Video comparison and monitoring device and its program

Abstract

The present invention provides a video comparison and monitoring device that can restore synchronization and perform video comparison and monitoring even when synchronization is lost between the active system and the backup system. [Solution] The video comparison monitoring device 10 includes a delay unit 11 that sequentially stores and outputs the video signal of the active system, a delay unit 12 that sequentially stores and outputs the video signal of the backup system, a difference evaluation value calculation unit 13 that calculates the difference evaluation value at the pixel level between the active system frame and the backup system frame, a fault detection unit 14 that detects faults from the distribution of occurrence of difference evaluation values ​​and detects a synchronization outage when the duration of the fault has elapsed to a predetermined time length, and a synchronization control unit 15 that synchronizes the active system frame and the backup system frame when a synchronization outage is detected.

Description

Technical Field

[0001] The present invention relates to a video comparison monitoring device and its program.

Background Art

[0002] Currently, when relaying video, in many cases, it is operated by a video relay system 100 composed of an active system and a standby system as shown in FIG. 8 (see Patent Document 1). The active system is a main system for transmitting video used as a broadcast service. The standby system is a backup system for switching the video used as a broadcast service when a failure occurs in the active system.

[0003] For example, in the active system, the video captured by camera 2 at the relay source (location A) is satellite-relayed by relay vehicle 3. Then, the active system receives the satellite-relayed video at the broadcast station (location B) with receiving device 4, and uses the video received by receiving equipment 7 in the broadcast station as the video for the broadcast service. In the standby system, the video captured by camera 2 at the relay source (location A) is wirelessly relayed by microwave transmission (FPU) with relay transmission device 5. Then, the standby system receives the wirelessly relayed video at the broadcast station (location B) with relay transmission device 6, and uses the video received by receiving equipment 7 in the broadcast station as the backup video when a failure occurs in the active system.

[0004] Conventionally, the switching between the active system and the standby system in the receiving equipment 7 of the video relay system 100 is performed by an operator O visually observing monitor M connected to the active system C and monitor M connected to the standby system S as shown in FIG. 9. [[ID=第32]]That is, the operator O identifies the faulty system by visually observing the monitor (S100), and determines whether to switch between the active system and the standby system (S101). When a failure occurs in the active system and it is determined that switching is necessary, the operator O performs a switching operation between the active system and the standby system in the receiving equipment 7 (S102).

[0005] On the other hand, there are FR (Full Reference) type evaluation devices that compare and evaluate two images on a pixel-by-pixel basis (see Patent Document 2). If this conventional evaluation device-based video comparison monitoring method could be applied to comparing the active and backup systems in the video relay system 100, it would eliminate the need for operators to detect failures. However, currently, there are no examples of applying the video comparison monitoring method to the video relay system 100. [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 2021-170700 [Patent Document 2] Japanese Patent Application Publication No. 8-205156 [Overview of the Initiative] [Problems that the invention aims to solve]

[0007] In video relay systems, the amount of latency may change during operation. Conventional video comparison monitoring methods have a problem where, if synchronization is lost due to changes in the delay amount between the active and backup systems, it becomes impossible to compare the active and backup systems, and the monitoring function cannot be continued. Therefore, with conventional video comparison monitoring methods, even if an FR-type evaluation device is used, the operator must constantly monitor the system.

[0008] Therefore, the object of the present invention is to provide a video comparison and monitoring device and program that can restore synchronization and perform video comparison and monitoring even when synchronization is lost between the active system and the backup system. [Means for solving the problem]

[0009] To solve the aforementioned problems, the video comparison monitoring device according to the present invention is a video comparison monitoring device that compares the video signal of the active system and the video signal of the backup system in a video relay system and monitors for faults, and comprises an active system video delay unit, a backup system video delay unit, a difference evaluation value calculation unit, a fault detection unit, and a synchronization control unit.

[0010] In this configuration, the video comparison and monitoring device sequentially stores and outputs the video signal of the active system using the active system video delay unit. Furthermore, the video comparison and monitoring device sequentially stores and outputs the video signal of the backup system using the backup system video delay unit.

[0011] The video comparison and monitoring device then uses a difference evaluation value calculation unit to calculate the pixel-level difference evaluation value between the active system frame, which is a frame-level video signal output from the active system video delay unit, and the backup system frame, which is a frame-level video signal output from the backup system video delay unit. The video comparison and monitoring device then uses a fault detection unit to detect faults from the distribution of differential evaluation values ​​at the pixel level, and when the duration of the fault has exceeded a predetermined time, it detects a loss of synchronization between the active system frame and the backup system frame.

[0012] The video comparison and monitoring device, via its synchronization control unit, resynchronizes the active and backup frames when the fault detection unit detects a synchronization issue. The synchronization control unit also performs initial synchronization between the active and backup frames upon startup. This allows the video comparison and monitoring device to quickly resynchronize when it detects a synchronization issue. Furthermore, the video comparison and monitoring device can be operated using a program that enables the computer to function as one of the aforementioned components. [Effects of the Invention]

[0013] According to the present invention, even if synchronization is lost between the active system and the backup system, synchronization can be restored and video comparison monitoring can be continued without the operator having to constantly monitor the system. [Brief explanation of the drawing]

[0014] [Figure 1] It is a system configuration diagram for explaining the outline of a video relay system including a video comparison monitoring device according to an embodiment of the present invention. [Figure 2] It is a block configuration diagram showing the configuration of a video comparison monitoring device according to an embodiment of the present invention. [Figure 3] It is an explanatory diagram for explaining a method of calculating a differential evaluation value in a differential evaluation value calculation unit. [Figure 4] It is an explanatory diagram for explaining the relationship between the occurrence distribution of the differential evaluation value and a failure, and the criteria for failure determination. [Figure 5] It is a graph showing the relationship between the frame unit MSE and the maximum pixel unit MSE and a failure. [Figure 6] It is an explanatory diagram for explaining a general example of the type of quality abnormality and its duration. [Figure 7] It is a flowchart showing the operation of a video comparison monitoring device according to an embodiment of the present invention. [Figure 8] It is a system configuration diagram showing the configuration of a general video relay system. [Figure 9] It is an explanatory diagram for explaining a monitoring method of a conventional video relay system. [Embodiments for Carrying Out the Invention]

[0015] Hereinafter, embodiments of the present invention will be described with reference to the drawings. [Outline of Video Relay System] First, referring to FIG. 1, the outline of a video relay system 1 including a video comparison monitoring device 10 according to an embodiment of the present invention will be described.

[0016] The video relay system 1 relays video at two locations (Location A, Location B) and monitors for failures in the video signal on the transmission path. The transmission path of the video relay system 1 consists of an active system and a backup system. Since this two-system video relay configuration is the same as the conventional configuration (Patent Document 1) described in Figure 8, a detailed explanation is omitted. The video relay system 1 is equipped with a video comparison and monitoring device 10 at location B (broadcasting station), which is the final relay point.

[0017] The video comparison and monitoring device 10 compares the video signal of the active system (hereinafter referred to as the active system video) and the video signal of the backup system (hereinafter referred to as the backup system video) to monitor for malfunctions. The active system video and the backup system video are, for example, 4K video, 8K video, etc. If the video comparison monitoring device 10 detects a malfunction, it will issue an alarm to notify the operator O of the malfunction. Operator O recognizes the malfunction through an alarm and switches the broadcasting service from the active system to the backup system at the receiving equipment 7 within the broadcasting station (S1). At this time, if the synchronization between the active system video and the backup system video is lost, the video comparison and monitoring device 10 restores the synchronization.

[0018] As a result, the video relay system 1 does not require constant monitoring by the operator; the operator only needs to switch from the active system to the backup system as needed when an alarm is triggered. The configuration and operation of the video comparison and monitoring device 10 will be described below.

[0019] [Configuration of the video comparison and monitoring system] Referring to Figure 2, the configuration of the video comparison and monitoring device 10 according to an embodiment of the present invention will be described. The video comparison monitoring device 10 includes delay units 11 and 12, a difference evaluation value calculation unit 13, a fault detection unit 14, a synchronization control unit 15, and an alarm notification unit 16.

[0020] The delay unit (active system video delay unit) 11 sequentially stores and outputs the active system video input from an external source. The delay unit 11 functions as a buffer that stores a predetermined number of frames. When the delay unit 11 receives a delay instruction from the synchronization control unit 15, it delays the output by the instructed number of frames.

[0021] The delay unit (backup video delay unit) 12 sequentially stores and outputs backup video input from an external source. The delay unit 12 functions as a buffer that stores a predetermined number of frames. When the delay unit 12 receives a delay instruction from the synchronization control unit 15, it delays the output by the instructed number of frames.

[0022] The difference evaluation value calculation unit 13 calculates the difference evaluation value for each pixel between the active frame, which is a frame-by-frame video signal output from the delay unit 11, and the backup frame, which is a frame-by-frame video signal output from the delay unit 12. The difference evaluation value calculation unit 13 calculates the difference (error) for each pixel between the active frame and the backup frame. Then, for each pixel, the difference evaluation value calculation unit 13 calculates the mean squared error (MSE) within a predetermined local region including that pixel as the difference evaluation value (pixel-level MSE) for that pixel. Furthermore, the difference evaluation value calculation unit 13 calculates a frame-level difference evaluation value (frame-level MSE) by averaging the difference evaluation values ​​for each pixel within the frame.

[0023] Now, with reference to Figure 3, the method for calculating the difference evaluation value in the difference evaluation value calculation unit 13 will be explained. As shown in Figure 3, the difference evaluation value calculation unit 13 is used for the current system frame F C and spare frame F S The following is entered. Here, the current system frame F C and spare frame F S Let the image size be X×Y pixels, and the size of the local region R be N×N pixels. Also, the current system frame F C and spare frame F S Let d(x,y) be the difference value for each pixel position (x,y) (1≦x≦X-N+1, 1≦y≦Y-N+1).

[0024] The difference evaluation value calculation unit 13 calculates the pixel unit MSE(x,y) for the pixel position (x,y) by averaging the difference values ​​in the local region R that includes the pixel position (x,y), as shown in equation (1) below.

[0025]

number

[0026] Furthermore, the difference evaluation value calculation unit 13 calculates the frame-level MSE by averaging the pixel-level MSE within the frame, as shown in equation (2) below.

[0027]

number

[0028] Returning to Figure 1, we will continue the explanation of the configuration of the video comparison and monitoring device 10. The difference evaluation value calculation unit 13 outputs a difference evaluation value (pixel-level MSE and frame-level MSE) for each frame to the fault detection unit 14.

[0029] The fault detection unit 14 detects faults from the distribution of differential evaluation values ​​at the pixel level, and when the duration of the fault has elapsed to a predetermined time, it detects a loss of synchronization between the active system frame and the backup system frame.

[0030] Here, referring to Figure 4, we will explain the relationship between the occurrence distribution of pixel-level differential evaluation values ​​(pixel-level MSE) for each frame and defects, as well as the criteria for defect detection. Figure 4(a) is a graph showing the relationship between the pixel-level MSE and its incidence rate under normal conditions where no defects occur.

[0031] Under normal conditions, as shown in Figure 4(a), the frame-level MSE values ​​are generally distributed normally, with the center of the normal distribution (A1). In this case, pixel-level MSE values ​​that deviate from A1 beyond a predetermined threshold W1 can be considered outliers in the distribution caused by quality abnormalities such as block noise. In other words, if the maximum MSE per pixel does not exceed the threshold W1, then there are no quality abnormalities within the frame, and it can be determined to be normal.

[0032] Figure 4(b) is a graph showing the relationship between pixel-level MSE and its occurrence rate in critical (high encoding difficulty) video where no failures occurred. In this case, no malfunction has occurred and the system is functioning normally. However, due to strong compression noise, the center of the normal distribution (A1) shifts to a higher pixel-level MSE value (A2) (A1→A2). Therefore, the threshold W1 for outliers in the distribution caused by quality anomalies such as block noise also needs to be set to a larger threshold W2.

[0033] Figure 4(c) is a graph showing the relationship between pixel-level MSE and its incidence rate when a localized defect occurs. A localized defect is a defect that occurs in a part of the frame, such as a localized scratch or noise. In this case, for a normal distribution shown in Figure 4(a), pixel-level MSEs occur that are outliers OL in the distribution, exceeding the threshold W1 as shown in Figure 4(c). In other words, the occurrence of local damage can be determined based on threshold W1 (or threshold W2 in the case of critical images).

[0034] Figure 4(d) is a graph showing the relationship between pixel-level MSE and its incidence rate when a frame defect occurs. A frame defect is a defect that occurs throughout the entire frame, such as momentary freezes, black frames, and color shifts (phase shifts). In this case, as shown in Fig. 4(d), there are no outliers in the distribution, but the center of the normal distribution has a value larger than A1 and A2. That is, the occurrence of a frame failure can be determined based on whether the value of the frame unit MSE, which is the center of the normal distribution, is greater than a predetermined value A3 that is larger than A1 and A2.

[0035] Fig. 4(e) is a graph showing the relationship between the pixel unit MSE and its occurrence rate when synchronization deviation occurs between the active frame and the standby frame. In this case, as shown in Fig. 4(e), the pixel unit MSE increases and an outlier OL exceeding the threshold value W1 occurs. Due to the occurrence of this outlier OL, in the case of synchronization deviation, a region that may be confused with the local failure shown in Fig. 4(c) will occur.

[0036] Fig. 5 shows the classification of the above four types of distributions (normal (a), (b), local failure (c), frame failure (d), synchronization deviation (e)). Fig. 5 is a graph showing the relationship between the frame unit MSE and the maximum pixel unit MSE and the failure. The horizontal axis X represents the frame unit MSE, and the vertical axis Y represents the maximum pixel unit MSE (maximum pixel unit MSE) within the frame.

[0037] As shown in Fig. 5, the region R1 of the upward-sloping stripe indicates the normal range, and the region R2 of the downward-sloping stripe indicates the range of frame failure. Also, the region R3 of the vertical stripe indicates the range of local failure, and the region R4 of the horizontal stripe indicates the range of synchronization deviation. That is, the region R1 (normal) is in the range where when X ≤ A1, Y ≤ W1; when A1 < X ≤ A2, Y ≤ a1X + b1 (Y is the straight line connecting (A1, W1) and (A2, W2)); when A2 < X ≤ A3, Y ≤ a2X + b2 (Y is the straight line passing through (A2, W2) with a predetermined slope a2). Also, the region R2 (frame failure) is in the range where X > A3 and Y ≤ a2X + b2.

[0038] The values ​​for A1, A2, A3, W1, W2, a1, a2, b1, and b2 should be set to values ​​optimized for the system. For example, A1=260, A2=2600, A3=3300, W1=52000, W2=520000, a1=151.7, b1=125528, a2=0.862, b2=517759.

[0039] Region R3 (local fault) and region R4 (out-of-sync) overlap within the same frame due to the presence of outlier OL in the distribution shown in Figure 4, as shown in Figure 5. Therefore, faults cannot be determined from a frame alone.

[0040] Therefore, the fault detection unit 14 measures the duration of the fault, which is determined by the difference evaluation value (pixel-level MSE and frame-level MSE) calculated for each frame by the difference evaluation value calculation unit 13, and determines whether it is a fault or a synchronization outage. Here, the number of frames is measured as the duration.

[0041] Here, with reference to Figure 6, we will describe some common examples of quality anomalies and their durations. For example, localized defects such as block noise / scratches may occur in only one frame (the first frame), but the defect may persist for more than a second. Furthermore, frame errors such as momentary freezes can cause visual discomfort if they continue for several frames, while freezes where the video motion stops or black screens where the entire video turns black can range from being perceived as a mere moment to lasting for tens of seconds, potentially leading to serious problems. Additionally, frame defects such as the entire video turning green or color distortions gradually degrade in quality and often persist for more than a minute.

[0042] Based on the above, the fault detection unit 14 sets T1, T2, and T3 as duration parameters shown in Figure 6 to detect quality abnormalities as quickly as possible without causing false detections. Furthermore, once a synchronization outage begins, it will not recover naturally unless resynchronization is performed. Therefore, T4 is set as a duration parameter to initiate resynchronization as quickly as possible without interfering with the alarm activation for the various quality abnormalities T1, T2, and T3.

[0043] For example, let's set T1=1, T2=6, T3=45, and T4=90 (frames). T1 is the first duration (frame) at which a local fault is detected. T2 is a second duration (frame) used to detect frame failures that last for a relatively short time. T3 is a third duration (frame) used to detect frame failures that are determined to be severe after a certain period of time. T4 is the fourth duration (frame) at which out-of-sync is detected. Furthermore, if the difference evaluation value calculated by the difference evaluation value calculation unit 13 after the elapsed time of each duration parameter is within the normal range, the failure detection unit 14 resets the elapsed time of the parameter.

[0044] Returning to Figure 2, we will continue our explanation of the configuration of the video comparison and monitoring device 10. After detecting a fault, the fault detection unit 14 performs the following processing according to the elapsed time parameters T1, T2, T3, and T4 as described in Figure 6.

[0045] (T1 progression) The fault detection unit 14 notifies the alarm issuing unit 16 of the occurrence of a local fault when a state other than normal, as described in Figure 5, has elapsed for T1, based on the frame-unit MSE and the maximum pixel-unit MSE. At this time, if the frame-unit MSE and the maximum pixel-unit MSE fall within the range of the frame fault described in Figure 5, the fault detection unit 14 notifies the alarm issuing unit 16 of the frame fault along with the occurrence of the fault. Otherwise, the fault detection unit 14 notifies the alarm issuing unit 16 of the local fault along with the occurrence of the fault.

[0046] (T2 progression) The fault detection unit 14, based on the frame-unit MSE and the maximum pixel-unit MSE, notifies the alarm generation unit 16 of the occurrence of a frame fault, such as a freeze or black screen, when a frame fault, which is an abnormal state as described in Figure 5, has elapsed for T2.

[0047] (T3 progression) The fault detection unit 14, based on the frame-unit MSE and the maximum pixel-unit MSE, notifies the alarm generation unit 16 of the occurrence of a serious fault, such as a green image or color abnormality, when a frame fault, which is an abnormal state as described in Figure 5, has elapsed for T3. The fault notified to the alarm generation unit 16 is the same as the one notified at T1.

[0048] (T4 stage) The fault detection unit 14 determines that synchronization has been lost when the normal and non-frame fault states described in Figure 5 have elapsed for T4, and notifies the alarm generation unit 16 of the occurrence of synchronization loss. If the fault detection unit 14 determines that synchronization has been lost, it instructs the synchronization control unit 15 to resynchronize. The fault detection unit 14 also remains on standby without monitoring for the presence of a fault until synchronization is restored. The fault detection unit 14 resumes monitoring when it receives notification from the synchronization control unit 15 that synchronization has been restored. This allows the fault detection unit 14 to suppress the activation of unnecessary alarms until synchronization is restored.

[0049] The synchronization control unit 15 performs initial synchronization between the active system frame and the backup system frame at startup, and resynchronizes the active system frame and the backup system frame if the fault detection unit 14 detects a synchronization failure. The synchronization control unit 15 performs matching of multiple frames between the active system frames sequentially stored in the delay unit 11 and the backup system frames sequentially stored in the delay unit 12. The synchronization control unit 15 then measures the delay amount between the frames that best match and synchronizes the active system frames and the backup system frames by delaying the frame that arrives earlier by that amount in the delay unit 11 or delay unit 12. Furthermore, if the fault detection unit 14 instructs the synchronization control unit 15 to resynchronize, the synchronization control unit 15 will notify the fault detection unit 14 once synchronization has been restored.

[0050] The alarm issuing unit 16 issues alarms notified by the fault detection unit 14. The alarms issued by the alarm issuing unit 16 are not particularly limited; any type of alarm is acceptable as long as it appropriately notifies the operator, such as by displaying text on an external display device or by sound.

[0051] With the configuration described above, the video comparison and monitoring device 10 can compare and monitor frames between the active and backup systems in the video relay system on a pixel-by-pixel basis and issue an alarm when a failure occurs. Furthermore, since the video comparison and monitoring device 10 can automatically restore synchronization if it is lost, there is no need for the operator to constantly monitor it. This video comparison and monitoring device 10 can be operated using a program that causes a computer (not shown in the figure) to function as one of the aforementioned components.

[0052] [Operation of the video comparison and monitoring device] Next, with reference to Figure 7 (and Figure 2 as appropriate for the configuration), the operation of the video comparison and monitoring device 10 according to an embodiment of the present invention will be described.

[0053] Although not shown in the diagram, it is assumed that the delay units 11 and 12 receive and sequentially store the active and backup system video signals from an external source. Furthermore, at startup, a variable for measuring the number of consecutive frames in which a fault occurred is initialized ("0"), and a variable indicating that a fault occurred in the previous frame is also initialized ("no fault").

[0054] In step S10, the synchronization control unit 15 performs initial synchronization between the active system frame and the backup system frame in the delay units 11 and 12.

[0055] In step S11, the difference evaluation value calculation unit 13 calculates a difference evaluation value from the active system frame and the backup system frame output from the delay units 11 and 12. Here, the difference evaluation value calculation unit 13 calculates the pixel-unit MSE according to equation (1) and the frame-unit MSE according to equation (2) as difference evaluation values ​​from the active system frame and the backup system frame.

[0056] In step S12, the fault detection unit 14 detects a fault based on the differential evaluation value calculated in step S12. Here, the fault detection unit 14 determines that there is a fault if the maximum value of the pixel-level MSE (maximum pixel-level MSE) and the frame-level MSE are outside the normal range shown in Figure 5.

[0057] If no fault is detected here (No in step S12), in step S13, the fault detection unit 14 initializes a variable ("0") for measuring (timing) the number of consecutive frames in which a fault occurred. Then, the operation returns to step S11.

[0058] On the other hand, if a fault is detected (Yes in step S12), in step S14, the fault detection unit 14 increments (+1) a variable for measuring (timing) the number of consecutive frames (duration) in which the fault occurred.

[0059] In step S15, the fault detection unit 14 determines whether or not a fault occurred in the previous frame based on the variable. If no failure occurred in the previous frame (No in step S15), the variable indicating that a failure occurred in the previous frame is set to "failure detected" (not shown as a step), and the operation returns to step S11.

[0060] On the other hand, if a failure occurred in the previous frame (Yes in step S15), in step S16, the failure detection unit 14 determines the duration (number of frames) of the failure and executes processing according to the durations T1, T2, T3, and T4 as explained in Figure 6.

[0061] If the duration of the fault is between T1 and T2, in step S17, the fault detection unit 14 notifies the alarm issuing unit 16 of the occurrence of a local fault, and the alarm issuing unit 16 issues an alarm for the occurrence of a local fault. Then, the operation returns to step S11.

[0062] If the duration of the failure is T2 or greater but less than T3, in step S18, the failure detection unit 14 notifies the alarm issuing unit 16 of the occurrence of a frame failure, and the alarm issuing unit 16 issues an alarm for the occurrence of a frame failure. Then, the operation returns to step S11.

[0063] If the duration of the failure is T3 or greater but less than T4, in step S19, the failure detection unit 14 notifies the alarm issuing unit 16 of the occurrence of a severe failure, causing the alarm issuing unit 16 to issue an alarm for a frame failure (severe failure). Then, the operation returns to step S11.

[0064] If the duration of the failure is T4 or longer, in step S20, the failure detection unit 14 notifies the alarm issuing unit 16 of the occurrence of a synchronization outage, and the alarm issuing unit 16 issues an alarm for the synchronization outage.

[0065] Then, in step S21, the fault detection unit 14 instructs the synchronization control unit 15 to resynchronize the active system frame and the backup system frame, and the synchronization control unit 15 performs the resynchronization. Then, the operation returns to step S11. If the duration of the failure is less than T1 (or for other reasons), the failure detection unit 14 returns to step S11.

[0066] Through the above operations, the video comparison and monitoring device 10 can compare and monitor frames between the active and backup systems in the video relay system on a pixel-by-pixel basis and issue an alarm when a failure occurs. Furthermore, the video comparison and monitoring device 10 can automatically restore synchronization if synchronization is lost.

[0067] Although embodiments of the present invention have been described above, the present invention is not limited to these embodiments and includes design modifications and the like that do not depart from the spirit of the present invention.

[0068] For example, in this case, the video comparison and monitoring device 10 uses the frame-by-frame MSE in equation (2) as the frame-by-frame difference evaluation value. However, the frame-by-frame difference evaluation value is not limited to MSE; other evaluation values ​​may be used. For example, PSNR (Peak Signal to Noise Ratio) may be used as the frame-by-frame difference evaluation value. Note that a smaller MSE value indicates a smaller error, while a larger PSNR value indicates a smaller error. Therefore, when using frame-by-frame PSNR instead of frame-by-frame MSE, the judgment of magnitude should be reversed. [Explanation of Symbols]

[0069] 1. Video relay system 10. Video Comparison and Monitoring Device 11. Delay section (current video delay section) 12. Delay section (backup video delay section) 13 Difference Evaluation Value Calculation Unit 14. Fault detection unit 15 Synchronization Control Unit 16 Alarm Activation Unit

Claims

1. A video comparison and monitoring device that compares the video signal of the active system and the video signal of the backup system in a video relay system and monitors for faults, A current system video delay unit that sequentially stores and outputs the video signals of the current system, A backup system video delay unit that sequentially stores and outputs the video signals of the backup system, A difference evaluation value calculation unit calculates a pixel-level difference evaluation value between the active system frame, which is a frame-level video signal output from the active system video delay unit, and the backup system frame, which is a frame-level video signal output from the backup system video delay unit. A fault detection unit detects a fault from the distribution of the difference evaluation values ​​at the pixel level, and when the duration of the fault has elapsed to a predetermined time, it detects a loss of synchronization between the active system frame and the backup system frame. A synchronization control unit performs initial synchronization between the active system frame and the backup system frame at startup, and resynchronizes the active system frame and the backup system frame when the fault detection unit detects a synchronization failure, A video comparison and monitoring device characterized by being equipped with the following features.

2. The video comparison and monitoring device according to claim 1, further comprising an alarm issuing unit that issues an alarm when the fault detection unit detects a fault.

3. The video comparison and monitoring device according to claim 2, characterized in that the fault detection unit, after detecting a fault, instructs the alarm issuing unit to issue an alarm indicating the continuation of the fault when a predetermined time period has elapsed that is less than the duration for detecting the synchronization out of sync.

4. The difference evaluation value calculation unit calculates the pixel-level MSE, which is the mean squared error within a predetermined local region including the pixel, for each pixel. The video comparison and monitoring device according to claim 1, characterized in that the fault detection unit detects faults from outliers when the distribution of occurrence of pixel-unit MSEs is considered to be a normal distribution.

5. The video comparison and monitoring device according to claim 1, characterized in that the fault detection unit determines the occurrence of a local fault and the occurrence of a synchronization out of sync based on a predetermined duration of the fault.

6. A program for causing a computer to function as a video comparison and monitoring device according to any one of claims 1 to 5.

Images