Monitoring system and monitoring method

Abstract

In large-scale surveillance systems equipped with numerous surveillance cameras, the more cameras connected, the more efficient the monitoring method required. Furthermore, displaying appropriate video footage itself can become an operational burden. [Solution] The monitoring system uses multiple surveillance cameras and one or more monitoring monitors operated by monitoring terminals, and includes an analysis device that captures video data captured by the surveillance cameras and performs analysis processing to detect events or anomalies from the video data, a recording device that records the video data, and a management server that switches the video data displayed on the monitoring monitor and switches the recording frame rate of the video data recorded by the recording device based on the analysis results of the analysis processing received from the analysis device.

Description

Technical Field

[0001] The present invention relates to a monitoring system and a monitoring method using a surveillance camera, and is particularly suitable for a large-scale monitoring system to which a large number of surveillance cameras are connected.

Background Art

[0002] In a conventional large-scale monitoring system, since it is necessary to monitor (monitor) a large number of camera images on a plurality of display monitors in a central monitoring room such as a central command post, as many display monitors as possible are arranged and the images are cyclically switched (while automatically switching) for monitoring. In addition, when an alarm is detected in conjunction with sensor equipment or the like, there is a function to automatically switch to the camera image in the vicinity.

[0003] If all the monitoring areas where surveillance cameras are installed can cooperate with other equipment such as sensors, it can be said that there is no problem in monitoring even in a large-scale system. However, in reality, it is rare for all monitoring systems to be constructed in such a form. In addition, even when there is no cooperation with sensor equipment or the like, it cannot be denied that an abnormal state (situation) may occur. In that case, manual operations (such as image switching operations) are required.

[0004] Regarding the recording function of each camera image, as the number of cameras increases and the required storage period becomes longer, the capacity of the storage medium becomes extremely large.

[0005] Here, Patent Document 1 cited as the prior art discloses a technique for ranking the importance of images transmitted from a plurality of cameras based on the detection frame area information of an intruder from an intruder detection sensor, and displaying the image on a monitoring monitor according to the ranking.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

[0007] In large-scale surveillance systems equipped with numerous cameras, the more cameras connected, the more efficient the monitoring method required. Furthermore, displaying appropriate images can itself become an operational burden. In other words, since it is difficult to set up a one-to-one relationship between surveillance cameras and monitors, the efficient number of monitors, their placement, and the control of monitor display switching are crucial. In addition, the installation space of the equipment is also a challenge, and it is desirable to make it as compact as possible.

[0008] Furthermore, regarding the camera video data recording function, an efficient recording method is desired that is appropriate to the capacity of the storage medium, given the enormous amount of storage required.

[0009] Therefore, the present invention aims to provide a technology that allows the system to automatically determine which surveillance camera the monitor operator (operator) should always view (automatically switching the monitor display image), and a technology that automatically changes the recording rate in real time based on the changes in the image being captured by the surveillance camera when saving camera images simultaneously with display. In addition, it aims to realize a camera image data recording function that efficiently utilizes the storage medium of the recording device. [Means for solving the problem]

[0010] To solve the above problems, one representative monitoring system according to the present invention is a monitoring system that uses multiple monitoring cameras and one or more monitoring monitors operated by monitoring terminals, and comprises an analysis device that takes in video data captured by the monitoring cameras and performs analysis processing to detect events or anomalies from the video data, a recording device that records the video data, and a management server that switches the video data displayed on the monitoring monitor and switches the recording frame rate of the video data recorded by the recording device based on the analysis results of the analysis processing received from the analysis device. [Effects of the Invention]

[0011] According to this invention, the user does not need to consciously switch to the necessary video footage, and problems such as missing important information due to incorrect operation can be avoided. In this case, by determining the priority and importance of each event or anomaly, the user can more reliably and instantly confirm the situation. Furthermore, since the recording frame rate can be changed for each event or anomaly, more optimal video data can be efficiently saved. Issues, structures, and effects other than those mentioned above will be clarified by the following explanation of the implementation methods. [Brief explanation of the drawing]

[0012] [Figure 1] Figure 1 is a block diagram showing an example of the system configuration of the monitoring system according to the present invention. [Figure 2] Figure 2 shows an example of multi-screen display on a monitoring screen. [Figure 3] Figure 3 shows an example of video data temporarily stored in memory. [Figure 4] Figure 4 shows an example of saving footage with a reduced recording frame rate during normal operation, i.e., when no event or anomaly is detected. [Figure 5] Figure 5 shows an example of increasing the recording frame rate and saving the footage densely when an event or anomaly is detected. [Modes for carrying out the invention]

[0013] The following describes embodiments for carrying out the present invention with reference to the drawings. However, the present invention is not limited to these embodiments. Furthermore, in the drawings, identical parts are denoted by the same reference numerals. [Examples]

[0014] Figure 1 is a block diagram showing an example of the system configuration of the monitoring system according to the present invention. Each device constituting the monitoring system is connected to the network switch 102, and necessary data is sent and received via this network switch 102.

[0015] Multiple surveillance cameras 101 are installed, for example, in important areas within the facility or in restricted perimeter areas.

[0016] The monitoring terminal 105 is operated by the user and switches the video data from the monitoring camera 101 displayed on the monitoring monitor 106 to check the situation inside the facility.

[0017] The monitoring monitor 106 displays video data from the monitoring camera 101 via operation by the monitoring terminal 105 or control by the management server. The monitoring monitor 106 consists of one or more display devices and can function as a multi-monitor to enable monitoring of multiple video data sets. The monitor display may also be capable of multi-screen display.

[0018] FIG. 2 is a diagram showing an example of multi-screen display on the display screen of the monitoring monitor 106. In FIG. 2, (1) is an example of nine multi-screen displays, (2) is an example of displaying an enlarged screen for close-up in the center of the display, and (3) is an example of displaying an enlarged screen for close-up in the upper left corner of the display. For example, it is possible to close up the video data to be noted from the multi-screen display and display an enlarged screen in the center of the display or at one of the four corners.

[0019] The analysis device 103 captures the video data captured by the monitoring camera 101, performs analysis processing on the video data, and notifies the management server 104 of the analysis result. The analysis processing of this video data includes analysis processing using AI technology (hereinafter referred to as "AI analysis processing").

[0020] However, when the monitoring camera 101 itself implements an analysis processing function, the analysis device 103 is not an essential configuration for system construction. That is, if the analysis processing including AI analysis processing can be performed by either the monitoring camera 101 or the analysis device 103, either can be adopted.

[0021] The recording device 107 records the camera video data captured by the monitoring camera 101. However, due to the high necessity of efficient recording, the recording method is devised under the control of the management server 104 to reduce the storage capacity. This will be described later.

[0022] The management server 104 has a function of managing the entire monitoring system. In the present invention, in particular, when receiving a signal from the analysis device 103, it notifies the monitoring terminal 105 and has a function of automatically switching the video displayed on the monitoring monitor 106 to the video data of the target monitoring camera.

[0023] Furthermore, the management server 104 changes the recording frame rate setting for video data in the recording device 107 based on the results of the analysis processing by either the surveillance camera 101 or the analysis device 103. In other words, the management server 104 has a control function that changes the recording frame rate of video data in normal and abnormal situations. For example, in normal situations, if the camera video data does not change much, it records at a reduced recording frame rate, and when an abnormality is detected, it controls the recording frame rate to be increased and records at a high frame rate.

[0024] As described above, the surveillance system according to the present invention is characterized in that it is equipped with a video data analysis function including AI analysis processing in a large-scale system equipped with multiple surveillance cameras, and the management server controls the automatic switching of camera images displayed on the surveillance monitor based on the analysis results, and also changes the settings for the video data recording frame rate.

[0025] In large-scale surveillance systems that require continuous monitoring, the burden on the operating user is significant. Furthermore, as the number of surveillance cameras increases, selecting the optimal camera becomes more difficult. As a result, the likelihood of missing important events increases.

[0026] Furthermore, while monitoring at central facilities such as command centers is important in this large-scale surveillance system, simply increasing the number of monitoring devices does not necessarily lead to more efficient monitoring. On the contrary, a larger number of monitoring devices may make it more difficult for monitors (operational users) to make instantaneous decisions about which monitors they need to focus on.

[0027] On the other hand, in recent years, it has become possible to use AI analysis processing on video data captured by cameras and other devices to recognize objects (such as people or obstacles) and determine abnormal conditions (unusual behavior) as anomalies. By incorporating this AI analysis processing into a surveillance system, including coordination with physical sensor equipment, it becomes possible to control the system to switch the video feed on the surveillance monitor to the video feed of the target camera as soon as an abnormal condition (unusual behavior) is detected as an anomaly, triggering an alert.

[0028] Therefore, either the surveillance camera 101 or the analysis device 103 performs event detection or anomaly detection using AI analysis processing for each surveillance camera, and a function is realized to automatically control (i.e., automatically switch) the video data displayed on the surveillance monitor based on the results. Specific examples of event detection or anomaly detection include, for instance, detecting the intrusion of people into restricted areas, or detecting disturbances in pedestrian flow (such as reverse flow) in places like train station concourses or event venues.

[0029] Furthermore, by using AI analysis processing, it is possible to assign importance levels to detected events or anomalies. This means that instead of simply determining whether something is normal or abnormal, the system can classify events or anomalies by importance and display higher-priority events on the monitoring screen based on that priority. As a result, operators can instantly see the optimal camera footage for monitoring at that moment and confirm the best situation. In this way, the system automatically determines the optimal camera footage to display, which ultimately reduces the burden on operators.

[0030] Furthermore, regarding the recording function, the management server 104 changes the recording frame rate of the video data to the recording device 107 based on the results of event detection or anomaly detection. As mentioned above, one method of changing this setting is to record at a reduced recording frame rate during normal (unusual) conditions, and then increase the recording frame rate to record at a high frame rate when an event or anomaly is detected.

[0031] Furthermore, by having a pre- and post-recording function as part of its recording capabilities, the recording frame rate can be increased and saved from a point before an abnormality is detected. However, strictly speaking, this pre- and post-recording function does not change the recording frame rate each time, but rather temporarily stores the footage at a certain high frame rate, filters out unnecessary information, and then saves it.

[0032] Here, the recording method described above will be explained using Figures 3 to 5. Figure 3 shows an example of video data temporarily stored in memory. This is an example of temporarily storing 30 frames of video data per second (30fps).

[0033] Figure 4 shows an example of saving footage with a reduced recording frame rate during normal operation, i.e., when no event or anomaly is detected. The pre-host function reduces the amount of video data stored in memory by 5 frames per second (5fps) by downsampling from the temporarily stored 30 frames per second video data (30fps).

[0034] Figure 5 shows an example of increasing the recording frame rate and saving data more densely when an event or anomaly is detected. From the temporarily stored video data of 30 frames per second (30fps), the recording is made more dense during normal operation, and video data of 15 frames per second (15fps) is actually saved to memory.

[0035] The recording method described above controls the recording frame rate of camera video data. During normal operation with minimal changes in state, the recording frame rate is lowered to reduce the proportion of storage media used by the recording device. The recording frame rate is increased only when an event or anomaly is detected. This eliminates the possibility of missing events or anomalies, and ensures that events and anomalies can be reliably identified when reviewing the recorded video data later. In addition, by changing the recording frame rate each time, even with the same storage media capacity, it becomes possible to utilize the fixed capacity efficiently.

[0036] Although embodiments for carrying out the present invention have been described above, the present invention is not limited to the embodiments described above, and various modifications are possible without departing from the spirit of the invention. [Explanation of Symbols]

[0037] 101... Surveillance camera, 102... Network switch, 103... Analysis device, 104...Management server, 105...Monitoring terminal, 106...Monitoring monitor, 107...Recording device

Claims

1. A surveillance system that uses multiple surveillance cameras and one or more surveillance monitors operated by surveillance terminals, An analysis device that captures video data captured by the aforementioned surveillance camera and performs analysis processing to detect events or anomalies from the video data, A recording device for recording the aforementioned video data, A management server switches the video data displayed on the monitoring monitor and the recording frame rate of the video data recorded in the recording device based on the analysis results of the analysis process received from the analysis device. A monitoring system equipped with the following features.

2. A monitoring system according to claim 1, The aforementioned analysis process includes analysis processing using AI technology. A monitoring system characterized by the following features.

3. A monitoring system according to claim 1, The surveillance camera, the surveillance terminal, the surveillance monitor, the analysis device, the recording device, and the management server are connected via a network switch. A monitoring system characterized by the following features.

4. A monitoring system according to any one of claims 1 to 3, The management server classifies the importance of the events or anomalies detected by the analysis device through the analysis process, and determines the priority order for switching the video data to be displayed on the monitoring monitor based on that importance. A monitoring system characterized by the following features.

5. A monitoring system according to any one of claims 1 to 3, The management server instructs the recording device to lower the recording frame rate during normal times, except when the event or abnormality detected by the analysis device through the analysis process is detected, and to increase the recording frame rate above the normal rate when the event or abnormality is detected. A monitoring system characterized by the following features.

6. A monitoring system according to claim 5, The recording device has a pre- and post-processing function, which increases the recording frame rate and saves the recording before detecting the event or anomaly. A monitoring system characterized by the following features.

7. A monitoring method that uses multiple surveillance cameras and one or more surveillance monitors operated by a monitoring terminal, The system captures video data from the aforementioned surveillance camera and performs analysis processing to detect events or anomalies from the video data. Based on the analysis results of the analysis process, the video data displayed on the monitoring screen is switched, and the recording frame rate when recording the video data is switched. A monitoring method characterized by the following features.

8. A monitoring method according to claim 7, As the aforementioned analysis process, an analysis process using AI technology is applied. A monitoring method characterized by the following features.

9. A monitoring method according to claim 7 or 8, The analysis process classifies the importance of the events or anomalies detected, and determines the priority order for switching the video data to be displayed on the monitoring screen based on that importance. A monitoring method characterized by the following features.

10. A monitoring method according to claim 7 or 8, The system switches to reduce the recording frame rate during normal times, other than when the aforementioned event or abnormality is detected by the analysis process, and to increase the recording frame rate above the normal rate when the aforementioned event or abnormality is detected. A monitoring method characterized by the following features.

Images