An intelligent video surveillance system based on multi-camera data fusion

Abstract

The present invention relates to the field of security technology, and in particular to an intelligent video monitoring system based on multi-camera data fusion, comprising: a plurality of multi-cameras for shooting a monitoring area; a rotation unit for adjusting the shooting angles, shooting positions and shooting foci of each camera in the multi-camera; a verification unit for determining and comparing the picture-in-picture features in the video streams shot by each camera in a single multi-camera within a first preset time, and comparing the picture-in-picture features of the video streams shot by a fixed single camera before and after replacement within a second preset time; an alarm device for receiving an alarm signal and emitting an audible and visual alarm; the system of the present invention avoids misjudgment caused by the shooting limitation of a single camera by setting the above devices, greatly improves the accuracy of judging whether the picture has been modified, and effectively enhances the security and accuracy of the intelligent video monitoring system based on multi-camera data fusion.

Description

Technical Field

[0001] The present invention relates to the field of security technologies, and in particular, to an intelligent video surveillance system based on multi-camera data fusion. Background Art

[0002] An intelligent video surveillance system based on multi-camera data fusion is an intelligent system that uses multiple cameras to collect video data from different perspectives and comprehensively processes this data through advanced data fusion technologies to achieve more accurate and comprehensive monitoring functions. Compared with traditional single-camera surveillance systems, it can overcome the limitations of a single perspective, provide richer scene information and more accurate target analysis results. Each camera independently collects a video stream, and these video streams contain scene information from different perspectives, such as the shape, color, and movement trajectory of objects. The data from different perspectives complement each other and have a wide range of applications in the security industry. However, many thieves use various technological means to interfere with the cameras. Therefore, there is an urgent need for an intelligent video surveillance system to provide better protection.

[0003] Chinese Patent Authorization Publication No.: CN118840787B discloses a safety warning system and method for an aerial passenger device based on target tracking. The invention discloses a safety warning system and method for an aerial passenger device based on target tracking, including: inputting the current video frame into the YOLOv8 target detection model to obtain a first detection result; obtaining the initial state of the first target box corresponding to the target object; inputting the frame number difference v, the time interval between each frame, and the initial state into the state transition model to predict the predicted position of the corresponding first target box and obtain the second detection result of the (i + v)-th video frame; finding the corresponding second target box that best matches the first target box, and taking the corresponding second target box with the best match as the updated target box; performing an audible and visual alarm according to the category of the first target box or the updated target box; repeating the above process until the category of the updated target box is a non-violation behavior, and then ending the loop; the invention can effectively monitor and warn the riding behavior of the aerial passenger device.

[0004] The Chinese patent authorization announcement number: CN114500871 B discloses a multi-channel video analysis method, device and medium. The invention discloses a multi-channel video analysis method, device and medium. Obtain the first real-time video data corresponding to multiple lenses of a multi-camera; perform picture-in-picture synthesis processing on the multiple first real-time video data, and use the synthesized picture-in-picture video data as the second real-time video data; send the second real-time video data to a preset intelligent analysis module to mark the target images in the second real-time video data through the preset intelligent analysis module; encode the second real-time video data with marks to obtain compressed video data; receive a request for the second real-time video data sent by a user, and push the compressed video data to the user. Through the above method, the purpose of analyzing multiple channels of video simultaneously by a low-computing-power device can be achieved.

[0005] However, the above method has the following problems: When there are problems or the monitoring video is tampered with, the comparison and prediction of video data cannot be well warned, and simply relying on images to monitor the device cannot effectively protect the monitoring camera from being invaded. Summary of the Invention

[0006] Therefore, the present invention provides an intelligent video monitoring system based on multi-camera data fusion to overcome the problems in the prior art that when there are problems or the monitoring video is tampered with, the comparison and prediction of video data cannot be well warned, and simply relying on images to monitor the device cannot effectively protect the monitoring camera from being invaded.

[0007] To achieve the above object, the present invention provides an intelligent video monitoring system based on multi-camera data fusion, including:

[0008] A plurality of multi-cameras for shooting a monitoring area;

[0009] A rotation unit connected to the multi-cameras for adjusting the shooting angles, shooting positions and shooting foci of the cameras in the multi-cameras, extracting a rotating single camera, adjusting the rotating single camera to be in the same shooting position as the remaining fixed single cameras in the same multi-camera in turn, and selecting a single fixed single camera to replace the rotating single camera;

[0010] A verification unit, which is respectively connected to the multi-camera and the rotation unit, is used to determine and compare the picture-in-picture features in the video streams captured by each camera in a single multi-camera within a first preset time, determine whether to send an alarm signal according to the comparison result, and respectively compare the video stream captured by the rotating single camera with the video streams captured by each fixed single camera, determine whether to send an alarm signal according to the comparison result, and compare the picture-in-picture features of the video streams captured by the fixed single camera before and after replacement within a second preset time, and determine whether to send an alarm signal according to the comparison result;

[0011] An alarm device, which is respectively connected to the multi-camera and the verification unit, is used to receive the alarm signal, and in the state of receiving the alarm signal, swap the shooting positions of the multi-cameras shooting the same monitoring area, and send out an audible and visual alarm;

[0012] Wherein, the shooting position is the corresponding position area in the monitoring area shot by the multi-camera.

[0013] Further, the verification unit determines the picture-in-picture features in the video streams captured by each camera in a single multi-camera, wherein,

[0014] If the picture captured by a single camera is completely included in the picture captured by another single camera, the features of the target object existing in the picture captured by the single camera are the picture-in-picture features;

[0015] If the picture captured by the single camera partially overlaps with the picture captured by the other single camera, the features of the target object existing in the overlapping part of the pictures are the picture-in-picture features;

[0016] The picture-in-picture features are the visual features and motion features of the same target object that appear in two single cameras at the same time; the visual features include the shape features, color features and texture features of the target object; the motion features include the motion direction and motion speed of the target object.

[0017] Further, the verification unit compares the picture-in-picture features in the video streams captured by each camera in a single multi-camera, wherein,

[0018] If the target object does not move within the first preset time, the verification unit compares the visual features of the target object in the video streams captured by each camera;

[0019] If the target object moves within the first preset time, the verification unit compares the motion features of the target object in the video streams captured by each camera;

[0020] Among them, the first preset time is positively correlated with the number of single - eye cameras included in a single multi - eye camera.

[0021] Furthermore, when the verification unit compares the visual features of the target object in the video streams captured by each eye camera, it determines whether to issue the alarm signal according to the comparison result, where

[0022] if the visual features are inconsistent, the verification unit issues the alarm signal.

[0023] Furthermore, when the verification unit compares the motion features of the target object in the video streams captured by each eye camera, it calculates the motion rate of the target object in the video streams captured by each eye camera, where

[0024] if the motion rates are inconsistent, the verification unit issues the alarm signal;

[0025] if the motion rates are consistent, the motion rates are compared with a preset rate. When the motion rate is greater than or equal to the preset rate, the verification unit issues the alarm signal;

[0026] The preset rate is positively correlated with the height of the target object.

[0027] Furthermore, the rotation unit extracts any single - eye camera from a single multi - eye camera, sets the single - eye camera as the rotation single - eye camera, and adjusts the shooting position of the rotation single - eye camera to be the same as the shooting positions of the remaining fixed single - eye cameras in the same multi - eye camera in sequence within a periodic time period,

[0028] The periodic time period is positively correlated with the number of single - eye cameras included in a single multi - eye camera.

[0029] Furthermore, the verification unit respectively compares the video stream captured by the rotation single - eye camera with the video streams captured by each fixed single - eye camera within the same periodic time period, and determines whether to issue the alarm signal according to the comparison result, where

[0030] if the picture - in - picture feature of the target object in the video stream captured by the rotation single - eye camera is inconsistent with the picture - in - picture feature of the same target object in the video streams captured by the fixed single - eye cameras, the verification unit issues the alarm signal.

[0031] Furthermore, the rotation unit selects a single fixed single - eye camera to replace the rotation single - eye camera, where

[0032] In a state where the rotation unit adjusts the shooting position of the rotating monocular camera to be consistent with that of the fixed monocular camera, the rotating monocular camera and the fixed monocular camera are replaced.

[0033] Further, the verification unit compares the picture-in-picture features of the video streams captured by the fixed monocular camera before and after replacement within the second preset time, and determines whether to issue an alarm signal according to the comparison result. Among them,

[0034] If the picture-in-picture features of the video streams captured by the fixed monocular camera before and after replacement are inconsistent within the second preset time, the verification unit issues the alarm signal;

[0035] The second preset time is positively correlated with the number of monocular cameras included in a single multi-camera.

[0036] Further, in a state where the alarm device receives the alarm signal, it switches the shooting positions of the multi-cameras that shoot the same monitoring area, emits a sound and light alarm, and closes the multi-camera with inconsistent picture-in-picture features.

[0037] Compared with the prior art, the beneficial effects of the present invention are that the system of the present invention can verify the video streams captured by the cameras from multiple dimensions by comparing the features of the same target object in the pictures captured by different cameras in the multi-camera. Since the shooting angles, focal lengths, etc. of different cameras are different, the feature information of the target object obtained is rich and diverse. In a monitoring scenario, one camera may capture the full frontal view of the target object, and another may capture the partial side details. By comprehensively comparing the features such as the shape, color, and texture of the target object from these different perspectives, it is possible to more comprehensively and accurately determine whether the picture conforms to the real situation, avoid misjudgment caused by the shooting limitation of a single camera, and greatly improve the accuracy of judging whether the picture has been modified. At the same time, the feature information of the target object captured by each camera can complement and confirm each other. If a certain monocular camera has a slight deviation in the color presentation of the target object due to light problems, other cameras can provide more accurate color references under different light conditions. Through comparison and correction, it is possible to more reliably identify whether there are abnormal changes in the features of the target object in the picture, so as to accurately judge the authenticity of the picture, effectively improving the security and accuracy of the intelligent video monitoring system based on multi-camera data fusion.

[0038] Furthermore, by comparing the motion characteristics of moving objects in the video streams captured by cameras with different resolutions, the accuracy of object motion analysis is improved. Under normal circumstances, the motion characteristics of moving objects have a certain regularity. By comparing the motion characteristics under different cameras, abnormal behaviors and events that do not conform to the normal mode can be detected in a timely manner, such as sudden acceleration, turning, and stopping of the object, thus realizing abnormal detection and early warning. In the field of intelligent security, such comparison helps to quickly discover suspicious behaviors and improve the security prevention ability, further enhancing the security and accuracy of the intelligent video surveillance system based on multi-camera data fusion.

[0039] Furthermore, the present invention provides a rotating single-camera. By periodically capturing the same video content of different fixed single-cameras, when the fixed cameras are mainly used to continuously monitor the static and dynamic conditions of a specific area, they may not be sensitive enough to capture some abnormal situations with subtle changes or intermittent occurrences. The periodic participation of the rotating camera in the comparison can bring observation perspectives at different time points and different dynamic conditions. When the rotating single-camera periodically compares with the fixed single-camera, if the fixed camera fails and the picture appears abnormal, the problem can be discovered more quickly by comparing the differences between the two pictures. At the same time, the periodic comparison of the rotating single-camera can serve as a supplementary and verification mechanism to ensure that even if the fixed single-camera has a short-term data loss or error, it can be repaired or restored by comparing with the data of the rotating single-camera, thus guaranteeing the integrity of the data of the entire monitoring system and further enhancing the security and accuracy of the intelligent video surveillance system based on multi-camera data fusion.

[0040] Furthermore, the present invention alternately replaces the rotating single-camera and the fixed single-camera, and at the same time, when receiving an alarm signal, it switches the shooting areas of different multi-cameras in the same monitoring area, which can effectively eliminate the monitoring blind spots. Due to the limitations of the installation position and angle of each camera, there may be a certain viewing blind spot, that is, an area that cannot be effectively monitored. By switching the positions of different cameras, the areas that were originally in the blind spot can be effectively monitored, and the areas that were previously under key monitoring can obtain different perspectives from the new camera view, thus realizing the optimization of the monitoring perspective and coverage range, enabling the entire monitoring area to be more comprehensively covered, reducing monitoring dead angles, and at the same time being able to enrich the monitoring angles. Different cameras have different viewing perspectives on the same monitoring area. After switching the cameras, this change in perspective helps to view the situation in the monitoring area with a new perspective. For some abnormal behaviors or events, they may be difficult to detect from the original camera perspective, but may become obvious after changing the perspective, further enhancing the security and accuracy of the intelligent video surveillance system based on multi-camera data fusion. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a schematic structural diagram of the intelligent video monitoring system based on multi-camera data fusion of the present invention;

[0042] Figure 2 This is a schematic diagram of the picture-in-picture feature with the entire screen covered in the embodiment of the present invention;

[0043] Figure 3 This is a schematic diagram of the picture-in-picture feature with partial overlap of the screens in the embodiment of the present invention;

[0044] Figure 4 This is a decision flowchart of the target object motion feature in the embodiment of the present invention;

[0045] Among them, 1 is the first monocular camera image; 2 is the second monocular camera image; 3 is the third monocular camera image; 4 is the fourth monocular camera image; 5 is the target object. Detailed implementation manners

[0046] In order to make the objectives and advantages of the present invention more clear and understandable, the present invention will be further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0047] The preferred implementation manners of the present invention will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these implementation manners are only used to explain the technical principles of the present invention and do not limit the protection scope of the present invention.

[0048] It should be noted that in the description of the present invention, the terms indicating directions or positional relationships such as "upper", "lower", "left", "right", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings. This is only for convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention.

[0049] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installation", "connection", and "coupling" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0050] Please refer to Figure 1 As shown, this is a schematic structural diagram of the intelligent video monitoring system based on multi-camera data fusion of the present invention. An intelligent video monitoring system based on multi-camera data fusion includes:

[0051] A number of multi-lens cameras, which are used to photograph a monitoring area;

[0052] A rotation unit, which is connected to the multi-lens cameras and is used to adjust the shooting angles, shooting positions and shooting foci of the lenses in the multi-lens cameras, and extract a rotating single-lens camera, adjust the rotating single-lens camera to have the same shooting position as the remaining fixed single-lens cameras in the same multi-lens camera in sequence, and select a single fixed single-lens camera to replace the rotating single-lens camera;

[0053] A verification unit, which is respectively connected to the multi-lens cameras and the rotation unit, and is used to determine and compare the picture-in-picture features in the video streams shot by the lenses in a single multi-lens camera within a first preset time, determine whether to send an alarm signal according to the comparison result, and respectively compare the video stream shot by the rotating single-lens camera with the video streams shot by each fixed single-lens camera, determine whether to send an alarm signal according to the comparison result, and compare the picture-in-picture features of the video streams shot by the fixed single-lens cameras before and after replacement within a second preset time, and determine whether to send an alarm signal according to the comparison result;

[0054] An alarm device, which is respectively connected to the multi-lens cameras and the verification unit, and is used to receive the alarm signal, and in the state of receiving the alarm signal, swap the shooting positions of the multi-lens cameras shooting the same monitoring area, and emit an audible and visual alarm;

[0055] Wherein, the shooting position is the corresponding position area in the monitoring area photographed by the multi-lens camera.

[0056] Specifically, the verification unit determines the picture-in-picture features in the video streams shot by the lenses in a single multi-lens camera, wherein,

[0057] If the picture shot by a single-lens camera is completely included in the picture shot by another single-lens camera, the features of the target object existing in the picture shot by the single-lens camera are picture-in-picture features;

[0058] If the picture shot by a single-lens camera partially overlaps with the picture shot by another single-lens camera, the features of the target object existing in the overlapping part of the pictures are picture-in-picture features;

[0059] The picture-in-picture features are the visual features and motion features of the same target object that appear in two single-lens cameras at the same time; the visual features include the shape features, color features and texture features of the target object; the motion features include the motion direction and motion rate of the target object.

[0060] Please cooperate with Figure 2 Refer to Figure 3As shown, they are respectively the schematic diagram of the picture-in-picture feature with the entire picture wrapped in the embodiment of the present invention and the schematic diagram of the picture-in-picture feature with partial overlap of the pictures in the embodiment of the present invention;

[0061] In implementation, the second monocular camera image 2 captured by the second monocular camera is included in the first monocular camera image 1 captured by the first monocular camera. The target object 5 is simultaneously captured in the first monocular camera image 1 and the second monocular camera image 2. Then, the visual features and motion features of the target object 5 are picture-in-picture features;

[0062] The third monocular camera image 3 captured by the third monocular camera and the fourth monocular camera image 4 captured by the fourth monocular camera partially overlap. The target object 5 is captured in the overlapping part. Then, the visual features and motion features of the target object 5 are picture-in-picture features.

[0063] Specifically, the system of the present invention can verify the video stream captured by the cameras from multiple dimensions by comparing the features of the same target object in the pictures captured by different cameras in the multi-camera. Since the shooting angles, focal lengths, etc. of different cameras are different, the feature information of the target object obtained is rich and diverse. In a monitoring scenario, one camera may capture the entire front view of the target object, and another may capture partial side details. By comprehensively comparing the features such as the shape, color, and texture of the target object from these different perspectives, it is possible to more comprehensively and accurately determine whether the picture conforms to the real situation, avoid misjudgment caused by the shooting limitations of a single camera, and greatly improve the accuracy of judging whether the picture has been modified. At the same time, the features of the target object captured by each camera can complement and corroborate each other. If a certain monocular camera has a slight deviation in the color presentation of the target object due to light problems, other cameras can provide more accurate color references under different light conditions. Through comparison and correction, it is possible to more reliably identify whether there are abnormal changes in the features of the target object in the picture, thereby accurately judging the authenticity of the picture, and effectively improving the security and accuracy of the intelligent video monitoring system based on multi-camera data fusion.

[0064] Specifically, the verification unit compares the picture-in-picture features in the video streams captured by each camera in a single multi-camera. Among them,

[0065] If the target object does not move within the first preset time, the verification unit compares the visual features of the target object in the video streams captured by each camera;

[0066] If the target object moves within the first preset time, the verification unit compares the motion features of the target object in the video streams captured by each camera;

[0067] In implementation, the visual feature is the most obvious differentiating feature of a stationary object in a video stream. If the target object does not move within the first preset time, it is determined that the target object is a stationary object, and the verification unit compares the visual features of the target objects in the video streams captured by each camera.

[0068] The motion feature is the most obvious differentiating feature of a moving object in a video stream. If the target object moves within the first preset time, it is determined that the target object is a moving object, and the verification unit compares the motion features of the target objects in the video streams captured by each camera.

[0069] Among them, the first preset time is positively correlated with the number of single cameras included in a single multi-camera.

[0070] It can be understood that the more single cameras a single multi-camera includes, the more video streams are captured per unit time, the longer the time for the verification unit to obtain the captured video streams, and the more accurate the judgment result of the target object.

[0071] Optionally, the number of single cameras included in a single multi-camera is 2, and the first preset time is 10 seconds;

[0072] The number of single cameras included in a single multi-camera is 3, and the first preset time is 15 seconds;

[0073] The number of single cameras included in a single multi-camera is 6, and the first preset time is 30 seconds.

[0074] Specifically, when the verification unit compares the visual features of the target objects in the video streams captured by each camera, it determines whether to send an alarm signal according to the comparison result. Among them,

[0075] If the visual features are inconsistent, the verification unit sends an alarm signal.

[0076] In implementation, the visual features of the target object include the shape feature, color feature and texture feature of the target object; if the target object in one of the two compared video streams appears yellow and appears red in the other video stream, the verification unit sends an alarm signal;

[0077] If the top shape of the target object in one of the two compared video streams contains a flower and this flower does not appear at the top of the target object in the other video stream, the verification unit sends an alarm signal;

[0078] It can be understood that the multi-camera is set at a position higher than the target object, and the external shape feature of the top of the target object captured will not be blocked by other objects;

[0079] If the target object in the two video streams being compared has a grid texture in one video stream and a striped texture in the other video stream, the verification unit issues an alarm signal.

[0080] Please refer to Figure 4 As shown, it is the determination flowchart of the motion characteristics of the target object in the embodiment of the present invention. The verification unit calculates the motion speed of the target object in the video streams captured by each camera while comparing the motion characteristics of the target object in the video streams captured by each camera. Among them,

[0081] If the motion speeds are inconsistent, the verification unit issues an alarm signal;

[0082] If the motion speeds are consistent, the motion speed is compared with a preset speed. When the motion speed is greater than or equal to the preset speed, the verification unit issues an alarm signal;

[0083] In practice, the motion characteristics of the target object include the motion direction and motion speed of the target object; cameras with different purposes analyze the motion direction of the target object by establishing a direction coordinate. If the motion directions of the target object in each video stream are inconsistent, the verification unit issues an alarm signal.

[0084] In practice, the camera calculates the motion speed of the target object moving in different video streams. If the motion speeds are inconsistent, the verification unit issues an alarm signal;

[0085] If the motion speeds are consistent, the motion speed is compared with a preset speed. The preset speed is set to 2 meters per second; if the motion speed is 4 meters per second which is greater than the preset speed of 2 meters per second, the verification unit issues an alarm signal.

[0086] The preset speed is positively correlated with the height of the target object.

[0087] It can be understood that in security monitoring, the target object being monitored is a person. The taller the person's height, the larger the stride and the greater the walking speed during normal walking, and the greater the preset speed.

[0088] Optionally, the height of the target object is 1.6 meters per second and the preset speed is 1.3 meters per second;

[0089] The height of the target object is 1.7 meters per second and the preset speed is 1.5 meters per second;

[0090] The height of the target object is 1.8 meters per second and the preset speed is 1.7 meters per second.

[0091] Specifically, by comparing the motion characteristics of moving objects in the video streams captured by cameras with different resolutions, the accuracy of object motion analysis is improved. Under normal circumstances, the motion characteristics of moving objects have a certain regularity. By comparing the motion characteristics under different cameras, abnormal behaviors and events that do not conform to the normal mode can be detected in a timely manner, such as sudden acceleration, turning, and stopping of the object, thereby realizing abnormal detection and early warning. In the field of intelligent security, this comparison helps to quickly detect suspicious behaviors, improve security prevention capabilities, and further enhance the security and accuracy of the intelligent video surveillance system based on multi-camera data fusion.

[0092] Specifically, the rotation unit extracts any one of the single cameras in a single multi-camera, sets the single camera as the rotating single camera, and adjusts the shooting position of the rotating single camera to be the same as that of the remaining fixed single cameras in the same multi-camera in turn within a periodic time period.

[0093] In implementation, the rotation unit extracts any one single camera as the rotating single camera, and sets the shooting image range of the rotating single camera to be the same as that of the fixed single camera.

[0094] The periodic time period is positively correlated with the number of single cameras included in a single multi-camera.

[0095] It can be understood that the more single cameras are included in a single multi-camera, the longer the time for the rotating single camera to match and compare with all the fixed single cameras will be. Under the condition of ensuring the accuracy of each comparison, the comparison time should not be less than 10 seconds, so the periodic time period will increase.

[0096] When the number of single cameras included in a single multi-camera is 2, the periodic time period is 10 seconds;

[0097] When the number of single cameras included in a single multi-camera is 3, the periodic time period is 12 seconds;

[0098] When the number of single cameras included in a single multi-camera is 6, the periodic time period is 15 seconds.

[0099] Specifically, the verification unit compares the video stream captured by the rotating single camera with the video streams captured by each fixed single camera in the same periodic time period, and determines whether to issue an alarm signal according to the comparison result. Among them,

[0100] If the picture-in-picture feature of the object in the video stream captured by the rotating single camera is inconsistent with the picture-in-picture feature of the same object in the video stream captured by the fixed single camera, the verification unit issues an alarm signal.

[0101] In implementation, inter-frame difference analysis is performed when switching the shooting range of the rotating monocular camera. For the video stream captured by the rotating monocular camera, the degree of difference between adjacent frames is calculated. At the moment of camera rotation, the difference between adjacent frames usually increases significantly, manifested as sudden changes in aspects such as the picture content, color, and brightness. This difference can be quantified by calculating indicators such as the difference in pixel values and the structural similarity index (SSIM) of the image.

[0102] Specifically, the present invention sets up a rotating monocular camera. By periodically shooting the same video content of different fixed monocular cameras, when the fixed cameras are mainly used to continuously monitor the static and dynamic conditions of a specific area, they may not be sensitive enough to capture some abnormal situations with subtle changes or intermittent occurrences. The periodic participation of the rotating camera in the comparison can bring observation perspectives at different time points and different dynamic conditions. When the rotating monocular camera periodically compares with the fixed monocular camera, if a failure occurs in the fixed camera resulting in an abnormal picture, the problem can be discovered more quickly by comparing the differences between the two pictures. At the same time, the periodic comparison of the rotating monocular camera can serve as a supplementary and verification mechanism to ensure that even if there is a temporary data loss or error in the fixed monocular camera, it can be repaired or restored by comparing with the data of the rotating monocular camera, thereby guaranteeing the integrity of the data of the entire monitoring system and further improving the security and accuracy of the intelligent video monitoring system based on multi-camera data fusion.

[0103] Specifically, the rotation unit selects a single fixed monocular camera to replace the rotating monocular camera, where

[0104] The rotation unit replaces the rotating monocular camera and the fixed monocular camera in a state where the shooting position of the rotating monocular camera is adjusted to be the same as that of the fixed monocular camera.

[0105] In implementation, a fixed monocular camera is randomly selected, the shooting position of the rotating monocular camera is adjusted to be the same as it, then the fixed monocular camera is set as the rotating monocular camera, and the rotating monocular camera is set as the fixed monocular camera.

[0106] Specifically, in the present invention, by alternately replacing the rotating monocular camera and the fixed monocular camera, and at the same time swapping the shooting areas of different multi-camera systems in the same monitoring area when an alarm signal is received, it is possible to effectively eliminate monitoring blind spots. Due to the limitations of the installation position and angle of each camera, there may be a certain viewing blind spot, that is, an area that cannot be effectively monitored. By swapping the positions of different cameras, the areas that were originally in the blind spot can be effectively monitored, and the areas that were previously under key monitoring can be observed from different angles from the perspective of the new camera, thereby optimizing the monitoring perspective and coverage range, enabling the entire monitoring area to be more comprehensively covered, reducing monitoring dead angles, and at the same time enriching the monitoring angles. Different cameras have different viewing perspectives on the same monitoring area. After swapping the cameras, this change in perspective helps to examine the situation in the monitoring area with a new perspective. For some abnormal behaviors or events, it may be difficult to detect from the original camera perspective, but it may become obvious after changing the perspective, further enhancing the security and accuracy of the intelligent video monitoring system based on multi-camera data fusion.

[0107] Specifically, the verification unit compares the picture-in-picture features of the video streams captured by the fixed monocular camera before and after replacement within the second preset time, and determines whether to send an alarm signal according to the comparison result. Among them,

[0108] If the picture-in-picture features of the video streams captured by the fixed monocular camera before and after replacement within the second preset time are inconsistent, the verification unit sends an alarm signal;

[0109] In practice, the second preset time is set to 10 seconds. The video stream within 10 seconds recorded by the fixed monocular camera before replacement before the replacement moment is compared with the video stream within 10 seconds recorded by the fixed monocular camera after replacement after the replacement moment. If the picture-in-picture features in the two video streams are inconsistent, the verification unit sends an alarm signal.

[0110] The second preset time is positively correlated with the number of monocular cameras included in a single multi-camera system.

[0111] It can be understood that the more monocular cameras are included in a single multi-camera system, the longer the comparison time and the longer the second preset time, while ensuring the accuracy of the comparison of the video streams before and after replacement.

[0112] Optionally, the number of monocular cameras included in a single multi-camera system is 2, and the second preset time is 3 seconds;

[0113] The number of monocular cameras included in a single multi-camera system is 3, and the second preset time is 5 seconds;

[0114] The number of monocular cameras included in a single multi-camera system is 6, and the second preset time is 10 seconds.

[0115] Specifically, in a state where an alarm signal is received, the alarm device switches the shooting positions of multiple multi-cameras that shoot the same monitoring area, emits an audible and visual alarm, and closes the multi-cameras with inconsistent picture-in-picture features.

[0116] So far, the technical solution of the present invention has been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to relevant technical features, and the technical solutions after these changes or substitutions will all fall within the protection scope of the present invention.

[0117] The above are only the preferred embodiments of the present invention and are not used to limit the present invention; for those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent video surveillance system based on multi-camera data fusion, characterized in that, Including: A plurality of multi-view cameras for photographing a monitoring area; A rotation unit connected to the multi-view cameras for adjusting the photographing angles, photographing positions, and photographing foci of the cameras in each view of the multi-view cameras, extracting a rotating single-view camera, adjusting the photographing positions of the rotating single-view camera to be the same as those of the remaining fixed single-view cameras in the same multi-view camera in sequence, and selecting a single fixed single-view camera to replace the rotating single-view camera; A verification unit connected to the multi-view cameras and the rotation unit respectively for determining and comparing the picture-in-picture features in the video streams photographed by the cameras in each view of a single multi-view camera within a first preset time, determining whether to send an alarm signal according to the comparison result, comparing the video stream photographed by the rotating single-view camera with the video streams photographed by each fixed single-view camera respectively, determining whether to send an alarm signal according to the comparison result, and comparing the picture-in-picture features of the video streams photographed by the fixed single-view cameras before and after replacement within a second preset time, and determining whether to send an alarm signal according to the comparison result; An alarm device connected to the multi-view cameras and the verification unit respectively for receiving the alarm signal, and in the state of receiving the alarm signal, switching the photographing positions of the multi-view cameras photographing the same monitoring area and emitting an audible and visual alarm; Wherein, the photographing position is the corresponding position area in the monitoring area photographed by the multi-view camera.

2. The intelligent video surveillance system based on multi-camera data fusion according to claim 1, wherein The verification unit determines the picture-in-picture features in the video streams photographed by the cameras in each view of a single multi-view camera, wherein, If the picture photographed by a single-view camera is completely included in the picture photographed by another single-view camera, the features of the target object existing in the picture photographed by the single-view camera are the picture-in-picture features; If the picture photographed by the single-view camera partially overlaps with the picture photographed by the other single-view camera, the features of the target object existing in the overlapping part of the pictures are the picture-in-picture features; The picture-in-picture features are the visual features and motion features of the same target object that appear in two single-view cameras at the same time; the visual features include the shape features, color features, and texture features of the target object; the motion features include the motion direction and motion speed of the target object.

3. The intelligent video monitoring system based on multi-camera data fusion according to claim 2, wherein The verification unit compares the picture-in-picture features in the video streams photographed by the cameras in each view of a single multi-view camera, wherein, If the target object does not move within the first preset time, the verification unit compares the visual features of the target object in the video streams photographed by the cameras in each view; If the target object moves within the first preset time, the verification unit compares the motion features of the target object in the video streams photographed by the cameras in each view; Wherein, the first preset time is positively correlated with the number of single-view cameras included in a single multi-view camera.

4. The intelligent video surveillance system based on multi-camera data fusion according to claim 3, wherein, When the verification unit compares the visual features of the target object in the video streams photographed by the cameras in each view, it determines whether to send the alarm signal according to the comparison result, wherein, If the visual features are inconsistent, the verification unit emits the alarm signal.

5. The intelligent video monitoring system based on multi-camera data fusion according to claim 4, characterized in that The verification unit calculates the motion rate of the target object in the video streams captured by each monocular camera while comparing the motion features of the target object in the video streams captured by each monocular camera. Among them, If the motion rates are inconsistent, the verification unit emits the alarm signal; If the motion rates are consistent, the motion rate is compared with a preset rate. In the state where the motion rate is greater than or equal to the preset rate, the verification unit emits the alarm signal; The preset rate is positively correlated with the height of the target object.

6. The intelligent video monitoring system based on multi-camera data fusion according to claim 5, characterized in that The rotation unit extracts any one monocular camera from a single multi - monocular camera, sets the monocular camera as the rotation monocular camera, and adjusts the shooting position of the rotation monocular camera to be consistent with the shooting positions of the remaining fixed monocular cameras in the same multi - monocular camera in turn at periodic time intervals, The periodic time interval is positively correlated with the number of monocular cameras included in a single multi - monocular camera.

7. The intelligent video monitoring system based on multi-camera data fusion according to claim 6, characterized in that The verification unit respectively compares the video stream captured by the rotation monocular camera with the video streams captured by each fixed monocular camera in the same periodic time interval, and determines whether to emit an alarm signal according to the comparison result. Among them, If the picture - in - picture feature of the target object in the video stream captured by the rotation monocular camera is inconsistent with the picture - in - picture feature of the same target object in the video stream captured by the fixed monocular camera, the verification unit emits the alarm signal.

8. The intelligent video surveillance system based on multi-camera data fusion according to claim 7, characterized in that The rotation unit selects a single fixed monocular camera to replace the rotation monocular camera. Among them, The rotation unit replaces the rotation monocular camera and the fixed monocular camera in the state where the shooting position of the rotation monocular camera is adjusted to be consistent with that of the fixed monocular camera.

9. The intelligent video monitoring system based on multi-camera data fusion according to claim 8, characterized in that, The verification unit compares the picture - in - picture features of the video streams captured by the fixed monocular camera before and after replacement within the second preset time, and determines whether to emit an alarm signal according to the comparison result. Among them, If the picture - in - picture features of the video streams captured by the fixed monocular camera before and after replacement within the second preset time are inconsistent, the verification unit emits the alarm signal; The second preset time is positively correlated with the number of monocular cameras included in a single multi - monocular camera.

10. The intelligent video surveillance system based on multi-camera data fusion according to claim 9, characterized in that, In the state of receiving the alarm signal, the alarm device switches the shooting positions of each multi - monocular camera shooting the same monitoring area, emits an audible and visual alarm, and closes the multi - monocular camera with inconsistent picture - in - picture features at the same time.

Images