Video surveillance system using distributed intelligence

Abstract

A distributed intelligent video surveillance system can be used to track events of interest. This system includes a main camera and multiple sub-cameras configured to communicate with it. One of the sub-cameras receives video footage and performs analysis and event generation for any events of interest detected within the footage. The sub-camera extracts metadata for any detected event of interest and transmits this metadata to the main camera. The main camera then transmits the extracted metadata to the other sub-cameras, alerting them to track the detected event of interest.

Description

Technical Field

[0001] This disclosure relates generally to video surveillance systems. More specifically, this disclosure relates to video surveillance systems using distributed intelligence between a main camera and sub-cameras. Background Technology

[0002] Multiple video surveillance systems employ cameras installed or otherwise deployed around a monitored area such as a city, a portion of a city, a facility, or buildings. Video surveillance systems may also include mobile cameras, such as drones carrying cameras. Each camera has a field of view corresponding to what that particular camera can see from its specific physical location. When an event or potential event is detected within one or more video streams provided by one or more cameras, the field of view provided by one or more other cameras can actually see the detected event more clearly. Furthermore, the detected event can be mobile, meaning that the detected event can move out of one camera's field of view and into another camera's field of view. Collaboration and communication between cameras can allow for improved surveillance by rapidly obtaining additional information about detected events, especially if cameras can quickly share relevant metadata about detected events. A desired outcome is a video surveillance system where the system is intelligently distributed among cameras and / or groups of cameras. Summary of the Invention

[0003] This disclosure relates to a video surveillance system. In one example, the video surveillance system is configured to provide monitoring of a protected area. The video surveillance system includes a camera group having a main camera and a plurality of sub-cameras configured to communicate with the main camera. Each of the main camera and the plurality of sub-cameras is configured to capture a video stream corresponding to the field of view of a particular camera. Each sub-camera includes an edge controller that monitors its corresponding video stream for one or more predefined events of interest, and when a predefined event of interest is detected, the edge controller generates metadata related to the corresponding event of interest. In some cases, the metadata includes video images and / or video segments representing the corresponding event of interest. Each sub-camera includes a transmitter for transmitting the generated metadata to the main camera, the generated metadata including video images and / or video segments representing the corresponding event of interest. Each sub-camera also includes a receiver for receiving the metadata. The main camera includes a receiver for receiving metadata from the transmitter of each of the plurality of sub-cameras, the metadata including video images and / or video segments representing the events of interest identified by the plurality of sub-cameras. The main camera includes a first transmitter for transmitting at least some metadata received from the transmitter of one of the multiple sub-cameras to a receiver of another of the multiple sub-cameras. The main camera also includes a second transmitter for transmitting metadata received from the transmitters of one or more of the multiple sub-cameras to a remote site for recording on a network video recorder. This metadata includes video images and / or video clips representing events of interest identified by one or more of the multiple sub-cameras.

[0004] In another example, a method for detecting and tracking events of interest using a distributed intelligent video surveillance system is provided. The distributed intelligent video surveillance system includes a main camera and multiple sub-cameras configured to communicate with the main camera. The method includes one of the sub-cameras receiving a video image and performing analysis and event generation for any events of interest detected within the video image. The sub-camera extracts metadata for any detected events of interest and transmits the extracted metadata to the main camera. The main camera transmits the extracted metadata to the other sub-cameras, enabling the other sub-cameras to be alerted to track the detected events of interest.

[0005] In another example, a video surveillance system is configured to provide monitoring of a protected area. The video surveillance system includes a camera group having a main camera and multiple sub-cameras configured to communicate with the main camera. Each sub-camera, including the main camera, is configured to capture a video stream corresponding to its specific field of view. Each sub-camera is configured to analyze its video stream for an event of interest. When an event of interest is detected, each sub-camera generates and captures metadata from its video stream relating to the event of interest, and sends the captured metadata to the main camera via a first communication protocol. The main camera is configured to receive metadata from each of the sub-cameras via the first communication protocol. The main camera is configured to communicate with other main cameras in the camera group and / or with a remote video surveillance system controller via a second communication protocol.

[0006] The foregoing summary is provided to facilitate understanding of the innovative features unique to this disclosure and is not intended as a complete description. A full understanding of this disclosure can be obtained by considering the entire specification, claims, drawings, and abstract as a whole. Attached Figure Description

[0007] This disclosure can be more fully understood by considering the following description of various examples in conjunction with the accompanying drawings, in which:

[0008] Figure 1 This is a schematic block diagram of an exemplary video surveillance system;

[0009] Figure 2 It can be used Figure 1 A schematic block diagram of an exemplary sub-camera in an exemplary video surveillance system;

[0010] Figure 3 It can be used Figure 1 A schematic block diagram of an exemplary main camera in an exemplary video surveillance system;

[0011] Figure 4 This is a flowchart illustrating an exemplary method;

[0012] Figure 5 yes Figure 1 A schematic diagram illustrating an example of a video surveillance system;

[0013] Figure 6 This is a flowchart illustrating an exemplary method;

[0014] Figure 7 This is a schematic diagram illustrating tracking by multiple cameras;

[0015] Figure 8 This is a flowchart illustrating an exemplary method;

[0016] Figure 9 This is a flowchart illustrating an exemplary method; and

[0017] Figure 10 This is a schematic block diagram illustrating the collaboration between adjacent master cameras to track events.

[0018] While this disclosure is subject to various modifications and alternatives, its details have been shown by way of example in the accompanying drawings and will be described in detail. However, it should be understood that this disclosure is not intended to limit it to the specific examples described. Rather, it is intended to cover all modifications, equivalents, and alternatives that fall within the substance and scope of this disclosure. Detailed Implementation

[0019] The following description should be read with reference to the accompanying drawings, in which similar elements in different drawings are numbered in the same manner. The drawings are not necessarily drawn to scale and depict examples that are not intended to limit the scope of this disclosure. While examples of various elements are shown, those skilled in the art will recognize that many of the examples provided have suitable alternatives that can be utilized.

[0020] This document assumes that all numbers are modified by the term “about” unless otherwise explicitly stated. Expressions of numerical ranges using endpoints include all numbers contained within that range (e.g., 1 to 5 includes 1, 1.5, 2, 2.75, 3, 3.80, 4, and 5).

[0021] As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” include plural references, unless otherwise expressly stated. As used in this specification and the appended claims, the term “or” is generally used in its meaning to include “and / or,” unless otherwise expressly stated.

[0022] It should be noted that references to "one embodiment," "some embodiments," or "other embodiments" in the specification indicate that the described embodiments may include specific features, structures, or characteristics, but each embodiment may not necessarily include that specific feature, structure, or characteristic. Furthermore, these phrases do not necessarily refer to the same embodiment. Additionally, when a specific feature, structure, or characteristic is described in conjunction with an embodiment, it is conceivable that, whether explicitly described or not, that feature, structure, or characteristic may be applied to other embodiments, unless otherwise expressly stated otherwise.

[0023] Figure 1This is a schematic block diagram of an exemplary video surveillance system 10. The exemplary video surveillance system 10 includes a first camera group 12, a second camera group 14, and a third camera group 16. Although three camera groups 12, 14, and 16 are shown in total, it should be understood that this is merely illustrative, as the video surveillance system 10 may include any number of camera groups. The first camera group 12 includes a main camera 18 and multiple sub-cameras 20, labeled 20a, 20b, 20c, 20d, and 20e, respectively. Although five sub-cameras 20 are shown in total, it should be understood that the first camera group 12 may include any number of sub-cameras 20, such as one, two, three, four, six, seven, or more sub-cameras 20. Each sub-camera 20 is configured to communicate with the main camera 18. Each main camera 18 and each sub-camera 20 is configured to capture a video stream corresponding to the field of view of a particular camera. In some cases, the main camera 18 can be considered to have a master-slave relationship with each of the sub-cameras in the sub-cameras 20, but this is not required. The main cameras 18, 22, and 26 in a camera group 12, 14, and 16 can work together with other main cameras 18, 22, and 26 in other camera groups 12, 14, and 16 to track an event as it propagates through time and space. A particular main camera 18, 22, and 26 can utilize its corresponding sub-cameras 20, 24, and 28 within its specific camera group 12, 14, and 16 to assist in tracking the propagation of the event.

[0024] It should be understood that within a specific camera group 12, 14, 16, signal strength is inversely proportional to the distance between two objects attempting to communicate with each other. An estimate of the distance between a specific sub-camera 20 and the main camera 18 can be determined using the following equation:

[0025] d(m)=10(FSPL(dB)–K–20log10(f)) / 20

[0026] in,

[0027] d = distance

[0028] f = frequency

[0029] FSPL = Free Space Path Loss

[0030] K is a constant, depending on the units of d and f.

[0031] In some cases, the determination of which sub-cameras form a specific camera group with a particular master camera is based on Received Signal Strength Indicator (RSSI). Each sub-camera may be included in a specific camera group with a master camera that is sufficiently close to it to communicate well with good signal strength and good SNR (signal-to-noise ratio). In some cases, the master camera may receive RSSI information from each sub-camera and determine itself which sub-cameras it will form a camera group with. In other cases, each sub-camera may determine which master camera it should be paired with based on RSSI information from nearby master cameras.

[0032] In the example shown, the second camera group 14 includes a main camera 22 and multiple sub-cameras 24, labeled 24a, 24b, 24c, 24d, and 24e, respectively. While a total of five sub-cameras 24 are shown, it should be understood that the second camera group 14 may include any number of sub-cameras 24, such as one, two, three, four, six, seven, or more sub-cameras 20. Each sub-camera 24 is configured to communicate with the main camera 22. Each main camera 22 and each sub-camera 24 are configured to capture a video stream corresponding to the field of view of a particular camera. In some cases, the main camera 22 may be considered to have a master-slave relationship with each sub-camera 24, but this is not required.

[0033] The third camera group 16 includes a main camera 26 and multiple sub-cameras 28, labeled 28a, 28b, 28c, 28d, and 28e, respectively. Although a total of five sub-cameras 28 are shown, it should be understood that the third camera group 16 may include any number of sub-cameras 28, such as one, two, three, four, six, seven, or more. Each sub-camera 28 is configured to communicate with the main camera 26. Each main camera 26 and each sub-camera 28 is configured to capture a video stream corresponding to the field of view of a particular camera. In some cases, the main camera 26 may be considered to have a master-slave relationship with each sub-camera 28, but this is not required.

[0034] An exemplary video surveillance system 10 includes an NVR (Network Video Recorder) 30. Each of the main cameras 18, 22, and 26 is configured to communicate with the NVR 30 via a network 32. Each of the sub-cameras 20, 24, and 28 is configured to communicate with the main camera 18, 22, or 26, which is part of a camera group 12, 14, or 16 to which both the specific sub-camera 20, 24, or 28 and the corresponding main camera 18, 22, or 26 belong. The main cameras 18, 22, and 26 are configured to communicate with the NVR 30 via the network 32. In some cases, as will be discussed, the main cameras 18, 22, and 26 may also be configured to communicate with other main cameras 18, 22, and 26 located within other camera groups 12, 14, and 16, sometimes via the network 32 and / or some other network (not shown).

[0035] In some cases, each of the sub-cameras 20, 24, and 28 is configured to communicate with its corresponding main camera 18, 22, and 26 using a first communication protocol. In some cases, each of the main cameras 18, 22, and 26 is configured to communicate with the NVR 30 via network 32 using a second communication protocol, which has a higher bandwidth and / or transmission rate than the first communication protocol. In some cases, the main cameras 18, 22, and 26 can communicate with other main cameras 18, 22, and 26 using the same second communication protocol or using a different communication protocol. The first and second communication protocols can be wireless communication protocols. In some cases, the first communication protocol can be a 4G cellular communication protocol, and the second communication protocol can be a 5G cellular communication protocol, but this is just an example. Therefore, in some cases, network 32 can represent a 5G cellular network. It should be understood that 5G cellular communication protocols can provide considerable bandwidth and / or transmission rate advantages over other cellular communication protocols (such as, but not limited to, 4G cellular communication protocols). In some cases, main cameras 18, 22, and 26 can be considered as 4G hotspots, enabling sub-cameras 20, 24, and 28 to communicate with their corresponding main cameras 18, 22, and 26 via 4G. Main cameras 18, 22, and 26 can also be considered as 5G hotspots, enabling them to communicate directly with other main cameras 18, 22, and 26. In some cases, the first and / or second communication protocols can be or may include other wireless communication protocols, such as Wi-Fi, BLE, ZigBee, Z-Wave, 6LoWPAN, Thread, 2G, 3G, 4G, 5G, LTE, NB-IoT, SigFox, LoraWAN, Ingenu, Weightless, EnOcean, Dash7, WirelessHART, and / or any other suitable wireless communication protocol.

[0036] Figure 2 This is a schematic block diagram of an exemplary sub-camera 34. The exemplary sub-camera 34 can be considered as an example of sub-cameras 20, 24, and 28. Sub-camera 34 includes an edge controller 36 configured to perform analysis on one or more predefined events of interest and monitor their corresponding video streams. When an event of interest is detected, edge controller 36 is configured to generate metadata associated with the corresponding event of interest. In some cases, the metadata includes video images and / or video clips representing the corresponding event of interest. Sub-camera 34 includes a transmitter 38 for sending the generated metadata to a main camera (such as main cameras 18, 22, and 26), the generated metadata including, for example, video images and / or video clips representing the corresponding event of interest, as well as other metadata. Sub-camera 34 includes a receiver 40 for receiving metadata. Sub-camera 34 also includes a video source 42. For example, video source 42 may be a camera built into sub-camera 42.

[0037] In some cases, one or more predefined events of interest (ROIs) searched by sub-camera 34 may include the presence of a person with one or more predetermined characteristics, a group of people with one or more predetermined characteristics, a person exhibiting one or more predetermined behaviors, a group of people exhibiting one or more predetermined behaviors, a living object with one or more predetermined characteristics, and / or a living object exhibiting one or more predetermined behaviors. In some cases, one or more predefined ROIs may include the presence of an inanimate object with one or more predetermined characteristics or a group of inanimate objects with one or more predetermined characteristics. Metadata generated by sub-camera 34 may include information identifying sub-camera 34 and / or one or more characteristics of the identified ROIs. In some cases, edge controller 36 may include a self-learning module that can identify ROIs considered anomalous based on normal patterns previously established by the self-learning module. In some cases, sub-camera 34 may determine whether to transmit metadata related to a specific event discovered to its main camera, sometimes based on the determined severity of the identified ROI.

[0038] Figure 3This is a schematic block diagram of an exemplary main camera 44. The exemplary main camera 44 can be considered as an example of main cameras 18, 22, and 26. The main camera 44 includes a first receiver 46 for receiving metadata from a transmitter 38 of each of the plurality of sub-cameras 34. This metadata includes video images and / or video clips representing events of interest identified by the provided plurality of sub-cameras 34. For example, the main camera 44 also includes a second receiver 47 for receiving metadata and other information from other main cameras. The first receiver 46 and the second receiver 47 can receive information via two different communication protocols. The main camera 44 also includes a first transmitter 48 for transmitting at least some of the metadata received from the transmitter 38 of one of the plurality of sub-cameras 34 to a receiver 40 of another of the plurality of sub-cameras 34. The main camera 44 also includes a second transmitter 50 for sending metadata received from transmitter 38 of one or more of the multiple sub-cameras 34 to a remote site for recording on a network video recorder (such as an NVR 30). The metadata includes video images and / or video clips representing events of interest identified by one or more of the multiple sub-cameras provided.

[0039] The exemplary main camera 44 also includes a controller 52. The controller 52 controls the operation of the main camera 44, including controlling communication between the main camera 44 and any of the sub-cameras 34, as well as communication with other main cameras 44 and remote sites (such as, but not limited to, NVR 30). The controller 52 may have analytical capabilities. The controller 52 may determine whether to transmit received information to sub-cameras within its specific camera group. In some cases, the controller 52 may include an inference engine 54 configured to process at least some metadata received from transmitters 38 of at least some of the sub-cameras 34 to track events of interest tracked within a protected area in time and space. In some cases, the inference engine 54 may also provide predictive analytics that can predict event movement. Therefore, the main camera 44 may be able to provide its sub-cameras and other main cameras 44 with information about the predicted movement of events, enabling its sub-cameras and other main cameras 44 to better guide their respective sub-cameras to detect and track events of interest. The exemplary main camera 44 also includes a video source 56. For example, video source 56 could be a camera built into main camera 44.

[0040] In some cases, the first transmitter 48 of the main camera 44 and the receiver 40 of at least one of the multiple sub-cameras 34 can communicate using a first communication protocol. The second transmitter 50 of the main camera 44 can communicate with a remote site using a second communication protocol, which differs from the first communication protocol. The second communication protocol can provide higher bandwidth than the first communication protocol. In some cases, the first communication protocol may include a 4G cellular communication protocol. The second communication protocol may include a 5G cellular communication protocol. These are just examples.

[0041] In some cases, such as Figure 1 As shown, a total of three camera groups 12, 14, and 16 are illustrated, each with a corresponding main camera 18, 22, and 26. The main camera 44 can be configured to notify the main cameras in adjacent camera groups of the tracked event of interest (ROI) and its metadata, sufficient to enable the adjacent camera groups to identify and track the ROI. Furthermore, in response to receiving metadata for a specific ROI from the transmitter 38 of one of its sub-cameras 34, the main camera 44 can send at least some of the metadata to the receiver 40 of another sub-camera in its own sub-camera 34. In response, each edge controller 36 of its plurality of sub-cameras 34 can use the received metadata to begin monitoring its corresponding video stream for the matching ROI.

[0042] Figure 4 This is a flowchart illustrating an exemplary method 57 for detecting and tracking events of interest using a distributed intelligent video surveillance system (such as video surveillance system 10), which includes a main camera (such as main camera 44) and multiple sub-cameras (such as sub-camera 34) configured to communicate with the main camera. Exemplary method 57 includes one of the sub-cameras receiving a video image, as shown in box 58. The sub-camera performs analysis and event generation for any events of interest found within the video image, as shown in box 60. The sub-camera extracts metadata for any discovered events of interest, as shown in box 62. The sub-camera transmits the extracted metadata to the main camera, as shown in box 64. The main camera transmits the extracted metadata to the other sub-cameras among its sub-cameras, enabling the other sub-cameras among its sub-cameras to also search for and track the discovered events of interest, as shown in box 66.

[0043] It should be understood that in some cases, each of the multiple sub-cameras may have a unique field of view, and the main camera may transmit the extracted metadata to the other sub-cameras, enabling each of the other sub-cameras to search for the discovered event of interest within the video stream corresponding to its own unique field of view. In some cases, two or more sub-cameras may each perform analysis and event generation for any event of interest, and each of the two or more sub-cameras may extract metadata and transmit the extracted metadata to the main camera. Exemplary metadata may identify certain event attributes. For example, for a person of interest, the metadata may identify clothing color, approximate height, gender, location, speed, wearing a hat, carrying a bag, the color of the bag, etc. The metadata may also include video images and / or video clips representing the corresponding event of interest. These are just examples.

[0044] In some cases, a master camera can be configured to transmit at least some of the extracted metadata to other master cameras, enabling these other master cameras to instruct their own sub-cameras to locate images corresponding to the extracted metadata. Each of the multiple sub-cameras can communicate with the master camera via a first communication protocol, and the master camera can communicate with other master cameras and with the video surveillance system controller via a second communication protocol that provides higher bandwidth than the first communication protocol. For example, the first communication protocol may include a 4G cellular communication protocol, and the second communication protocol may include a 5G cellular communication protocol. This is just one example.

[0045] Figure 5 This is a schematic diagram illustrating an example of a video surveillance system. Figure 5 A facility 68 is shown divided into three areas: Area 1 (labeled 70), Area 2 (labeled 72), and Area 3 (labeled 74). Each of facility 68 and areas 70, 72, and 74 can typically represent any of various types of facilities, such as an area within a warehouse, a portion of a parking lot, etc. In this example, Area 1 (labeled 70) includes a first camera group 76 with a main camera 78 and a total of three sub-cameras 80 labeled 80a, 80b, and 80c. Area 2 (labeled 72) includes a second camera group 82 with a main camera 84 and a total of six sub-cameras 86 labeled 86a, 86b, 86c, 86d, 86e, and 86f. Area 3 (labeled 74) includes a third camera group 87 with a main camera 88 and a total of five sub-cameras 90 labeled 90a, 90b, 90c, 90d, and 90e.

[0046] Sub-camera 86 is shown highlighted, meaning that one or more sub-cameras among sub-cameras 86 have detected a possible event occurring within facility 68. In this particular example, for instance, sub-camera 86f initially detects a suspected intruder. Because each sub-camera among sub-cameras 86 communicates with the main camera 84, the main camera 84 is able to automatically share metadata related to the possible intruder with the other sub-cameras 86. Thus, sub-cameras 86a and 86b are each able to sequentially search for and detect a suspected intruder as it enters and passes through the field of view of each of sub-cameras 86a and 86b, and report to the main camera 84. Because of this, the main camera 84 is able to track the intruder's movement and determine that the suspected intruder is moving out of area two, marked 72, and into area one, marked 70. Therefore, the main camera 84 communicates with the main camera 78 in area one 70 to share relevant metadata with the group of cameras 76 located in area one, marked 70. If the suspected intruder moves in a different direction instead, the main camera 84 may instead share relevant metadata with the group of cameras 87 located in area three, marked 74.

[0047] Figure 6 This is a flowchart illustrating an exemplary method 92 for tracking events using multiple cameras. Each camera, including various sub-cameras, performs individual camera analysis, as shown in box 94. In one example, these cameras may not support 5G. As shown in box 96, if an event is triggered, whichever or more cameras detect the event will send their camera ID and event metadata to their specific master camera. The specific master camera will poll the sub-cameras that are part of the camera group to which it belongs and will send the event metadata along with its sub-cameras, as shown in box 98. Each sub-camera can then perform analysis to find events based on the received metadata and notify the master camera of any events discovered and their corresponding metadata, as shown in box 100. If an event is seen in a subsequent timeframe, the camera ID and nearby cameras are recorded, as shown in box 102. The master camera can then communicate with nearby master cameras to share relevant metadata, allowing nearby master cameras to guide their camera groups in discovering and tracking events, as shown in box 104. In some cases, and as shown in box 106, event propagation is recorded and notified to relevant authorities.

[0048] Figure 7This is a schematic diagram illustrating multiple camera tracking. In this example, each of the multiple sub-cameras 108, labeled 108a, 108b, 108c, and 108d, transmits video images, video clips, and other relevant metadata to the main camera 110. Box 112 illustrates some steps that can be performed by the main camera 110. The main camera 110 acquires all metadata from the surrounding cameras, as shown in box 114. The main camera 110 can use the metadata provided by the other cameras to stitch together multiple events, as shown in box 116. An inference engine (such as inference engine 54) prioritizes the events and predicts which cameras will be most likely to capture events of interest within their field of view, as shown in box 118. The cameras prioritized are notified to perform analysis and provide the results to the main camera, as shown in box 120. Based on event movement, the main camera notifies neighboring main cameras, as shown in box 122.

[0049] Figure 8 This is a flowchart illustrating an exemplary method 124 for performing single-camera tracking. Images are received, as shown in box 126. Analysis and event generation begin within the camera, as shown in box 128. Attributes of discovered objects are extracted, as shown in box 130. Object tracking algorithms can be applied to track events or objects, as shown in box 132. Objects (such as people, vehicles, etc.) are tracked within the camera's field of view, as shown in box 134. Metadata is collected, as shown in box 136, and sent to the main camera, as shown in box 138. Table 140 provides examples of metadata types.

[0050] Figure 9 It shows that it can be accessed via Figure 3 The flowchart illustrates the exemplary method 142 executed by the inference engine 54. Temporal, spatial, and event attributes are stitched together from multiple cameras, as shown in box 148. The LVC (Master Camera) can search for event-related attributes across all sub-cameras and assess the importance or severity of the event, as shown in box 144. For example, if people are gathered together and there is fighting and crowds in different scenes, the severity can be marked as high. In addition to event details, the inference engine 54 can also detect anomalous patterns in video clips and subsequently assess the situation. Event priority / severity rules or model 146 can be used to inform the determined event priority / severity. If an event is of high priority and an anomalous pattern is detected from a sub-camera, the LVC can send an alert signal to the nearest other LVC and / or remote system (such as, but not limited to, NVR 30), as shown in box 150. Exemplary but non-limiting techniques that can be applied within the inference engine 54 are envisioned, including but not limited to case-based reasoning, expert systems, fuzzy logic, neural networks, deep learning models / methods, and / or any other suitable techniques.

[0051] Figure 10 Figure 154 is a schematic block diagram illustrating the collaboration between adjacent master cameras, particularly when tracking an event moving from one area to another. In Figure 154 are Area 1, labeled 156; Area 2, labeled 158; and Area 3, labeled 160. While three areas 156, 158, and 160 are shown in total, it should be understood that in some cases, a moving event can extend across a large number of areas. In this particular example, the event begins at a starting point 162 within Area 1, labeled 156, and ends at an ending point 164 within Area 3, labeled 160. The event travels along event path 166 through each of the areas 156, 158, and 160. While the event remains within Area 1, labeled 156, LVC (Master Camera) 168 manages its child cameras (not shown) to track the event. As the event moves toward Area 2, labeled 158, LVC 168 in Area 1 transmits metadata about the event to LVC 170 in Area 2, labeled 158. As an event travels toward Zone 3, labeled 160, LVC 170 in Zone 2 transmits metadata about the event to LVC 172 in Zone 3, also labeled 160. In some cases, the metadata collected by the three LVCs 168, 170, and 172 in all three zones can be sent to a remote site, such as NVR 30, and recorded there. In some cases, the metadata from each sub-camera may include video images and / or video clips representing the corresponding detected event of interest.

[0052] Although several illustrative embodiments of this disclosure have been described thus, those skilled in the art will readily understand that other embodiments can be made and used within the scope of the appended claims. However, it should be understood that this disclosure is illustrative in many respects only. Changes may be made to details, particularly those relating to shape, size, arrangement of parts, and exclusion and order of steps, without departing from the scope of this disclosure. The scope of this disclosure is, of course, defined by the language expressed in the appended claims.

Claims

1. A video surveillance system configured to provide monitoring of a protected area, the video surveillance system comprising: A camera group, comprising a main camera and a plurality of selected sub-cameras configured to communicate with the main camera, wherein each of the main camera and the plurality of selected sub-cameras is configured to capture a video stream corresponding to the field of view of its respective camera; The main camera is configured to determine the Received Signal Strength (RSSI) parameter associated with each of the plurality of active sub-cameras, and to determine, based on the RSSI parameter, which of the plurality of active sub-cameras to include in the camera group as one of the plurality of selected sub-cameras; Each of the plurality of selected sub-cameras includes: An edge controller monitors the corresponding video stream for one or more predefined events of interest, and when it detects one of the predefined events of interest, the edge controller performs analysis to generate metadata related to the event of interest, wherein the metadata includes video images and / or video clips representing the corresponding event of interest; A transmitter for sending generated metadata to the main camera, the generated metadata including the video image and / or the video clip representing the corresponding event of interest; Receiver, the receiver being used to receive metadata; The main camera includes: A receiver configured to receive the metadata from the transmitter of each of the plurality of selected sub-cameras in the camera group, the metadata including the video images and / or video clips representing the event of interest identified by the plurality of selected sub-cameras in the camera group; A first transmitter, configured to transmit at least some of the metadata received from the transmitter of one of the plurality of selected sub-cameras to the receiver of another of the plurality of selected sub-cameras; and A second transmitter is configured to send the metadata received from the transmitters of one or more of the plurality of selected sub-cameras to a remote site for recording on a network video recorder, the metadata including video images and / or video clips representing the event of interest identified by the one or more of the plurality of selected sub-cameras.

2. The video surveillance system according to claim 1, wherein the first transmitter of the main camera and the receiver of at least one of the plurality of selected sub-cameras communicate using a first communication protocol.

3. The video surveillance system according to claim 2, wherein the second transmitter of the main camera communicates with the remote site using a second communication protocol, wherein the second communication protocol is different from the first communication protocol.

4. The video surveillance system according to claim 3, wherein the second communication protocol includes the 5G cellular communication protocol.

5. The video surveillance system according to claim 1, wherein one or more of the predefined events of interest include the appearance of a person having one or more predetermined characteristics, a group of people having one or more predetermined characteristics, a person exhibiting one or more predetermined behaviors, a group of people exhibiting one or more predetermined behaviors, a living object having one or more predetermined characteristics, and / or a living object exhibiting one or more predetermined behaviors.

6. The video surveillance system of claim 1, wherein the main camera includes an inference engine that processes at least some of the metadata received from the transmitter of at least some of the plurality of selected sub-cameras to track events of interest tracked within the protected area in time and space.

7. The video surveillance system of claim 6, further comprising an adjacent camera group, wherein the main camera is configured to notify the main cameras in the adjacent camera group of the tracked event of interest and metadata, the metadata being sufficient to enable the adjacent camera group to identify and track the tracked event of interest.

8. The video surveillance system of claim 1, wherein, in response to receiving metadata of a specific event of interest from the transmitter of one of the plurality of selected sub-cameras, the main camera sends at least some of the metadata to the receiver of the plurality of selected sub-cameras, wherein, In response, each edge controller of the plurality of selected sub-cameras uses the received metadata to begin monitoring the corresponding video stream for the corresponding event of interest.

9. A method for detecting and tracking events of interest using a distributed intelligent video surveillance system, the distributed intelligent video surveillance system comprising a camera group consisting of a main camera and multiple sub-cameras, each sub-camera being configured to communicate metadata with the main camera, the method comprising: Determine the Received Signal Strength (RSSI) parameter for communication between each of the multiple active master cameras and each of the multiple active sub-cameras; Identify the main camera of the camera group from the plurality of active main cameras and / or identify the plurality of sub-cameras of the camera group from the plurality of active sub-cameras based on the RSSI parameters; One of the multiple sub-cameras in the camera group receives the video image; The sub-camera performs analysis and event generation for any events of interest found within the video image; The sub-camera extracts metadata for any events of interest it detects; The sub-camera transmits the extracted metadata to the main camera via a first communication protocol; and The main camera transmits the extracted metadata to other sub-cameras in the multiple sub-cameras of the camera group through the first communication protocol, so that the other sub-cameras in the multiple sub-cameras can also track the discovered events of interest; The main camera transmits at least some of the extracted metadata to one or more other main cameras of one or more other camera groups via a second communication protocol, so that the one or more main cameras can instruct their own sub-cameras to find the image corresponding to the extracted metadata. The second communication protocol provides a higher bandwidth communication channel than the first communication protocol.

Images