A method and system for information push based on regional intelligent clustering

Abstract

This invention discloses a method and system for information push based on regional intelligent clustering. All cameras register with a management server. Each camera captures and identifies pedestrians, then multicasts a first message containing the identified pedestrian's features, the camera's own geographic location information, and a timestamp to other cameras. Upon receiving the first message, each camera sends a second message to the management server, linking itself to the camera in the first message. Upon receiving the second message, the management server groups the related cameras into the same region and sends information push instructions to the cameras within that region according to a strategy for information push within that region. This invention achieves regional intelligent clustering by recognizing the same pedestrian among cameras. Therefore, when information needs to be pushed to a specific region, the management server can directly issue instructions to the cameras in that region, enabling precise information push based on region.

Description

Technical Field

[0001] This application belongs to the field of information push technology, and in particular relates to an information push method and system for regional intelligent clustering. Background Technology

[0002] With the development of the digital economy, information recommendation (advertising) is no longer limited to traditional media, but widely covers multiple channels such as the Internet, mobile applications, and social media. This information recommendation method greatly improves the promotion effect of information and provides a solid user base for the business products promoted by the recommended information.

[0003] However, many current information recommendation methods fail to accurately target the right audience, making it difficult to convert recommendations into actual consumption and thus increasing the cost of information recommendation. Furthermore, to achieve accurate information recommendations, many systems employ complex technologies to filter out precise user groups, requiring significant human and material resources, and even then, they still cannot accurately pinpoint the target audience for information delivery.

[0004] Therefore, how to make accurate and intelligent information recommendations is an urgent problem to be solved. Summary of the Invention

[0005] The purpose of this application is to provide a method and system for information push based on regional intelligent clustering, so as to accurately identify potential user groups within a region and push information to the identified user groups to achieve better push results.

[0006] To achieve the above objectives, the technical solution of this application is as follows:

[0007] A method for information push based on regional intelligent clustering includes:

[0008] All cameras are registered to the management server, and each camera joins the same multicast group through its own gateway router.

[0009] The camera captures and identifies pedestrians. When the number of times the same pedestrian is identified during continuous capture exceeds the first threshold, the camera records the pedestrian's features and timestamp, and multicasts a first message containing the identified pedestrian's features, the camera's own geographical location information, and the timestamp to other cameras.

[0010] After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it is within the range threshold, they check whether they have recorded the same pedestrian features. When they find that they have recorded the same pedestrian features, they further determine whether the difference between the timestamp recorded by themselves and the timestamp in the first message is less than the time difference threshold. If it is less, they send the second message associated with themselves and the camera in the first message to the management server.

[0011] After receiving the second message, the management server groups the related cameras into the same area and sends information push instructions to the cameras in the designated area according to the information push strategy.

[0012] Furthermore, the information push method for regional intelligent clustering also includes:

[0013] After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it exceeds the range threshold, they send a rejection message to their own gateway router.

[0014] After receiving a rejection message, the gateway router records a multicast suppression entry on the interface that received the rejection message, thus suppressing the first message sent by the camera that sent the first message.

[0015] Furthermore, the information push method for regional intelligent clustering also includes:

[0016] If the gateway router that receives the rejection message has only one outgoing interface, it will continue to send the rejection message to the upstream multicast router. After receiving the rejection message, the upstream multicast router will record the multicast suppression entry and suppress the first message sent by the camera that sent the first message from being sent to the gateway router.

[0017] Furthermore, the information push method for regional intelligent clustering also includes:

[0018] When the camera sends its first message, it includes the IP address of its own gateway router.

[0019] After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it does, they also send a rejection message to the gateway router of the camera that sent the first message.

[0020] The gateway router of the camera that sends the first message records a multicast suppression entry and suppresses the first message sent from the camera that sent the rejection message.

[0021] Furthermore, the information push method for regional intelligent clustering also includes:

[0022] The camera captures and identifies a specified target. When the number of times the specified target is identified during continuous capture exceeds a first threshold, a first message containing the characteristics of the identified specified target, the camera's own geographical location information, and a timestamp is sent.

[0023] After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it is within the range threshold, they check whether they have recorded the same specified target features.

[0024] When a record is found to have a specified target feature, it is further determined whether the difference between the timestamp of its own record and the timestamp in the first message is less than the time difference threshold. If it is less than the threshold, the second message associated with the camera in the first message is sent to the management server.

[0025] If the specified target feature is not recorded, wait for a period of time. If the specified target is identified during the waiting period, send a second message, which is associated with the camera in the first message, to the management server.

[0026] This application also proposes a regional intelligent clustering information push system, including a camera, a gateway router, and a management server, wherein:

[0027] All cameras are registered to the management server, and each camera joins the same multicast group through its own gateway router.

[0028] The camera captures and identifies pedestrians. When the number of times the same pedestrian is identified during continuous capture exceeds the first threshold, the camera records the pedestrian's features and timestamp, and multicasts a first message containing the identified pedestrian's features, the camera's own geographical location information, and the timestamp to other cameras.

[0029] After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it is within the range threshold, they check whether they have recorded the same pedestrian features. When they find that they have recorded the same pedestrian features, they further determine whether the difference between the timestamp recorded by themselves and the timestamp in the first message is less than the time difference threshold. If it is less, they send the second message associated with themselves and the camera in the first message to the management server.

[0030] After receiving the second message, the management server groups the related cameras into the same area and sends information push instructions to the cameras in the designated area according to the information push strategy.

[0031] Furthermore, the regional intelligent clustering information push system also includes:

[0032] After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it exceeds the range threshold, they send a rejection message to their own gateway router.

[0033] After receiving a rejection message, the gateway router records a multicast suppression entry on the interface that received the rejection message, thus suppressing the first message sent by the camera that sent the first message.

[0034] Furthermore, the regional intelligent clustering information push system also includes:

[0035] If the gateway router that receives the rejection message has only one outgoing interface, it will continue to send the rejection message to the upstream multicast router. After receiving the rejection message, the upstream multicast router will record the multicast suppression entry and suppress the first message sent by the camera that sent the first message from being sent to the gateway router.

[0036] Furthermore, the regional intelligent clustering information push system also includes:

[0037] When the camera sends its first message, it includes the IP address of its own gateway router.

[0038] After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it does, they also send a rejection message to the gateway router of the camera that sent the first message.

[0039] The gateway router of the camera that sends the first message records a multicast suppression entry and suppresses the first message sent from the camera that sent the rejection message.

[0040] Furthermore, the regional intelligent clustering information push system also includes:

[0041] The camera captures and identifies a specified target. When the number of times the specified target is identified during continuous capture exceeds a first threshold, a first message containing the characteristics of the identified specified target, the camera's own geographical location information, and a timestamp is sent.

[0042] After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it is within the range threshold, they check whether they have recorded the same specified target features.

[0043] When a record is found to have a specified target feature, it is further determined whether the difference between the timestamp of its own record and the timestamp in the first message is less than the time difference threshold. If it is less than the threshold, the second message associated with the camera in the first message is sent to the management server.

[0044] If the specified target feature is not recorded, wait for a period of time. If the specified target is identified during the waiting period, send a second message, which is associated with the camera in the first message, to the management server.

[0045] This application proposes a method and system for information push based on regional intelligent clustering. By recognizing the same pedestrian among cameras, regional intelligent clustering is achieved. When information needs to be pushed to a specific area, the management server can directly send instructions to the cameras in that area to push the information, thus achieving accurate information push according to the region. Attached Figure Description

[0046] Figure 1 This is a flowchart of an information push method for regional intelligent clustering according to this application.

[0047] Figure 2 This is a schematic diagram of the network structure of an information push system for regional intelligent clustering according to this application. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0049] Example 1, as Figure 1 As shown, a method for information push based on regional intelligent clustering is provided, including:

[0050] Step S1: All cameras register with the management server, and each camera joins the same multicast group through its own gateway router.

[0051] This embodiment applies to an information push system for smart cities. Information pushes can include traffic information, public safety information, or product promotion information, and this application is not limited to the specific types of information pushed. For a given administrative block, there may be multiple different residential areas or multiple different commercial blocks. If information is pushed indiscriminately to everyone within the entire administrative block, people will receive too much clutter, much of which will be considered spam and will not receive attention. This application uses cameras to identify pedestrians and performs intelligent regional clustering, pushing information only to the areas where it is needed, thus achieving precise information delivery.

[0052] To achieve the above objectives, each camera in this embodiment has its own unique serial number and built-in geographical location information, such as longitude: 116.4074°E and latitude: 39.9042°N. When a camera registers with the management server, it can submit its own geographical location information. Thus, the management server also possesses the geographical location information of each camera, or it can obtain the geographical location information through configuration.

[0053] The management server sends a unified multicast group address, such as 226.1.1.1, to all registered cameras to establish a multicast tree for camera synchronization messages. After registering with the management server, each camera needs to initiate an IGMP join message and send it to its own gateway router. The gateway router then sends a PIM (*, G) join message to the upstream router of the PIM, thus enabling the camera to join the (*, G) multicast forwarding tree related to the network registration multicast address.

[0054] Step S2: The camera captures and identifies pedestrians. When the number of times the same pedestrian is identified during continuous capture exceeds the first threshold, the identified pedestrian features and timestamp are recorded, and the first message containing the identified pedestrian features, the camera's own geographical location information, and the timestamp is multicast to other cameras.

[0055] Specifically, the camera periodically (e.g., every 10 seconds) captures images and performs target recognition on the images, extracting pedestrian features, such as facial feature vectors [0.1, -0.2, 0.3, ..., 0.7, -0.8, 0.9], and establishing a record table for the facial feature vectors.

[0056] If the same pedestrian is identified by a camera more than a first threshold number of times during continuous capture, such as more than 2 times, then it is determined that the pedestrian is staying or loitering within the range of this camera.

[0057] For example, if a person is detected at least twice within five consecutive snapshots, it indicates that they lingered or loitered in front of the camera, rather than simply passing by. It's important to note that if the interval between two successful snapshots of the same person exceeds four times (excluding four), the condition of "at least twice within five consecutive snapshots" is not met, and the person is not considered a "loitering face." This is because if the time interval is too long, the pedestrian might have gone somewhere else and then returned.

[0058] After determining whether a pedestrian is staying or loitering within the camera's range, the camera saves the pedestrian's features as a "loitering face" and adds a timestamp. The timestamp records the time of the most recent capture and is continuously updated when the pedestrian is captured again.

[0059] If the camera "hovers over a face" and subsequently fails to capture the corresponding face, and therefore does not update the timestamp, the image will be aged out and deleted after a certain period (e.g., 1 hour).

[0060] The camera simultaneously broadcasts pedestrians matching the "loitering face" feature across the entire network using a unified multicast address (e.g., 226.1.1.1). This message contains pedestrian characteristics (e.g., [0.1, -0.2, 0.3, ..., 0.7, -0.8, 0.9]), the camera's geographical location information (e.g., 116.4074°E latitude: 39.9042°N), a timestamp (2024-11-12, 9:55:30~10:00:00), and its own camera ID (0x1111). This message is called the "first message." Since all cameras use unified multicast, the gateway router's multicast forwarding table uses (*,G) entries, where G is the network registered multicast address.

[0061] After receiving the first message, the gateway router sends it to the (*,G) multicast forwarding tree. The gateway router and other multicast routers in the network then send the first message to other interfaces of the (*,G) multicast forwarding tree, excluding the incoming interface, based on bidirectional PIM, thus enabling full network transmission on the (*,G) multicast forwarding tree.

[0062] Step S3: After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it is within the range threshold, they check whether they have recorded the same pedestrian features. When they find that they have recorded the same pedestrian features, they further determine whether the difference between their own recorded timestamp and the timestamp in the first message is less than the time difference threshold. If it is less, they send the second message associated with themselves and the camera in the first message to the management server.

[0063] In this embodiment, the camera that sends the first message is referred to as camera A. Other cameras, such as camera B, after receiving the first message, first determine the distance between the geographical information contained in the received message and their own geographical location information. If it is within a certain distance range (which can be preset by the administrator, and the distance range can be calculated by straight-line distance, such as 100M, or by walking route in the actual map), then it will be processed; otherwise, it will not be processed, thereby saving resources, because it is meaningless if the distance is too far.

[0064] If camera B is within 100 meters of camera A, then camera B will further check its local "loitering face" database for similar feature vectors. If the feature vector exists, it will then determine if the timestamp of that face is less than a certain threshold (called the time difference threshold), for example, less than 30 minutes. If so, it means that this camera and the camera that sent the message are in the same area. Therefore, camera B sends its own camera ID, for example 0x2222, and the camera ID from the first message (i.e., camera A, 0x1111) to the management server. This message is called the "camera association" message, which is a second message linking itself to the camera in the first message. If the time difference threshold is exceeded, camera B will not take any action.

[0065] In this embodiment, the time difference is controlled within a certain threshold. If the interval is too long, it may be a face that appeared the day before. Although the straight-line distance is close, it may not be possible for the face to travel from camera A to camera B in a short time. Therefore, it may not belong to the same area and will not be processed.

[0066] Step S4: After receiving the second message, the management server divides the related cameras into the same area and sends information push instructions to the cameras in the designated area according to the information push strategy in the designated area.

[0067] The management server receives a large number of "camera association" messages. It extracts the association information between cameras from the messages and obtains the associated camera IDs, such as "Camera A (ID: 0x1111) - Camera B (ID: 0x2222)".

[0068] Through this association, a set of cameras can be established. For example, there are also association messages between "camera B and camera C" and association information between "camera A and camera C". Then "camera A, camera B, and camera C" form a set and belong to the same region, such as region 10.

[0069] If the management server receives association messages between camera A and camera D, but not between camera D and other cameras, then camera D can also be clustered into region 10, provided it is associated with camera A but not with cameras B or C. Any two related cameras can form a region cluster. Any camera can be added to a region as long as it is associated with one of the cameras in that cluster.

[0070] The association information of the management server has an aging period. If no update message is received before the aging period expires, the corresponding association item needs to be deleted.

[0071] Clearly, this set represents a commercial area that is close to each other and walkable, or extended to areas that are vehicularly connected. This kind of regional aggregation is much more scientific than the aggregation of geographic information by the management server, because being close to each other does not necessarily mean belonging to the same commercial area; it could be two areas that are insurmountable by walls, mountains, or rivers.

[0072] In this embodiment, cameras are intelligently clustered, and cameras in the same area are grouped into a set.

[0073] Information push can employ different strategies, such as sending within a designated area. Therefore, when information needs to be pushed to a specific area, the management server can directly send instructions to the cameras in that area to push the information, achieving precise information push based on region. For example, shops in a commercial area can push product or service information to other shops within the commercial area, or the government can push public information about a specific residential community to a particular neighborhood.

[0074] Example 2, to reduce the mutual interference of multicast messages from cameras at long distances, further improves upon Example 1. The regional intelligent clustering information push method further includes:

[0075] After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it exceeds the range threshold, they send a rejection message to their own gateway router.

[0076] After receiving a rejection message, the gateway router records a multicast suppression entry on the interface that received the rejection message, thus suppressing the first message sent by the camera that sent the first message.

[0077] Specifically, if camera B determines that it is "a certain distance" from camera A, it sends a "register rejection" IGMP message to its gateway router B, containing the IP address of camera A (20.20.20.20) and the IP address of camera B (30.30.30.30). Upon receiving this message, gateway router B records a multicast suppression entry on the corresponding interface (the outgoing interface that received the message) in the (*,G) table: the IP address of camera A (20.20.20.20). Subsequently, when this gateway router receives the first message with the source IP address of camera A (20.20.20.20) and the destination address of G (226.1.1.1), it will no longer send it to this outgoing interface, and camera B will no longer be able to receive the first message sent by camera A.

[0078] Example 3 further improves upon Example 2, further reducing the propagation of multicast messages in the network. The regional intelligent clustering information push method also includes:

[0079] If the gateway router that receives the rejection message has only one outgoing interface, it will continue to send the rejection message to the upstream multicast router. After receiving the rejection message, the upstream multicast router will record the multicast suppression entry and suppress the first message sent by the camera that sent the first message from being sent to the gateway router.

[0080] Specifically, if gateway router B's (*,G) multicast forwarding table entry has only this one outgoing interface connected to camera B, it will continue to send a "register rejection" PIM message to the upstream multicast router, containing the IP address of camera A (20.20.20.20). After receiving the above message, the upstream multicast router records a multicast suppression entry in its local (*,G) multicast forwarding table entry: the IP address of camera A1 (20.20.20.20). Thereafter, when the upstream multicast router receives the first message with the source IP address of camera A, it will no longer forward it to gateway router B.

[0081] It should be noted that if network router B has only one outgoing interface and camera B refuses to receive it, then network router B has no need to receive the first message. This can further suppress unnecessary multicast messages in the network.

[0082] However, if a new outgoing interface is added to the multicast forwarding table entry of gateway router B (*,G), and this interface does not suppress the first message with the source IP address of camera A, then network router B needs to send a "Registration Rejection Cancellation" PIM message to the upstream multicast router, containing the IP address of camera A that was originally to be suppressed. Upon receiving this message, the upstream multicast router deletes the marker for the IP address (20.20.20.20) of camera A1 from its local (*,G) multicast forwarding table entry. Afterwards, when this upstream multicast router receives a multicast message with the characteristic source IP address of camera A, it will then forward it to gateway router B.

[0083] Example 4 further improves upon Example 2, further reducing the propagation of multicast messages in the network. The regional intelligent clustering information push method also includes:

[0084] When the camera sends its first message, it includes the IP address of its own gateway router.

[0085] After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it does, they also send a rejection message to the gateway router of the camera that sent the first message.

[0086] The gateway router of the camera that sends the first message records a multicast suppression entry and suppresses the first message sent from the camera that sent the rejection message.

[0087] Specifically, when camera A broadcasts a facial feature vector matching the "loitering face" characteristic across the entire network using a unified multicast address (e.g., 226.1.1.1), it also includes the gateway router's IP address in the message. When camera B detects that it is too far from camera A, it sends a rejection registration message to its directly connected gateway router B, and also sends a "remote rejection registration" message to camera A's gateway router A (the gateway router A's IP address is taken from the first message sent by camera A), containing the IP addresses of both camera A and camera B. Upon receiving this message, gateway router A, based on camera A's IP address in the message, records a multicast suppression entry on its outgoing interface connected to camera A, registering camera B's IP address (30.30.30.30). From then on, gateway router A will directly reject any subsequent messages received from camera B.

[0088] In this embodiment, since camera B interacts with the gateway router of camera A, instead of camera B telling camera A and camera A telling gateway router A, camera A avoids centrally processing notifications from a large number of other remote cameras and centrally registering with gateway router A.

[0089] Example 5: To better verify connected regions, more precise processing is required. Cameras that should indeed belong to the same region should be grouped together, even if they only captured the image once. The region intelligent clustering information push method described in this example also includes:

[0090] The camera captures and identifies a specified target. When the number of times the specified target is identified during continuous capture exceeds a first threshold, a first message containing the characteristics of the identified specified target, the camera's own geographical location information, and a timestamp is sent.

[0091] After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it is within the range threshold, they check whether they have recorded the same specified target features.

[0092] When a record is found to have a specified target feature, it is further determined whether the difference between the timestamp of its own record and the timestamp in the first message is less than the time difference threshold. If it is less than the threshold, the second message associated with the camera in the first message is sent to the management server.

[0093] If the specified target feature is not recorded, wait for a period of time. If the specified target is identified during the waiting period, send a second message, which is associated with the camera in the first message, to the management server.

[0094] Specifically, the designated target is the staff member verifying the connectivity area. After camera A identifies the designated target, it processes it according to the method in Example 1, or it can send the first message to the management server.

[0095] If the designated target is within the range of nearby camera B within a short period of time, but camera B does not capture the designated target twice within five consecutive attempts, then according to the processing flow of Example 1, camera A and camera B cannot form a single area.

[0096] This embodiment improves upon the previous one. After receiving the first message from camera A regarding the specified target, the management server extracts the camera ID, facial feature vector, and timestamp. If the facial feature vector belongs to the specified target, it sends a network-wide multicast message containing camera A's ID, geographical location, the aforementioned facial feature vector, and timestamp. This message is called a "regional connectivity query" message and is a type of the first message, where the facial feature vector is the specified target. This first message can also be sent by camera A via multicast.

[0097] After receiving the information, other camera B extracts it. It then first determines the distance between the received message's geographic information and its own geographic location. Processing is only performed if the distance is within a certain range, thus saving resources, as distances that are too great are meaningless. Next, it checks if there are corresponding target features locally. If so, it checks if the difference between its own recorded timestamp and the timestamp in the first message is less than a time difference threshold. If less, it sends a second message, associating itself with the camera in the first message, to the management server.

[0098] If not, wait for a period of time to see if the specified target can be captured. The waiting time is: time difference threshold - (current time - timestamp in the message). If the specified target is captured within the waiting time, a second message containing the ID of this camera and the ID of camera A is sent back to the management server. If the specified target is not found after the waiting time, camera B clears the first message retrieved from the management server. Therefore, continuing to wait will result in a timeout, which is meaningless.

[0099] When the management server receives the second message, it will cluster camera B and camera A as the same region.

[0100] Camera A in this application will multicast the facial feature vectors that match the "loitering face" across the entire network using a unified multicast group address. When other cameras receive the message, they first determine the distance between the geographical information contained in the received message and their own geographical location. If the distance is within a certain range, they will check their local "loitering face" database. If the face exists, they will then determine whether the timestamp of the face is less than a certain threshold compared to the timestamp of the newly received message. If so, it means that this camera and the camera that sent the message are in the same area. Therefore, they will send their own camera ID and the camera ID in the message (i.e., camera A) to the management server to achieve regional clustering.

[0101] The camera reduces multicast messages on the network by sending "registration rejection" messages, and also achieves bidirectional message suppression and multicast tree pruning. In addition, the management server can proactively initiate "regional connectivity queries" for specific facial feature vectors, allowing the camera to perform special processing on specific facial feature vectors and achieve analysis of specific connected regions.

[0102] Cameras can analyze their associated camera information locally by sending feature multicast messages to each other, thereby creating an information table of associated cameras. This information table allows the management server to perform regional clustering, reducing the pressure on the management server to perform regional clustering through centralized computing.

[0103] Example 6, corresponding to the above-described method for information push based on regional intelligent clustering, proposes a system for information push based on regional intelligent clustering, including a camera, a gateway router, and a management server, wherein:

[0104] All cameras are registered to the management server, and each camera joins the same multicast group through its own gateway router.

[0105] The camera captures and identifies pedestrians. When the number of times the same pedestrian is identified during continuous capture exceeds the first threshold, the camera records the pedestrian's features and timestamp, and multicasts a first message containing the identified pedestrian's features, the camera's own geographical location information, and the timestamp to other cameras.

[0106] After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it is within the range threshold, they check whether they have recorded the same pedestrian features. When they find that they have recorded the same pedestrian features, they further determine whether the difference between the timestamp recorded by themselves and the timestamp in the first message is less than the time difference threshold. If it is less, they send the second message associated with themselves and the camera in the first message to the management server.

[0107] After receiving the second message, the management server groups the related cameras into the same area and sends information push instructions to the cameras in the designated area according to the information push strategy.

[0108] This embodiment of a regional intelligent clustering information push system also includes the following embodiments:

[0109] In one specific embodiment, the regional intelligent clustering information push system further includes:

[0110] After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it exceeds the range threshold, they send a rejection message to their own gateway router.

[0111] After receiving a rejection message, the gateway router records a multicast suppression entry on the interface that received the rejection message, thus suppressing the first message sent by the camera that sent the first message.

[0112] In another specific embodiment, the regional intelligent clustering information push system further includes:

[0113] If the gateway router that receives the rejection message has only one outgoing interface, it will continue to send the rejection message to the upstream multicast router. After receiving the rejection message, the upstream multicast router will record the multicast suppression entry and suppress the first message sent by the camera that sent the first message from being sent to the gateway router.

[0114] In another specific embodiment, the regional intelligent clustering information push system further includes:

[0115] When the camera sends its first message, it includes the IP address of its own gateway router.

[0116] After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it does, they also send a rejection message to the gateway router of the camera that sent the first message.

[0117] The gateway router of the camera that sends the first message records a multicast suppression entry and suppresses the first message sent from the camera that sent the rejection message.

[0118] In another specific embodiment, the regional intelligent clustering information push system further includes:

[0119] The camera captures and identifies a specified target. When the number of times the specified target is identified during continuous capture exceeds a first threshold, a first message containing the characteristics of the identified specified target, the camera's own geographical location information, and a timestamp is sent.

[0120] After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it is within the range threshold, they check whether they have recorded the same specified target features.

[0121] When a record is found to have a specified target feature, it is further determined whether the difference between the timestamp of its own record and the timestamp in the first message is less than the time difference threshold. If it is less than the threshold, the second message associated with the camera in the first message is sent to the management server.

[0122] If the specified target feature is not recorded, wait for a period of time. If the specified target is identified during the waiting period, send a second message, which is associated with the camera in the first message, to the management server.

[0123] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for information push based on regional intelligent clustering, characterized in that, The information push method for regional intelligent clustering includes: All cameras are registered to the management server, and each camera joins the same multicast group through its own gateway router. The camera captures and identifies pedestrians. When the number of times the same pedestrian is identified during continuous capture exceeds the first threshold, the camera records the pedestrian's features and timestamp, and multicasts a first message containing the identified pedestrian's features, the camera's own geographical location information, and the timestamp to other cameras. After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it is within the range threshold, they check whether they have recorded the same pedestrian features. When they find that they have recorded the same pedestrian features, they further determine whether the difference between the timestamp recorded by themselves and the timestamp in the first message is less than the time difference threshold. If it is less, they send the second message associated with themselves and the camera in the first message to the management server. After receiving the second message, the management server groups the related cameras into the same area and sends information push instructions to the cameras in the designated area according to the information push strategy.

2. The information push method for regional intelligent clustering according to claim 1, characterized in that, The information push method for regional intelligent clustering also includes: After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it exceeds the range threshold, they send a rejection message to their own gateway router. After receiving a rejection message, the gateway router records a multicast suppression entry on the interface that received the rejection message, thus suppressing the first message sent by the camera that sent the first message.

3. The information push method for regional intelligent clustering according to claim 2, characterized in that, The information push method for regional intelligent clustering also includes: If the gateway router that receives the rejection message has only one outgoing interface, it will continue to send the rejection message to the upstream multicast router. After receiving the rejection message, the upstream multicast router will record the multicast suppression entry and suppress the first message sent by the camera that sent the first message from being sent to the gateway router.

4. The information push method for regional intelligent clustering according to claim 2, characterized in that, The information push method for regional intelligent clustering also includes: When the camera sends its first message, it includes the IP address of its own gateway router. After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it does, they also send a rejection message to the gateway router of the camera that sent the first message. The gateway router of the camera that sends the first message records a multicast suppression entry and suppresses the first message sent from the camera that sent the rejection message.

5. The information push method for regional intelligent clustering according to claim 1, characterized in that, The information push method for regional intelligent clustering also includes: The camera captures and identifies a specified target. When the number of times the specified target is identified during continuous capture exceeds a first threshold, a first message containing the characteristics of the identified specified target, the camera's own geographical location information, and a timestamp is sent. After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it is within the range threshold, they check whether they have recorded the same specified target features. When a record is found to have a specified target feature, it is further determined whether the difference between the timestamp of its own record and the timestamp in the first message is less than the time difference threshold. If it is less than the threshold, the second message associated with the camera in the first message is sent to the management server. If the specified target feature is not recorded, wait for a period of time. If the specified target is identified during the waiting period, send a second message, which is associated with the camera in the first message, to the management server.

6. An information push system for regional intelligent clustering, characterized in that, The regional intelligent clustering information push system includes cameras, gateway routers, and a management server, wherein: All cameras are registered to the management server, and each camera joins the same multicast group through its own gateway router. The camera captures and identifies pedestrians. When the number of times the same pedestrian is identified during continuous capture exceeds the first threshold, the camera records the pedestrian's features and timestamp, and multicasts a first message containing the identified pedestrian's features, the camera's own geographical location information, and the timestamp to other cameras. After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it is within the range threshold, they check whether they have recorded the same pedestrian features. When they find that they have recorded the same pedestrian features, they further determine whether the difference between the timestamp recorded by themselves and the timestamp in the first message is less than the time difference threshold. If it is less, they send the second message associated with themselves and the camera in the first message to the management server. After receiving the second message, the management server groups the related cameras into the same area and sends information push instructions to the cameras in the designated area according to the information push strategy.

7. The information push system for regional intelligent clustering according to claim 6, characterized in that, The regional intelligent clustering information push system also includes: After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it exceeds the range threshold, they send a rejection message to their own gateway router. After receiving a rejection message, the gateway router records a multicast suppression entry on the interface that received the rejection message, thus suppressing the first message sent by the camera that sent the first message.

8. The information push system for regional intelligent clustering according to claim 7, characterized in that, The regional intelligent clustering information push system also includes: If the gateway router that receives the rejection message has only one outgoing interface, it will continue to send the rejection message to the upstream multicast router. After receiving the rejection message, the upstream multicast router will record the multicast suppression entry and suppress the first message sent by the camera that sent the first message from being sent to the gateway router.

9. The information push system for regional intelligent clustering according to claim 7, characterized in that, The regional intelligent clustering information push system also includes: When the camera sends its first message, it includes the IP address of its own gateway router. After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it does, they also send a rejection message to the gateway router of the camera that sent the first message. The gateway router of the camera that sends the first message records a multicast suppression entry and suppresses the first message sent from the camera that sent the rejection message.

10. The information push system for regional intelligent clustering according to claim 6, characterized in that, The regional intelligent clustering information push system also includes: The camera captures and identifies a specified target. When the number of times the specified target is identified during continuous capture exceeds a first threshold, a first message containing the characteristics of the identified specified target, the camera's own geographical location information, and a timestamp is sent. After receiving the first message, other cameras determine whether the distance between themselves and the camera that sent the first message exceeds the range threshold. If it is within the range threshold, they check whether they have recorded the same specified target features. When a record is found to have a specified target feature, it is further determined whether the difference between the timestamp of its own record and the timestamp in the first message is less than the time difference threshold. If it is less than the threshold, the second message associated with the camera in the first message is sent to the management server. If the specified target feature is not recorded, wait for a period of time. If the specified target is identified during the waiting period, send a second message, which is associated with the camera in the first message, to the management server.

Images