Video splicing method, system and processor for multi-camera monitoring

Abstract

The present invention belongs to the field of video surveillance technology, and discloses a video splicing method and system for multi-camera surveillance, and a processor, comprising: dividing a surveillance area into a plurality of partition units, arranging at least one camera in each partition unit; and creating an interactive interface for a user to input a preset surveillance target; obtaining the partition video data of each partition unit in real time, performing frame number synchronization processing on the plurality of groups of partition video data, so as to align them on a time axis; cutting each group of partition video data respectively according to a plurality of time periods, and capturing and splicing the preset surveillance target, obtaining the marked video units corresponding to the preset surveillance target, and then connecting them in series to form a collection, and creating a corresponding video directory on the interactive interface; through fine time management and efficient video splicing, the continuity and relevance of the surveillance video are improved, so that the user can quickly access the relevant video clips, and the efficiency of the surveillance system and the user experience are greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of video surveillance technology, and in particular to a video splicing method and system for multi-camera surveillance, and a processor. Background Art

[0002] As an important component of security and supervision, video surveillance systems are widely used in many fields, including but not limited to urban security, traffic management, retail, and other places that require real-time monitoring. With the advancement of technology, the coverage and clarity of video surveillance have been significantly improved. However, with the increase in the number of cameras, how to efficiently integrate and process video data from different cameras, achieve seamless splicing and rapid analysis of multiple video sources, and provide a comprehensive and coherent monitoring view has become a major challenge facing the industry.

[0003] Multi-camera video stitching in the existing technology mainly relies on image processing technology, such as image registration, stitching algorithm, etc., to achieve continuous video playback and unified visual effects; however, in the case of complex monitoring scenes and a large number of cameras, the existing technology often causes the stitched multi-camera video to be a huge amount of information, making it difficult for users to instantly and accurately identify and track key information in the video, such as specific people or dynamic targets, and lacks a flexible user interaction interface, making it difficult for users to quickly locate specific monitoring images or key information according to their own needs, which greatly reduces the effectiveness of video surveillance.

[0004] In view of this, it is necessary to improve the monitoring video processing technology in the prior art to solve the technical problems of being unable to obtain key information in real time and having poor monitoring effect. Summary of the invention

[0005] The purpose of the present invention is to provide a video stitching method and system for multi-camera monitoring, and a processor to solve the above technical problems.

[0006] To achieve this object, the present invention adopts the following technical solutions:

[0007] A video splicing method for multi-camera monitoring, comprising:

[0008] Divide the monitoring area into a plurality of partition units, arrange at least one camera in each of the partition units; and create an interactive interface for the user to input a preset monitoring target;

[0009] The plurality of groups of cameras are operated to perform partition monitoring on the monitoring area, and the partition video data of each partition unit is obtained in real time, and the frame number synchronization processing is performed on the plurality of groups of the partition video data so that all the partition video data are aligned on the time axis;

[0010] The time axis is divided into a plurality of time periods, each group of the partitioned video data is cut into a plurality of video-time units according to the plurality of time periods, and the preset monitoring target is captured and spliced ​​for the plurality of video-time units in each time period to obtain a marker-video unit corresponding to the preset monitoring target;

[0011] Based on the time axis, the tag-video units in each of the time periods are connected in series to form a tag-video set;

[0012] A video catalog corresponding to the tag-video set is created on the interactive interface.

[0013] Optionally, dividing the monitoring area into a plurality of partition units and arranging at least one camera in each of the partition units specifically includes:

[0014] According to the size and layout of the monitoring area, the monitoring area is divided into a two-dimensional grid mapping plane; the two-dimensional grid mapping plane includes a plurality of grid units;

[0015] On the two-dimensional grid mapping plane, determine the specific position and range of each partition unit, and set a preset identifier for each partition unit;

[0016] Obtain monitoring requirements for each partition unit, select a suitable type of camera, and deploy at least one camera in each partition unit;

[0017] Configure the parameters of each camera and connect all said cameras to a central monitoring system.

[0018] Optionally, the step of performing frame synchronization processing on the plurality of groups of the partitioned video data so that all the partitioned video data are aligned on the time axis specifically includes:

[0019] Before the plurality of groups of cameras are operated, the internal clock of each camera is synchronously set so that the plurality of groups of partition video data obtained have a unified timeline;

[0020] Set the standard video acquisition frame rate so that the camera of each partition unit generates the same number of video frames in the same time interval;

[0021] According to the unified timeline, and using the timestamp information of each frame video, the video frames from different cameras are matched and integrated to perform a timestamp alignment operation on the video frames of the multiple groups of the partitioned video data;

[0022] Establishing a buffer and temporarily storing the video frames with aligned timestamps in the buffer;

[0023] The data processing center retrieves the video frames with aligned timestamps in the buffer and reassembles them into a plurality of continuous video data streams to obtain a plurality of groups of the partitioned video data aligned on the time axis.

[0024] Optionally, the time axis is divided into a plurality of time periods, and each group of the partitioned video data is cut into a plurality of video-time units according to the plurality of time periods; specifically including:

[0025] According to the monitoring cycle and monitoring requirements, the time axis is divided into several time periods;

[0026] According to the division of time periods, the video data of each group of partitions is cut, and the video data in each time period is cut into independent video-time units;

[0027] Verify the cut video-time units to ensure that the time range of each video-time unit corresponds to the predetermined time period;

[0028] The verified video-time unit is named according to the naming rule of the corresponding time period and stored in the file system.

[0029] Optionally, the capturing and splicing of the preset monitoring target for a plurality of the video-time units in each time period to obtain a tag-video unit corresponding to the preset monitoring target specifically includes:

[0030] Configure the target recognition system according to the preset monitoring target input by the user in the interactive interface;

[0031] The target recognition system analyzes the video frames of each time period of the video-time unit one by one, and uses image recognition technology to perform target recognition on each video frame to distinguish the video frames into target video frames and non-target video frames;

[0032] Extracting the target video frames for overall classification, splicing the target video frames at the same time stamp using an image edge processing algorithm to obtain a fused target video frame;

[0033] The fused target video frames are processed continuously according to the order of the timeline to obtain a marker-video unit corresponding to the preset monitoring target;

[0034] The tag-video unit of each time period is stored separately in the file system.

[0035] Optionally, based on the time axis, the tag-video units in each time period are connected in series to form a tag-video set; specifically including:

[0036] Based on the monitoring cycle, a complete timeline is constructed;

[0037] Retrieving the mark-video unit of each time period from the file system, and sorting the mark-video unit of each time period according to the order of the time axis;

[0038] Verify the continuity of the timeline of each of the mark-video units, and when the timeline of any of the mark-video units is interrupted, mark the start timestamp and interruption timestamp of the mark-video unit;

[0039] A video concatenation module is used to concatenate the sorted tag-video units according to the time axis to obtain a tag-video set, and the tag-video set is stored in the file system.

[0040] Optionally, creating a video directory corresponding to the tag-video set on the interactive interface; specifically comprising:

[0041] Analyze each tag-video set and extract the main information;

[0042] Naming the tag-video set according to the extracted main information;

[0043] On the interactive interface, a corresponding video directory is created according to the naming of the tag-video set;

[0044] The files of the tag-video set are associated with the created video directory.

[0045] The present invention also provides a video splicing system for multi-camera monitoring, which is used to implement the video splicing method for multi-camera monitoring as described above, and the video splicing system includes:

[0046] The partition setting module is used to divide the monitoring area into multiple partition units and plan the layout of cameras for each partition unit;

[0047] A synchronization control module is used to perform frame synchronization processing on multiple groups of partitioned video data so that all partitioned video data are aligned on the time axis;

[0048] A video data processing module, used for segmenting, splicing and serializing the partitioned video data;

[0049] Target recognition system, used to capture preset monitoring targets in video-time units;

[0050] File system, used to store various video information;

[0051] An interactive interface unit, used for allowing a user to input a preset monitoring target and presenting a video directory corresponding to a tag-video set;

[0052] The central monitoring system is used to control the operation of the multiple groups of cameras.

[0053] The present invention also provides a processor, comprising a memory and at least one processor, wherein instructions are stored in the memory;

[0054] The processor calls the instruction in the memory so that the processor executes the video stitching method for multi-camera monitoring as described above.

[0055] Compared with the prior art, the present invention has the following beneficial effects: when working, by subdividing the monitoring area into multiple partition units and deploying at least one camera in each unit to cover the entire monitoring range, the user can set the preset monitoring target through the interactive interface, and multiple cameras synchronously collect the partition video data, and synchronize the frame number on the time axis to ensure the continuity of the video stream; then, the video data is cut according to the time period, and the target is captured and spliced ​​for each segment to form a tag-video unit containing the preset target, and then these units are connected in series in chronological order to form a tag-video set, and an intuitive video directory is created for these sets on the interactive interface, which is convenient for users to search and watch; this video splicing method provides an instant and accurate monitoring target capture technical solution, which solves the shortcomings of the prior art in real-time acquisition of key monitoring information and target identification. Specifically, through fine time management and efficient video splicing, the continuity and relevance of the monitoring video are improved, so that users can quickly access related video clips, greatly improving the efficiency of the monitoring system and user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0057] The structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with this technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantive technical significance. Any structural modification, change in proportion or adjustment of size, without affecting the effects and purposes that can be achieved by the present invention, should still fall within the scope of the technical contents disclosed by the present invention.

[0058] Figure 1This is one of the flowcharts of the video stitching method of the first embodiment;

[0059] Figure 2 This is a second flow chart of the video stitching method of the first embodiment;

[0060] Figure 3 This is the third flowchart of the video stitching method of the first embodiment. DETAILED DESCRIPTION

[0061] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] In the description of the present invention, it should be understood that the terms "upper", "lower", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. It should be noted that when a component is considered to be "connected" to another component, it may be directly connected to the other component or there may be a centrally arranged component at the same time.

[0063] The technical solution of the present invention is further described below with reference to the accompanying drawings and through specific implementation methods.

[0064] Embodiment 1:

[0065] Combination Figures 1 to 3 As shown, an embodiment of the present invention provides a video stitching method for multi-camera monitoring, comprising:

[0066] S1, dividing the monitoring area into a plurality of partition units, arranging at least one camera in each partition unit; and creating an interactive interface for the user to input a preset monitoring target;

[0067] S2, multiple groups of cameras are operated to monitor the monitoring area in different zones, and the zone video data of each zone unit is obtained in real time. The multiple groups of zone video data are synchronized in the number of frames to align all the zone video data on the time axis;

[0068] S3, dividing the time axis into a number of time periods, cutting each group of partitioned video data into a number of video-time units according to the number of time periods, capturing and splicing a number of video-time units in each time period for a preset monitoring target, and obtaining a marker-video unit corresponding to the preset monitoring target;

[0069] S4, based on the timeline, concatenate the tag-video units in each time period to form a tag-video set;

[0070] S5, creating a video directory corresponding to the tag-video set on the interactive interface.

[0071] The working principle of the present invention is as follows: when working, the monitoring area is subdivided into multiple partition units, and at least one camera is deployed in each unit to cover the entire monitoring range. The user can set the preset monitoring target through the interactive interface, and multiple cameras synchronously collect the partition video data, and synchronize the frame number on the time axis to ensure the continuity of the video stream; then, the video data is cut according to the time period, and the target is captured and spliced ​​for each segment to form a tag-video unit containing the preset target, and then these units are connected in series in chronological order to form a tag-video set, and an intuitive video directory is created for these sets on the interactive interface to facilitate user retrieval and viewing; this video splicing method provides an instant and accurate monitoring target capture technical solution, which solves the shortcomings of the existing technology in real-time acquisition of key monitoring information and target identification, and improves the continuity and relevance of the monitoring video through fine time management and efficient video splicing, so that users can quickly access related video clips, greatly improving the efficiency of the monitoring system and user experience.

[0072] In this embodiment, it is specifically explained that step S1 specifically includes:

[0073] S11, dividing the monitoring area into a two-dimensional grid mapping plane according to the size and layout of the monitoring area; the two-dimensional grid mapping plane includes a plurality of grid units;

[0074] This step establishes the infrastructure for surveillance coverage. Dividing the surveillance area into a two-dimensional grid mapping plane allows the system to be specifically configured for each grid cell, ensuring the comprehensiveness of the surveillance range. Dividing the grid cells allows for more detailed control and positioning, which is crucial for video data acquisition and processing in subsequent steps.

[0075] S12, determining the specific position and range of each partition unit on the two-dimensional grid mapping plane, and setting a preset identifier for each partition unit;

[0076] On the grid mapping plane, the location and boundaries of each partition unit are accurately determined, which helps to clearly divide the camera's area of ​​responsibility and avoid overlap or blind spots; setting a unique preset identifier for each partition can simplify monitoring management and facilitate identification and reference of specific partition units in subsequent operations.

[0077] S13, obtaining monitoring requirements of each partition unit, selecting a suitable type of camera, and arranging at least one camera in each partition unit;

[0078] Select the appropriate type of camera (such as fixed, rotating, zoom, night vision, etc.) according to the monitoring requirements of each partition, such as monitoring range, target type and environmental conditions; ensure that the selected cameras are properly arranged in the partition to achieve the best monitoring effect in the area.

[0079] S14, configure the parameters of each camera and connect all cameras to the central monitoring system;

[0080] Configure the specific parameters of each camera (such as viewing angle, resolution, frame rate, etc.) to meet the predetermined monitoring effect requirements, while ensuring that all cameras have balanced performance and work together; connect all cameras to the central monitoring system to ensure smooth transmission and centralized management of video data, providing a basis for video synchronization and data analysis.

[0081] S15, creating an interactive interface for the user to input a preset monitoring target;

[0082] Develop an interface for user interaction, through which users can input parameters of preset monitoring targets and enable users to easily set and modify monitoring targets; the interactive interface may also include other functions, such as real-time video preview, historical data retrieval, alarm setting, etc., to improve the usability and flexibility of the monitoring system.

[0083] In this embodiment, it is specifically described that step S2 specifically includes:

[0084] S21, multiple groups of cameras are operated to perform partition monitoring on the monitoring area, and the partition video data of each partition unit is obtained in real time.

[0085] S22, before the multiple groups of cameras are operated, the internal clock of each camera is synchronously set so that the obtained multiple groups of partitioned video data have a unified timeline;

[0086] In order to make the video data recorded by all cameras consistent in time, clock synchronization must be performed before they start running; by synchronizing the internal clock, it can be ensured that all video data is consistent in timeline.

[0087] S23, setting a standard video acquisition frame rate so that the camera of each partition unit generates the same number of video frames in the same time interval;

[0088] Determine a unified video acquisition frame rate so that all cameras capture the same number of video frames in the same time interval; this standardized operation helps stabilize video quality and simplify subsequent data processing. At the same time, during the video acquisition process, the central monitoring system needs to continuously detect the video frames output by the camera to ensure that each frame at a certain time point can be matched, and to perform real-time verification and processing of inconsistent frame numbers.

[0089] S24, according to the unified timeline, and using the timestamp information of each frame video,

[0090] Matching and integrating video frames from different cameras to perform time stamp alignment operations on video frames of multiple sets of partitioned video data;

[0091] In the data processing center, a unified timeline is used to match and integrate video frames from different cameras; video frames from different cameras are matched and integrated using the timestamp information of the video frames. Timestamp alignment is the core step to integrate video data streams on a unified timeline, ensuring the time synchronization of video recordings between cameras.

[0092] S25, establishing a buffer, and temporarily storing the video frames with aligned timestamps in the buffer;

[0093] In order to deal with network delays or inconsistent camera processing speeds, a buffer is established to manage video frame data; the buffer will temporarily store the arriving frame data and output the cached frames at the appropriate time to ensure timeline synchronization.

[0094] S26, the data processing center retrieves the video frames in the buffer with aligned timestamps and reassembles them into a plurality of continuous video data streams to obtain a plurality of groups of partitioned video data aligned on the time axis.

[0095] The data processing center reads the video frames in the buffer and reassembles them into a continuous video data stream. This merging process ensures that the video streams are aligned on the time axis to form a seamless and coherent overall monitoring view.

[0096] In this embodiment, it is specifically described that step S3 specifically includes:

[0097] S31, dividing the time axis into several time periods according to the monitoring cycle and monitoring requirements;

[0098] This step involves dividing the entire monitoring timeline into multiple time periods based on the monitoring cycle and specific monitoring needs. This division can be based on a certain length of time (such as every hour or every day) or a specific event (such as during a special event). The division of the timeline makes the monitoring process more orderly and provides a clear time frame for subsequent video data processing.

[0099] Index the video data captured by each camera to quickly access the video content at any specific point in time; the index should include key information such as timestamps and video file locations to facilitate subsequent rapid retrieval and processing;

[0100] S32, cutting each group of partitioned video data according to the division of the time period, and cutting the video data in each time period into independent video-time units;

[0101] This step can be achieved using automated video editing tools, and each video-time unit contains the complete video content within that time period.

[0102] Based on the division of time periods, each group of partitioned video data is cut into separate video-time units; this means that the video data in each time period is isolated and processed independently for more detailed analysis and storage; the cutting process needs to be precise and efficient to ensure that each video-time unit completely contains all the video data in its corresponding time period.

[0103] S33, verifying the cut video-time units to ensure that the time range of each video-time unit corresponds to the predetermined time period; and the video content is complete and correct; if time mismatch or video incompleteness is found, correct or re-cut immediately.

[0104] Verify the cut video-time units to ensure that the time range of each unit correctly corresponds to the pre-set time period; this step is to avoid time mismatch errors during the cutting process; some video units need to be adjusted and re-cut during the verification process to ensure time accuracy and data integrity.

[0105] S34, naming the verified video-time unit according to the naming rule of the corresponding time period, and storing it in the file system.

[0106] The verified video-time units are named according to their corresponding time periods. Such naming rules facilitate subsequent retrieval and identification. Naming is usually based on the characteristics of the time period, such as date and time. The named video-time units are stored in the file system, which not only facilitates management but also facilitates subsequent data access and processing.

[0107] S35, configure the target recognition system according to the preset monitoring target input by the user in the interactive interface; preset monitoring targets (such as specific people, vehicles, etc.), the configuration process includes setting the characteristic parameters of the target, such as color, shape, size, movement mode, etc.; ensure that the target recognition system can adapt to different environments and conditions, such as light changes, occlusion, etc.

[0108] S36, analyzing the video frames of the video-time unit of each time period one by one through the target recognition system, using image recognition technology to perform target recognition in each video frame to distinguish the video frames into target video frames and non-target video frames; wherein the image recognition technology can be a deep learning model to increase the intelligence of recognition.

[0109] In this step, the target recognition system analyzes the video frames of each video-time unit in each time period one by one. Using advanced image recognition technology (such as deep learning algorithms), the system recognizes and distinguishes between preset targets and non-target elements in the video. This process is automated and requires the system to be able to identify targets efficiently and accurately while avoiding misidentification or omissions.

[0110] S37, extracting target video frames for overall classification, and splicing target video frames at the same time stamp using an image edge processing algorithm to obtain a fused target video frame;

[0111] The target video frames extracted from the video-time unit will be classified and integrated, and the target video frames at the same time stamp will be spliced ​​using an image edge processing algorithm to create a more complete target perspective; this step aims to generate a fused, clearer target video frame to provide a better data foundation for subsequent steps.

[0112] S38, performing serial processing on the fused target video frames in the order of the timeline to obtain a tag-video unit corresponding to the preset monitoring target;

[0113] Next, these fused target video frames are processed in a continuous manner in the order of the timeline; the purpose of this step is to generate a tag-video unit corresponding to the preset monitoring target, which contains the complete activity record of the target in a specific time period. Through continuous processing, the fluency and coherence of the video content are guaranteed, providing users with a more intuitive and comprehensive monitoring view.

[0114] S39, storing the mark-video unit of each time period in the file system respectively.

[0115] Finally, the marker-video units in each time period are stored separately in the file system; this storage process should include appropriate naming rules to facilitate subsequent rapid retrieval and access; a systematic storage method is key to ensuring data integrity and accessibility, while also facilitating future data analysis and event review.

[0116] In this embodiment, it is specifically described that step S4 specifically includes:

[0117] S41, constructing a complete timeline according to the monitoring cycle; the complete timeline covers all the tag-video units.

[0118] The goal of this step is to create a complete timeline covering the entire surveillance cycle. The timeline serves as the basis for organizing and processing video data, ensuring the order and continuity of video units; a complete timeline is essential for understanding and analyzing the temporal order and continuity of surveillance videos.

[0119] S42, calling the mark-video unit of each time period from the file system, and sorting the mark-video unit of each time period according to the order of the time axis;

[0120] Extract the markers of each time period - video units from the file system and sort them in the order of the timeline, which ensures the temporal continuity and logic of the video units; the key to this step is to ensure that all video units can be arranged in the order of actual occurrence, which is convenient for subsequent concatenation and analysis.

[0121] S43, verifying the continuity of the timeline of each mark-video unit, and when the timeline of any mark-video unit is interrupted, marking the start timestamp and interruption timestamp of the mark-video unit;

[0122] During the sorting process, it is necessary to verify whether the timeline of each marker-video unit is continuous. If the timeline is found to be interrupted, the start and interruption timestamps of these marker-video units are specially marked; this step helps to identify the activity of the preset monitoring target.

[0123] S44, using a video concatenation module to concatenate the sorted tag-video units according to the time axis, to obtain a tag-video set, and to store the tag-video set in a file system.

[0124] Use the video concatenation module to concatenate the sorted tag-video units along the timeline. This step is the key to the entire surveillance video analysis. It integrates scattered video units into a continuous tag-video set. After the concatenation is completed, these tag-video sets are stored in the file system, which is not only convenient for long-term storage and management, but also convenient for users to access and review the surveillance content at any time.

[0125] In this embodiment, it is specifically described that step S5 specifically includes:

[0126] S51, analyzing each tag-video set to extract main information therein;

[0127] Analyze each tag-video collection and extract key information from it, such as the type of monitoring target, time of occurrence, location, important events, etc. This information will help users understand the content and focus of each video collection. The extracted information must be accurate in order to provide a correct description of the video collection and easy-to-manage information.

[0128] S52, naming the tag-video set according to the extracted main information;

[0129] Use the key information extracted from the video collection to name it. The name should be concise and meaningful, and be able to reflect the main content of the video collection, such as the nature, time or objects involved in the event, which can greatly improve the recognizability and searchability of the video collection.

[0130] S53, on the interactive interface, creating a corresponding video directory according to the name of the tag-video set;

[0131] Create corresponding video directories for each named tag-video collection on the interactive interface; these directories should be easy for users to browse and access, and the layout should take into account user usage habits. Video directories are the main way for users to access and manage video materials, so they must be designed to be intuitive and easy to use.

[0132] S54, associating the file of the tag-video collection with the created video directory.

[0133] The corresponding files of each tag-video set are associated with the video directory created on the interactive interface. When the user selects any video directory, the associated tag-video set can be directly accessed. Through this design, the user can quickly find the required video screen, which improves the retrieval efficiency and user experience.

[0134] Embodiment 2:

[0135] The present invention also provides a video splicing system for multi-camera monitoring, which is used to implement the video splicing method for multi-camera monitoring as in the first embodiment. The video splicing system includes:

[0136] The partition setting module is used to divide the monitoring area into multiple partition units and plan the layout of cameras for each partition unit;

[0137] A synchronization control module is used to perform frame synchronization processing on multiple groups of partitioned video data so that all partitioned video data are aligned on the time axis;

[0138] A video data processing module, used for segmenting, splicing and serializing the partitioned video data;

[0139] Target recognition system, used to capture preset monitoring targets in video-time units;

[0140] File system, used to store various video information;

[0141] An interactive interface unit, used for allowing a user to input a preset monitoring target and presenting a video directory corresponding to a tag-video set;

[0142] Central monitoring system, used to control the operation of multiple groups of cameras.

[0143] The working principle of this system is as follows: the video stitching system of multi-camera monitoring divides the monitoring area into multiple partition units through its partition setting module, and arranges cameras in each unit to ensure comprehensive monitoring coverage; the synchronization control module is responsible for synchronizing the video data captured by cameras in different partitions to ensure the consistency of all video data on the timeline; the video data processing module cuts, splices and concatenates these data to form a continuous video stream; the target recognition system further analyzes the video and captures the preset monitoring targets; the file system is responsible for storing the processed video information, and the interactive interface unit provides users with the functions of setting monitoring targets and accessing video directories; finally, the central monitoring system controls the operation of multiple groups of cameras as a whole to ensure the coordination and efficient operation of the entire system.

[0144] The main advantages of this system are its efficient monitoring coverage and accurate video data processing capabilities. Through meticulous partition settings and camera layout, the system ensures full coverage without blind spots in the monitoring area. Synchronous control on the timeline and accurate processing of video data, including cutting, splicing and concatenation, ensure the continuity and accuracy of video data; the target recognition system can effectively identify and track preset targets, greatly improving monitoring efficiency; in addition, the user-friendly interactive interface and efficient data storage management provide users with a convenient video retrieval and management experience, enhancing the practicality of the monitoring system and user satisfaction.

[0145] Embodiment three:

[0146] The present invention also provides a processor, comprising a memory and at least one processor, wherein instructions are stored in the memory;

[0147] The processor calls the instructions in the memory so that the processor executes the video stitching method for multi-camera monitoring as described in the first embodiment.

[0148] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A video splicing method for multi-camera monitoring, characterized in that: include: Divide the monitoring area into a plurality of partition units, and arrange at least one camera in each of the partition units; And create an interactive interface for users to input preset monitoring targets; The plurality of groups of cameras are operated to perform partition monitoring on the monitoring area, and the partition video data of each partition unit is obtained in real time, and the frame number synchronization processing is performed on the plurality of groups of the partition video data so that all the partition video data are aligned on the time axis; The time axis is divided into a plurality of time periods, each group of the partitioned video data is cut into a plurality of video-time units according to the plurality of time periods, and the preset monitoring target is captured and spliced ​​for the plurality of video-time units in each time period to obtain a marker-video unit corresponding to the preset monitoring target; Based on the time axis, the tag-video units in each of the time periods are connected in series to form a tag-video set; Creating a video directory corresponding to the tag-video set on the interactive interface; The method of dividing the monitoring area into a plurality of partition units and arranging at least one camera in each of the partition units specifically includes: According to the size and layout of the monitoring area, the monitoring area is divided into a two-dimensional grid mapping plane; the two-dimensional grid mapping plane includes a plurality of grid units; On the two-dimensional grid mapping plane, determine the specific position and range of each partition unit, and set a preset identifier for each partition unit; Obtain monitoring requirements for each partition unit, select a suitable type of camera, and deploy at least one camera in each partition unit; Configuring parameters of each camera and connecting all said cameras to a central monitoring system; The step of performing frame synchronization processing on the plurality of groups of the partitioned video data so that all the partitioned video data are aligned on the time axis specifically includes: Before the plurality of groups of cameras are operated, the internal clock of each camera is synchronously set so that the plurality of groups of partition video data obtained have a unified timeline; Set the standard video acquisition frame rate so that the camera of each partition unit generates the same number of video frames in the same time interval; According to the unified timeline, and using the timestamp information of each frame video, the video frames from different cameras are matched and integrated to perform a timestamp alignment operation on the video frames of the multiple groups of the partitioned video data; Establishing a buffer and temporarily storing the video frames with aligned timestamps in the buffer; The data processing center retrieves the video frames with aligned timestamps in the buffer and reassembles them into multiple continuous video data streams to obtain multiple groups of the partitioned video data aligned on the time axis; The time axis is divided into a plurality of time periods, and each group of the partitioned video data is cut into a plurality of video-time units according to the plurality of time periods; specifically, the method includes: According to the monitoring cycle and monitoring requirements, the time axis is divided into several time periods; According to the division of time periods, the video data of each group of partitions is cut, and the video data in each time period is cut into independent video-time units; Verify the cut video-time units to ensure that the time range of each video-time unit corresponds to the predetermined time period; The verified video-time unit is named according to the corresponding time period as a naming rule, and stored in the file system; The capturing and splicing of the preset monitoring target for a plurality of the video-time units in each time period to obtain a marker-video unit corresponding to the preset monitoring target specifically includes: Configure the target recognition system according to the preset monitoring target input by the user in the interactive interface; The target recognition system analyzes the video frames of each time period of the video-time unit one by one, and uses image recognition technology to perform target recognition on each video frame to distinguish the video frames into target video frames and non-target video frames; Extracting the target video frames for overall classification, splicing the target video frames at the same time stamp using an image edge processing algorithm to obtain a fused target video frame; The fused target video frames are processed continuously according to the order of the timeline to obtain a marker-video unit corresponding to the preset monitoring target; The tag-video unit of each time period is stored in the file system respectively; Wherein, based on the time axis, the tag-video units in each time period are connected in series to form a tag-video set; specifically including: Based on the monitoring cycle, a complete timeline is constructed; Retrieving the mark-video unit of each time period from the file system, and sorting the mark-video unit of each time period according to the order of the time axis; Verify the continuity of the timeline of each of the mark-video units, and when the timeline of any of the mark-video units is interrupted, mark the start timestamp and interruption timestamp of the mark-video unit; A video concatenation module is used to concatenate the sorted tag-video units according to the time axis to obtain a tag-video set, and the tag-video set is stored in the file system.

2. The video stitching method for multi-camera monitoring according to claim 1 is characterized in that: Creating a video directory corresponding to the tag-video set on the interactive interface; specifically comprising: Analyze each of the tag-video sets to extract key information therein; Naming the tag-video set according to the extracted main information; On the interactive interface, creating a corresponding video directory according to the naming of the tag-video set; The files of the tag-video set are associated with the created video directory.

3. A video splicing system for multi-camera monitoring, characterized in that: Used to implement the video stitching method for multi-camera monitoring according to any one of claims 1 to 2, the video stitching system comprises: The partition setting module is used to divide the monitoring area into multiple partition units and plan the layout of cameras for each partition unit; A synchronization control module is used to perform frame synchronization processing on multiple groups of partitioned video data so that all partitioned video data are aligned on the time axis; A video data processing module, used for segmenting, splicing and serializing the partitioned video data; Target recognition system, used to capture preset monitoring targets in video-time units; File system, used to store various video information; An interactive interface unit, used for allowing a user to input a preset monitoring target and presenting a video directory corresponding to a tag-video set; The central monitoring system is used to control the operation of the multiple groups of cameras.

4. A processor, characterized in that: The device comprises a memory and at least one processor, wherein instructions are stored in the memory; The processor calls the instruction in the memory so that the processor executes the video stitching method for multi-camera monitoring as described in any one of claims 1 to 2.

Images