence detected" in the image change display area 3200 of the monitor screen 3000 (FIG. 5) displayed on the monitor 5 (step S425-3).

Although illustration is omitted, the log output means 144 of the monitoring result output means 14 performs "outside station 1 freeze detection" and "outside station 1 freeze detection" in step S415 of FIG. 17 and step S425 of FIG. 1 silence detection” is stored as log information in the storage means 20 together with its date and time, and is displayed in the log display area 3300 of the monitoring screen 3000 (FIG. 5).

In addition, in the example of analysis pattern 4, the "process start" and "process stop" of the analysis process are set, for example, by the "process start" button (code 3002) or the "process stop" button (code 3003) accepts and executes instruction information from the operator. By doing so, the analysis processing can be started or stopped at an arbitrary timing desired by the operator. In addition, among the multi-screens shown in FIG. 2, for example, when the resource video of “Pic25” “Report IP8” is changed to the resource video of “Temporary FPU”, the analysis processing of “Temporary FPU” For example, a "processing start" button (reference numeral 3004) and a "processing stop" button (reference numeral 3005) may be provided on the monitoring screen 3000 (FIG. 5) to accept and execute instruction information from the operator.

By doing so, the processing of the analysis pattern 4 makes it possible for the operator to reliably recognize the abnormality of the resource video freeze and silence of the outside station 1 .

As described above, according to the image monitoring apparatus 1 and the image monitoring program according to the present embodiment, it is possible to detect a change in the state of the multi-screen and reliably notify the operator of the change. In addition, it is possible to detect abnormalities in the resource video such as freezing and blown-out highlights, and to reliably notify the operator.

1 video monitoring device 5 monitor 6 speaker 10 control means 11 video reading means 12 analysis pattern selection means 13 analysis processing means 14 monitoring result output means 15 processing setting means 20 storage means 30 input/output means 131 image switching detection means 132 freeze detection means 133 Whiteout detection means 134 Silence detection means 135 Positional deviation detection means 141 Tally output means 142 Audio output means 143 Monitoring screen display means 144 Log output means 200 Setting file 210 Initial setting information 220 Analysis pattern setting information 1000 Input image 2000 Output image 3000 Monitoring Screen 3100 Multi-screen area 3200 Image change display area 3300 Log display area

Claims (6)

A video monitoring device for monitoring each resource video on a multi-screen displaying a plurality of resource videos on one screen,
storage means for storing a setting file that stores a resource name of the resource video and an analysis pattern indicating the content of analysis processing according to the type of the resource video in association with each screen position in the multi-screen; ,
video reading means for reading an input image in which the resource name of the resource video is displayed in each of the resource videos on the multi-screen;
dividing the input image into child screens for each of the resource videos, extracting the resource names displayed on the child screens by character recognition, referring to the setting file, and analyzing patterns corresponding to the extracted resource names. analysis pattern selection means for selecting
analysis processing means for executing analysis processing indicated by the selected analysis pattern for each resource video;
monitoring result output means for outputting result information regarding the state change of each of the resource videos detected by the analysis process;
A video monitoring device comprising: The content of analysis processing indicated by the analysis pattern includes detection of image switching of the resource video, detection of freezing of the resource video, detection of overexposure of the resource video, and detection of displacement of the resource video from a default position. , and detection of silence in the resource video. 3. The video monitoring according to claim 1, wherein, when a state change of the resource video is detected by the analysis processing, the monitoring result output means outputs audio information indicating the state change. Device. When a state change of the resource video is detected by the analysis processing, the monitoring result output means displays an output image including a tally video at a screen position in the multi-screen of the resource video where the state change is detected. 4. The video monitoring apparatus according to any one of claims 1 to 3, wherein the video monitoring apparatus outputs When the state change of the resource video is detected by the analysis processing, the monitoring result output means displays a multi-screen area including the tally video and the resource video for which the state change is detected by the analysis processing. 5. The video monitoring apparatus according to claim 4, generating and outputting a monitoring screen including a resource name of and an image change display area for displaying result information regarding a state change of the resource video. A video monitoring program for causing a computer to function as the video monitoring device according to any one of claims 1 to 5.

Images