Security camera with video analytics and direct network communication with neighboring cameras

Abstract

A potentially event monitoring includes a plurality of cameras monitoring a monitored area. The cameras capture video streams showing a portion of the monitored area and perform video analysis on the captured video streams. When a camera identifies an event of interest within the captured video stream, the camera generates event information associated with the identified event of interest. The camera uses relative position information to determine which of the plurality of cameras is positioned to capture the event of interest now and / or in the future and instructs that camera to track the event.

Description

Technical Field

[0001] This disclosure pertains to security systems with cameras, and more specifically to security systems with multiple cameras configured to communicate with each other. Background Technology

[0002] Various areas have cameras positioned within designated security zones. Some cameras possess video analytics capabilities and are therefore able to detect various events independently, without having to transmit their video to a remote server for further analysis. This saves valuable time. When the first camera detects an event, especially one that may be moving, it can benefit from neighboring cameras that also attempt to track the event as it moves. Systems and methods for cameras capable of communicating directly with their neighboring cameras are still needed to track events, including moving events, more quickly and efficiently. Summary of the Invention

[0003] This disclosure relates generally to security systems having cameras, and more specifically to security systems having multiple cameras configured to communicate with each other. Examples present in a surveillance camera include memory, a camera, a network port, and one or more processors operatively coupled to the memory, the camera, and the network port. The memory is configured to store network addresses for each of one or more adjacent cameras, wherein each of the one or more adjacent cameras is network-accessible to the surveillance camera via a network. The camera has a field of view and is configured to capture a video stream of that field of view. The one or more processors are configured to perform video analysis on the video stream captured by the camera to search for potentially interesting events within the video stream. When a potentially interesting event is identified in the video stream, the one or more processors are configured to: generate event information associated with the identified potentially interesting event; retrieve from the memory the network address associated with a selected neighboring camera among the one or more neighboring cameras; assemble one or more messages for transmission to the selected neighboring camera among the one or more neighboring cameras, each of the one or more messages including at least a portion of the event information and a target network address, wherein the target network address is set to the network address of the selected neighboring camera among the one or more neighboring cameras. The one or more processors are further configured to: transmit the one or more messages over the network via the network port, wherein once the one or more messages are received by the selected neighboring camera among the one or more neighboring cameras, the selected neighboring camera is alerted to use at least a portion of the event information to closely monitor the identified potentially interesting event in the field of view of the selected neighboring camera among the one or more neighboring cameras.

[0004] Another example of a method for monitoring an area for an event of interest includes a plurality of cameras positioned around the monitored area. Each of the plurality of cameras stores relative position information and network address information for at least some of the other cameras in the plurality of cameras. The cameras capture a video stream displaying a portion of the monitored area and perform video analysis on the captured video stream. When a camera identifies an event of interest within the captured video stream, the camera generates event information associated with the identified event of interest. The camera uses the relative position information to determine which of the plurality of cameras is positioned to capture the event of interest now and / or in the future. The camera uses the network address information of the camera identified as positioned to capture the event of interest now and / or in the future to address network communications to the camera to initiate a search for the identified event of interest, the network communications including at least some of the event information in the event information.

[0005] Another example exists in a non-transitory computer-readable storage medium having instructions stored thereon. When executed by one or more processors of a camera, the instructions cause the one or more processors of the camera to capture a video stream of the display field of view and perform video analysis on the captured video stream to identify events of interest within the field of view. When the camera detects an event of interest within the captured video stream, the one or more processors of the camera are caused to: capture a video segment of the event of interest; use location information stored within the camera to determine a specific camera positioned to capture the event of interest now and / or in the future; instruct the specific camera to capture and track the event of interest; and transmit the video segment to a remote site.

[0006] The foregoing summary is provided to facilitate understanding of certain features of this disclosure and is not intended to be 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] A more complete understanding of this disclosure can be achieved by considering the following description of various exemplary embodiments in conjunction with the accompanying drawings, wherein:

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

[0009] Figure 2 It can be used Figure 1 A schematic block diagram of an exemplary camera in an exemplary surveillance system, the exemplary camera including one or more processors;

[0010] Figure 3This is a flowchart illustrating an exemplary method;

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

[0012] Figure 5 This is a schematic block diagram of an exemplary monitoring system.

[0013] 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 it is not intended to limit aspects of this disclosure to the specific exemplary embodiments described. Rather, it is intended to cover all modifications, equivalents, and alternatives that fall within the substance and scope of this disclosure. Detailed Implementation

[0014] The following description should be read with reference to the accompanying drawings, in which like reference numerals indicate like elements. The drawings are not necessarily drawn to scale and are not intended to limit the scope of this disclosure. In some drawings, elements deemed unnecessary for understanding the relationships between the illustrated parts may have been omitted for clarity.

[0015] 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).

[0016] 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.

[0017] It should be noted that references to "implementation scheme," "some implementation schemes," "other implementation schemes," etc., in the specification indicate that the described implementation schemes may include specific features, structures, or characteristics, but each implementation scheme need not necessarily include specific features, structures, or characteristics. Furthermore, these phrases do not necessarily refer to the same implementation scheme. Additionally, when a specific feature, structure, or characteristic is described in connection with an implementation scheme, it is conceivable that, whether explicitly described or not, that feature, structure, or characteristic may be applied to other implementation schemes, unless otherwise expressly stated to the contrary.

[0018] Figure 1This is a schematic block diagram illustrating an exemplary surveillance system 10. The exemplary surveillance system 10 includes multiple cameras 12, each labeled 12a, 12b, 12c, and 12d. Although a total of four cameras 12 are shown, it should be understood that in some cases, the surveillance system 10 may have fewer than four cameras 12. In some cases, the surveillance system 10 may generally have more than four cameras 12.

[0019] Each of the plurality of cameras 12 includes network addresses 14, each labeled 14a, 14b, 14c, and 14d. Each of the plurality of cameras 12 includes a field of view (FOV) 16, each labeled 16a, 16b, 16c, and 16d, which at least partially defines the content that each of the plurality of cameras 12 can see. For fixed cameras, the corresponding FOV 16 is also fixed. For adjustable cameras such as PTZ (pan-tilt zoom) cameras, the corresponding FOV 16 is adjustable.

[0020] A network address 14 for each camera 12 identifies a specific addressable location on network 18 for that camera 12. While network 18 is generally shown as a wired network, in some cases, network 18 can be a wireless network communicating using any of a variety of different wireless communication protocols. In some cases, for example, camera 12 can communicate via a 5G cellular network. See reference... Figure 2 As will be discussed, each camera 12 may also know (e.g., store) the network address 14, and in some cases, may know the FOV 16 for other cameras 12 or at least for each of the adjacent cameras 12.

[0021] The exemplary monitoring system 10 also includes a remote site 20 operatively coupled to the network 18, and thus each camera in the camera 12 is able to communicate with the remote site 20 via the network 18. In some cases, the remote site 20 may be viewed as a remote computer with video analytics capabilities or even a remote cloud-based server, capable of determining whether a specific video clip of a potential event displays a real event or a false event, i.e., a false alarm. In some cases, the remote site 20 may include a display capable of showing video clips of a potential event, allowing a human observer to determine whether the video clip displays a real event or a false event.

[0022] Figure 2 It can be seen as Figure 1The example shown is a schematic block diagram of an exemplary surveillance camera 22. The exemplary surveillance camera 22 includes a memory 24 configured to store network addresses 26 for each of one or more adjacent cameras, wherein each of these adjacent cameras is network-accessible to the surveillance camera 22 via a network (such as network 18). In some cases, the network address 26 may include a network address for the surveillance camera 22 itself, which is similar to... Figure 1 The network address 14 is shown. An exemplary surveillance camera 22 includes a camera 28 having a field of view (FOV) 30 and is configured to capture a video stream of the FOV. The FOV 30 can be considered as... Figure 1 The FOV 16 is shown. The exemplary surveillance camera 22 also includes a network port 32, which allows the surveillance camera 22 to communicate via network 18. The surveillance camera 22 can communicate, for example, via the Open Network Video Interface Forum (ONVIF).

[0023] The exemplary surveillance camera 22 further includes one or more processors 34 operatively coupled to memory 24, camera 28, and network port 32. The one or more processors 34 are configured to, for example, perform video analysis on a video stream captured by camera 28 to search for potential events of interest within the video stream. When a potential event of interest is identified in the video stream, the one or more processors 34 are configured to: generate event information associated with the identified potential event of interest, and retrieve from memory 24 the network address associated with one(s) of a selected neighboring camera among the one or more neighboring cameras 12. The one or more processors 34 are configured to: assemble one or more messages for transmission to the selected neighboring camera among the one or more neighboring cameras 12, each message including at least a portion of the event information and a target network address, wherein the target network address is set to the network address of the selected neighboring camera among the one or more neighboring cameras 12. The one or more processors 34 are configured to transmit the one or more messages via network port 32 through network 18. Once the message or the message is received by the selected neighboring camera among the one or more neighboring cameras 12, the selected neighboring camera among the one or more neighboring cameras 12 is alerted to use at least a portion of the event information to closely monitor the identified potential events of interest in the field of view of the camera 28 of the selected neighboring camera among the one or more neighboring cameras 12.

[0024] In some cases, a peer-to-peer communication protocol is used to send one or more messages to a selected neighboring camera among the one or more neighboring cameras 12 via network port 32 through network 18. In some cases, network 18 includes the Internet, and the network address of the selected neighboring camera among the one or more neighboring cameras 12 is an Internet Protocol (IP) address.

[0025] In some cases, the event information included in the one or more messages includes one or more of the following: an image captured by camera 28 representing the identified event of potential interest, and a video clip captured by camera 28 representing the identified event of potential interest. Alternatively or additionally, the event information may include metadata, which may be generated at least in part based on one or more images of the identified event of potential interest captured by camera 28. The one or more messages may include at least some of the metadata. The one or more messages may, for example, include metadata specifying one or more parameters that can be used to identify one or more objects of interest in the images and / or video clips captured by camera 28 representing the identified event of potential interest. For example, the one or more parameters may identify object type (car, person, suitcase, clothing item, glasses, hat, etc.), object color (red, yellow, green, etc.), object size (height, width, etc.), object relationship (e.g., object A is close to object B in time and / or space, object A is behind object B, etc.), object movement (e.g., movement speed, movement direction, etc.), and / or any other parameters as needed. These parameters can be determined by the one or more processors when performing video analysis on the video stream captured by camera 22.

[0026] In some cases, the one or more processors 34 may be configured to: transmit at least some of the event information from the event information to the remote station 20, and receive from the remote station 20 a determination as to whether the identified potential event of interest represents a true event of interest or a false event of interest. The one or more processors 34 may be further configured to: ignore any subsequent detection of the identified potential event of interest when the remote station 20 reports that the identified potential event of interest is a false event of interest.

[0027] In some cases, memory 24 may be configured to store relative position information of each of the one or more adjacent cameras 12, and processors 34 may be configured to identify the direction of movement of an identified potential event of interest in the field of view 30 of camera 28, and to use the relative position information of each of the one or more adjacent cameras 12 and the direction of movement of the identified potential event of interest to identify a selected adjacent camera among the one or more adjacent cameras 12. For example, the relative position information may identify a first adjacent camera located due south of camera 28, whose field of view is adjacent to the southern edge of the field of view of camera 28, and the relative position information may identify a second adjacent camera located due north of camera 28, whose field of view is adjacent to the northern edge of the field of view of camera 28. This is just one example.

[0028] In some cases, the memory 24 may be configured to store relative FOV 30 information of the camera 28 relative to each of the one or more adjacent cameras 12, and the one or more processors 34 may be configured to identify the movement direction of the identified potential event of interest in the FOV 30 of the camera 28, and to identify the selected adjacent camera among the one or more adjacent cameras 12 using the relative FOV 30 information of the camera 28 of each of the one or more adjacent cameras 12 and the movement direction of the identified potential event of interest.

[0029] In some cases, the one or more processors 34 may be configured to use machine learning (ML) and / or artificial intelligence (AI) when performing video analysis on the video stream to identify potentially interesting events within the video stream. Video analysis may include one or more of the following: crowd detection, crowd counting, group detection, face detection, face recognition, animal detection, vehicle detection, license plate detection, fire detection, speed detection, direction detection, detection of weapons such as guns or knives, detection of animals jumping over their cages, detection of traffic violations, and detection of camera tampering. Video analysis may include tracking one or more objects in the field of view 30 of camera 28 as it changes in time and space. In some cases, camera 28 may be a pan-tilt-zoom (PTZ) camera, and video analysis may include changing the field of view of the PTZ camera to track the one or more objects in the field of view of camera 28.

[0030] Figure 3This is a flowchart illustrating an exemplary method 36 for monitoring a monitored area for an event of interest. The monitored area includes a plurality of cameras (such as camera 12) positioned around the monitored area. Each of the plurality of cameras stores relative position information and network address information for at least some of the other cameras. The cameras capture video streams displaying a portion of the monitored area, as indicated in box 38. The cameras perform video analysis on the captured video streams, as indicated in box 40.

[0031] When a camera identifies an event of interest (OP) within a captured video stream, as indicated in box 42, the camera generates event information associated with the identified OOP, as indicated in box 42a. The camera uses relative position information to determine which of the plurality of cameras is positioned to capture the OOP now and / or in the future, as indicated in box 42b. The camera uses the network address information of the camera identified as positioned to capture the OOP now and / or in the future to address network communications to begin searching for the identified OOP, these network communications including at least some of the event information, as indicated in box 42c.

[0032] In some cases, the event information includes metadata, which is generated at least in part based on one or more images of the identified potentially interesting event captured by the camera, and at least some of the metadata is included in network communications to the cameras identified as being positioned to capture the event of interest now and / or in the future.

[0033] In some cases, and as indicated in box 42d, the camera may send at least some of the event information to a remote station and receive back whether the event of interest represents a real event or a false event. When the remote station returns a particular event of interest as a false event, the camera may ignore any subsequent detection of that particular event of interest, as indicated in box 42e.

[0034] Figure 4 This is a flowchart illustrating an exemplary method 50 that can be executed by one or more processors of a camera (such as the one or more processors 34) when the one or more processors execute instructions stored on a non-transitory computer-readable storage medium. When the instructions are executed by the one or more processors of the camera, the one or more processors of the camera capture a video stream of the display field of view, as indicated in box 52, and perform video analysis on the captured video stream to identify events of interest within the field of view, as indicated in box 54.

[0035] When a camera detects an event of interest within the captured video stream, as indicated in box 56, the camera's one or more processors capture a video segment of the event of interest, as indicated in box 56a. The camera's one or more processors use location information stored within the camera to determine the specific camera positioned to capture the event of interest now and / or in the future, as indicated in box 56b. The camera's one or more processors instruct that specific camera to capture and track the event of interest, as indicated in box 56c. The camera's one or more processors transmit the video segment to a remote site, as indicated in box 56d.

[0036] Figure 5 This is a schematic block diagram of an exemplary surveillance system 60. The exemplary surveillance system 60 includes multiple cameras 62, each labeled 62a, 62b, 62c, and 62d. Each camera 62 is an intelligent camera running machine learning (ML) algorithms and / or artificial intelligence (AI) algorithms. In this example, camera 62d has identified a situation, as indicated in box 64. As indicated in box 66, camera 62d saves a temporary video clip displaying the situation and transmits that temporary video clip to a control center (e.g., Figure 1 (Remote site 20 indexed). At box 68, temporary video clips are verified by an operator at the control center and ignored if identified as a false alarm by the operator. If the situation is deemed of interest by the operator at the control center, as indicated in box 70, the control center immediately accesses video input from other cameras in the area, and, if appropriate, the operator at the control center immediately takes action to resolve the situation, as indicated in box 72. In some cases, the operator may send temporary video clips to management agencies such as police departments, enabling them to take immediate action. In some cases, real-time video input may be provided to management agencies simultaneously. If the situation is indicated as a false alarm by the operator at the control center, the ML and / or AI algorithms in camera 62 are updated so that any future situations of a similar kind detected by camera 62 are ignored by camera 62, as indicated in box 74.

[0037] Those skilled in the art will recognize that this disclosure can be presented in various forms different from the specific embodiments described and contemplated herein. Therefore, changes in form and detail may be made without departing from the scope and spirit of this disclosure as set forth in the appended claims.

Claims

1. A surveillance camera (12a), the surveillance camera (12a) comprising: The memory (24) is configured to store the network address (26) for each of one or more adjacent cameras (12b-12d), wherein each of the one or more adjacent cameras (12b-12d) is network-accessible to the surveillance camera (12a) via the network (18); A camera (28) having a field of view (30) is configured to capture a video stream of the field of view; Network port (32); One or more processors (34), said one or more processors (34) being operatively coupled to said memory (24), said camera (28) and said network port (32), said one or more processors (34) being configured to: Video analysis is performed on the video stream captured by the camera (28) to find potential events of interest within the video stream. The video analysis includes object detection for detecting one or more objects within the video stream and one or more object characteristics for detecting each of the one or more objects. When a potentially interesting event is identified in the video stream: Generate event information associated with the identified potential events of interest, the event information including metadata that specifies one or more parameters for each of one or more objects associated with the identified potential events of interest, including one or more object characteristics, including one or more of object type, object color, object size, and spatial and / or temporal relationships between two or more objects; Retrieve from the memory (24) the network address (26) associated with a selected neighboring camera among the one or more neighboring cameras (12b-12d); Assemble one or more messages for transmission to the selected neighboring camera among the one or more neighboring cameras (12b-12d), each of the one or more messages including at least a portion of the event information, which includes metadata specifying one or more parameters for each of the one or more objects associated with the identified event of potential interest, and a target network address, wherein the target network address is set to the network address of the selected neighboring camera among the one or more neighboring cameras (12b-12d); as well as The one or more messages are sent via the network port (32) through the network (18), and once received by the selected neighboring camera among the one or more neighboring cameras (12b-12d), the selected neighboring camera among the one or more neighboring cameras (12b-12d) is alerted to perform video analysis on the video stream captured by the selected neighboring camera among the one or more neighboring cameras (12b-12d) to identify one or more objects associated with the identified potential event of interest, using at least a portion of the event information, including metadata specifying one or more parameters for each of the one or more objects associated with the identified potential event of interest.

2. The surveillance camera according to claim 1, wherein the one or more processors (34) are further configured to: transmit at least some of the event information in the event information to a remote station (20), and receive from the remote station (20) a determination of whether the identified potential event of interest represents a true event of interest or a false event of interest.

3. The surveillance camera according to claim 2, wherein the one or more processors (34) are further configured to: ignore any subsequent detection of the identified potential interest event when the remote site (20) reports back that the identified potential interest event is a false interest event.

4. The surveillance camera according to claim 1, wherein the memory (24) is further configured to store relative position information of each of the one or more adjacent cameras (12b-12d), and the one or more processors (34) are configured to identify the movement direction of one or more objects of the identified potential event of interest in the field of view (30) of the camera (28), and to identify the selected adjacent camera among the one or more adjacent cameras (12b-12d) using the relative position information of each of the one or more adjacent cameras (12b-12d) and the movement direction of the one or more objects of the identified potential event of interest.

5. The surveillance camera according to claim 1, wherein the memory (24) is further configured to store relative field-of-view information of the camera of each of the one or more adjacent cameras (12b-12d), and the one or more processors (34) are configured to identify the movement direction of one or more objects of potential interest in the field of view (30) of the camera (28), and to identify the selected adjacent camera of the one or more adjacent cameras (12b-12d) using the relative field-of-view information of the camera of each of the one or more adjacent cameras (12b-12d) and the movement direction of the one or more objects of potential interest in the identified event.

6. The surveillance camera of claim 1, wherein the one or more messages are sent via the network (18) through the network port (32) using a peer-to-peer communication protocol to the selected neighboring camera among the one or more neighboring cameras (12b-12d).

7. The surveillance camera of claim 1, wherein the one or more processors (34) are configured to use machine learning (ML) and / or artificial intelligence (AI) when performing video analysis on the video stream to identify potential events of interest within the video stream.

8. The surveillance camera according to claim 1, wherein the camera is a pan-tilt-zoom (PTZ) camera, and wherein the video analysis includes: Track one or more objects in the field of view (30) of the camera (28); as well as The field of view (30) of the PTZ camera is changed to track one or more objects in the field of view (30) of the camera (28).

9. A method for monitoring a monitored area for an event of interest, the monitored area comprising a plurality of cameras (12a-12d) arranged around the monitored area, each of the plurality of cameras (12a-12d) storing relative position information and network address information of at least some of the other cameras among the plurality of cameras (12a-12d), the method comprising: Camera (12a) of the plurality of cameras (12a-12d) captures a video stream displaying a portion of the monitored area; The camera (12a) performs video analysis on the captured video stream, the video analysis including object detection for detecting one or more objects in the video stream and one or more object characteristics for detecting each of the one or more objects; When the camera (12a) identifies an event of interest within the captured video stream: The camera (12a) generates event information associated with the identified event of interest, the event information including metadata that specifies one or more parameters for each of one or more objects associated with the identified potential event of interest, including one or more object characteristics, including one or more of object type, object color, object size, and spatial and / or temporal relationships between two or more objects. The camera (12a) uses the relative position information to determine which of the plurality of cameras (12a-12d) is positioned to capture the event of interest now and / or in the future, thereby generating a selected neighboring camera; and The camera (12a) uses the network address information of the selected neighboring camera to address network communications of the selected neighboring camera that include at least a portion of the event information, the at least portion of which includes metadata specifying one or more parameters for each of one or more objects associated with the identified potential event of interest, the network communications instructing the selected neighboring camera to perform video analysis on the video stream captured by the selected neighboring camera to identify one or more of the objects associated with the identified event of interest.

Images