Automatic video monitoring system applied to construction site

Abstract

The invention relates to the technical field of monitoring systems, in particular to an automatic video monitoring system applied to a construction site, which comprises a monitoring video frame acquisition module, a construction site core area acquisition module, a construction core area offset degree acquisition module and a camera angle adjustment module, analyzing change conditions of image information of different cameras between adjacent monitoring video frames to obtain a plurality of construction site core areas; analyzing the similar condition of the shape change of the core area of the construction site among different cameras, and dividing all the cameras into a plurality of construction monitoring camera groups; in the same construction monitoring camera group, the moving mode of the construction site core area with the similar area is analyzed, the construction core area offset degree is obtained, and then the twisting angle of the camera is adjusted in a self-adaptive mode. The monitoring efficiency of the construction site is improved, and the potential safety hazard of the construction site is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of monitoring systems, and in particular to an automated video monitoring system applied to a construction site. Background Art

[0002] The automated video surveillance system at the construction site can achieve all-weather and all-round supervision of the construction site. By installing cameras at different monitoring points on the construction site, different construction areas within the construction site can be observed in real time to monitor various conditions on the construction site, such as the activities of staff, the operating status of equipment, and the supply of building materials, etc., so as to achieve centralized management and dynamic monitoring and eliminate safety hazards to the greatest extent.

[0003] When monitoring construction sites, existing monitoring methods usually use fixed time intervals to adjust the camera angle; however, in actual environments, the construction area that the camera focuses on may be offset due to the influence of random events such as construction personnel or the environment, resulting in the fixed camera angle adjusted at fixed time intervals not being able to adjust the angle of the construction area in a timely manner, reducing the monitoring accuracy of the monitoring video, reducing the monitoring efficiency of the construction site, and increasing the safety hazards of the construction site. Summary of the invention

[0004] The present invention provides an automated video monitoring system for construction sites to solve the existing problem that the construction area focused on by a camera may be offset due to the influence of random events such as construction personnel or the environment, resulting in a fixed camera angle adjusted at a fixed time interval failing to adjust the angle of the focused construction area in a timely manner.

[0005] The present invention adopts the following technical solution for an automated video monitoring system applied to a construction site: Includes the following modules: A monitoring video frame acquisition module, used to obtain a number of monitoring video frames from different cameras, wherein the monitoring video frames contain a construction shape descriptor; The construction site core area acquisition module is used to analyze the changes in image information between adjacent surveillance video frames of different cameras based on the construction shape descriptor, divide the surveillance video frames into regions, and obtain several construction site core areas for each camera; The construction core area offset acquisition module is used to analyze the similarity of the shape changes of the construction site core area between different cameras based on the construction shape descriptor, and divide all cameras into several construction monitoring camera groups; in the same construction monitoring camera group, the movement mode of the construction site core area with similar area is analyzed to obtain the construction core area offset of different cameras; The camera angle adjustment module is used to adaptively adjust the twisting angle of the camera according to the offset of the core construction area.

[0006] Preferably, the method for obtaining the core area of ​​the construction site is: For any camera, based on the construction shape descriptor, the difference in distance between the image information variation range and the camera center in adjacent surveillance video frames under the camera is analyzed to obtain the construction information fluctuation of the surveillance video frame; according to the construction information fluctuation, the core area of ​​the construction site is divided from the surveillance video frame.

[0007] Preferably, the method for obtaining the construction information fluctuation is: According to the construction shape descriptor, the construction information change area in each monitoring video frame is obtained; the distance between the internal center of the construction information change area and the camera center is analyzed to obtain the construction information fluctuation degree of the monitoring video frame.

[0008] Preferably, the method for obtaining the construction monitoring camera group is: According to the construction shape descriptor, the shape matching of the construction shape descriptors between different cameras is analyzed to obtain the shape matching degree of the construction core area between different comparison cameras; according to the shape matching degree of the construction core area, all cameras are divided into several construction monitoring camera groups.

[0009] Preferably, the method for obtaining the shape matching degree of the construction core area is: Any camera is taken as the target camera, and the cameras other than the target camera are taken as the reference cameras. For any reference camera, the DTW distance of the construction shape descriptor between the target camera and the reference camera is taken as the shape matching degree of the construction core area between the target camera and the reference camera.

[0010] Preferably, the method for obtaining the deviation of the construction core area is: For any construction monitoring camera group, analyze the proximity of the core area of ​​the construction site between different cameras in the construction monitoring camera group, and divide the construction monitoring camera group into several camera groups with the same construction range; For any camera in any camera group of the same construction range, analyze the offset direction and offset distance of the camera center in continuous monitoring video frames to obtain the construction monitoring screen centrality of the camera; Analyze the area distribution of the core area of ​​the construction site of the camera to obtain the distribution breadth of the camera's monitoring center; The centrality of the construction monitoring screen is combined with the distribution breadth of the monitoring center to obtain the camera's construction core area offset.

[0011] Preferably, the method for obtaining the camera groups of the same construction range is: Obtain the core area chain code sequence of each camera in the construction monitoring camera group; analyze the data differences between the core area chain code sequences, and divide the construction monitoring camera group into several camera groups of the same construction range.

[0012] Preferably, the method for obtaining the centrality of the construction monitoring screen is: According to the coordinate movement trend of the center of the core area of ​​the construction site in the continuous monitoring video frame of the camera, the construction camera center offset vector of the camera is obtained; by comparing the distance transformation between the construction camera center offset vector and the center of the core area of ​​the construction site, the centrality of the camera's construction monitoring screen is obtained.

[0013] Preferably, the method for acquiring the distribution breadth of the monitoring centers is: The overall fluctuation of the camera's core area of ​​the construction site in all surveillance video frames is analyzed to obtain the distribution breadth of the camera's monitoring center.

[0014] Preferably, after calculating the construction core area deviation, the method further includes: The deviation of the construction core area is normalized.

[0015] The beneficial effects of the technical solution of the present invention are as follows: analyzing the changes in image information between adjacent monitoring video frames of different cameras, dividing the monitoring video frames into regions to obtain the core area of ​​the construction site; wherein the core area of ​​the construction site is used to reflect the area with frequent activities on the construction site, so that the area that may contain safety hazards during the construction process is obvious; then analyzing the similarities in the shape changes of the core area of ​​the construction site between different cameras, and dividing all cameras into several construction monitoring camera groups; wherein the construction monitoring camera groups are used to reflect the cameras that focus on the same construction details, so that the cameras that monitor the same local site details are classified into one category, so that the monitoring effect of focusing on the same construction details is better ; Finally, in the same construction monitoring camera group, the movement mode of the core area of ​​the construction site with a similar area is analyzed to obtain the offset of the camera's construction core area; wherein the offset of the construction core area is used to reflect the degree of non-overlap between the camera's monitoring center and the construction area that needs to be paid attention to, making the camera's monitoring offset phenomenon more obvious; the present invention analyzes the changes in construction information in the key focus area when different cameras monitor the construction site, judges the offset state between the key focus area and the camera's monitoring center, and thus timely adjusts the camera's monitoring angle, thereby improving the monitoring accuracy of the monitoring video, improving the monitoring efficiency of the construction site, and reducing the safety hazards of the construction site. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] 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 work.

[0017] Figure 1 The present invention is a structural block diagram of an automated video monitoring system applied to a construction site. DETAILED DESCRIPTION

[0018] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation, structure, features and effects of an automated video monitoring system for construction sites proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0019] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0020] The following is a detailed description of a specific solution of an automated video monitoring system applied to a construction site provided by the present invention in conjunction with the accompanying drawings.

[0021] See also Figure 1 , which shows a structural block diagram of an automated video monitoring system applied to a construction site provided by an embodiment of the present invention, the system includes the following modules: The monitoring video frame acquisition module 101 obtains a number of monitoring video frames from different cameras.

[0022] It should be noted that when monitoring a construction site, existing monitoring methods usually use a fixed time interval to adjust the camera angle; however, in an actual environment, the construction area that the camera focuses on may be offset due to the influence of random events such as construction personnel or the environment, resulting in the fixed camera angle adjusted at a fixed time interval not being able to adjust the angle of the construction area that is focused on in a timely manner, thereby reducing the monitoring accuracy of the monitoring video, reducing the monitoring efficiency of the construction site, and increasing the safety hazards of the construction site.

[0023] In a specific implementation of an embodiment of the present invention, a method for acquiring monitoring video frames is as follows: the monitoring video frames of all cameras in the same construction site within the past hour are retrieved from a construction video database; taking any two adjacent monitoring video frames as an example, the first monitoring video frame is used as a reference monitoring video frame, and the second monitoring video frame is used as a target monitoring video frame; in a frame difference map after subtracting the two monitoring video frames, a grayscale change threshold is preset. , change the gray value greater than The pixel points are retained to obtain all the retained pixel points; the coordinate positions occupied by all the retained pixel points are used as the changed pixel points; the chain code sequence of all the changed pixel points is obtained, and the chain code sequence is used as the construction shape descriptor of the target monitoring video frame; the construction shape descriptor of all monitoring video frames is obtained. This example is described as an example, and this embodiment is not specifically limited. It may depend on the specific implementation situation.

[0024] It is particularly noted that the acquisition of chain code sequences is a well-known technology and will not be described in detail in this embodiment. In addition, the number of cameras in this embodiment is described by taking 10 as an example, wherein the number of cameras may be determined according to the specific implementation situation and will not be described in detail in this embodiment.

[0025] It should be noted that this embodiment does not consider the construction shape descriptor of the first monitoring video frame.

[0026] So far, several monitoring video frames of different cameras are obtained through the above method.

[0027] The construction site core area acquisition module 102 analyzes the changes in image information between adjacent monitoring video frames of different cameras according to the construction shape descriptor, divides the monitoring video frames into regions, and obtains a number of construction site core areas for each camera.

[0028] It should be noted that in the construction scene, the camera will continuously record the construction status in the construction site. Usually, there will be personnel or machines performing construction operations on the construction site, and the rest of the construction site area is basically static because it is not affected by the operation of personnel or machines. Since the construction operation process is a dynamic process, there will be certain changes in the image information at different times. The areas with these changes are mainly areas where construction activities are frequent on the construction site, and these areas need to be monitored. Therefore, the changes in image information between adjacent monitoring video frames of different cameras can be analyzed based on the construction shape descriptor, and the monitoring video frames can be divided into regions to obtain several core areas of the construction site for each camera.

[0029] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the core area of ​​the construction site is as follows: for any camera, based on the construction shape descriptor, the difference in distance between the image information variation range and the camera center in adjacent monitoring video frames under the camera is analyzed to obtain the construction information fluctuation degree of the monitoring video frame; according to the construction information fluctuation degree, the core area of ​​the construction site is divided from the monitoring video frame. The specific process is as follows: It should be noted that when construction workers or machines are performing construction operations, the range of changes in the affected image information is generally relatively fixed; if construction workers change shifts normally or enter the construction site by mistake, the original range of changes in the image information, which was originally relatively fixed in size, will fluctuate greatly. Normally, the camera will aim the camera center at the construction area to observe changes in construction operations in the construction area and a certain area around it. If construction workers change shifts normally or enter the construction site by mistake, the affected construction area will be a certain distance from the camera center; therefore, the construction information change area in the monitoring video frame can be obtained based on the construction shape descriptor; the distance between the internal center of the construction information change area and the camera center is analyzed to obtain the construction information fluctuation degree of the monitoring video frame; the greater the construction information fluctuation degree, the more obvious the fluctuation of the construction information change area in the monitoring video frame difference image, reflecting that the monitoring video frame difference image is more likely to contain construction anomalies. The specific process is as follows: Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the construction information fluctuation is: according to the construction shape descriptor, obtain the construction information change area in each monitoring video frame; analyze the distance between the internal center of the construction information change area and the camera center to obtain the construction information fluctuation of the monitoring video frame. The specific process is as follows: Taking any surveillance video frame of the camera as an example, the image area occupied by the construction shape descriptor in the surveillance video frame is taken as the construction information change area; the normalized value of the Euclidean distance between the center of the construction information change area and the center of the surveillance video frame is taken as the construction information fluctuation degree of the surveillance video frame. The acquisition of the Euclidean distance is a well-known technology and will not be repeated in this embodiment.

[0030] It is particularly noted that the normalization process in this embodiment is based on The normalization function can be determined according to the specific implementation situation, and will not be described in detail in this embodiment.

[0031] It should be noted that if the fluctuation of the construction information is greater, it means that the fluctuation of the regional change of the construction information in the monitoring video frame difference image is more obvious, which reflects that the monitoring video frame difference image is more likely to contain construction anomalies.

[0032] Furthermore, a construction information fluctuation threshold is preset , only retain construction information with fluctuation greater than The monitoring video frames and the construction information fluctuation is greater than The construction information change area in the monitoring video frame is used as the core area of ​​the construction site; and all the core areas of the construction site of the camera are obtained.

[0033] It is particularly noted that, in this embodiment, This example is described as an example, and this embodiment is not specifically limited. It may depend on the specific implementation situation.

[0034] It should be noted that the cameras described later in this embodiment are all cameras covering the core area of ​​the construction site.

[0035] So far, several construction site core areas of the camera are obtained through the above method.

[0036] The construction core area offset acquisition module 103 analyzes the similarities of shape changes of the core area of ​​the construction site between different cameras based on the construction shape descriptor, and divides all cameras into several construction monitoring camera groups; in the same construction monitoring camera group, the movement mode of the core area of ​​the construction site with similar area is analyzed to obtain the construction core area offset of different cameras.

[0037] It should be noted that in an actual construction site, multiple cameras will be installed to conduct video surveillance of different local construction areas in the construction site; and for cameras that focus on monitoring the same local construction area, the camera centers of such cameras are basically in the same construction area; therefore, based on the construction shape descriptor, we can analyze the similarities of shape changes in the core areas of the construction site between different cameras and divide all cameras into several construction monitoring camera groups.

[0038] It should be further explained that, for the same construction monitoring camera group, the internal cameras correspond to the core areas of the construction site and show the same range of changes as a whole, which to a certain extent reflects the range of changes in the core areas of the construction site. Within this range of changes, the content of the construction area focused on by the camera's imaging center will shift, forming different movement modes. Therefore, in the same construction monitoring camera group, the movement modes of the core areas of the construction site with similar areas can be analyzed to obtain the offset degrees of the construction core areas of different cameras. The greater the offset degree of the construction core area, the greater the offset between the camera's imaging center and the construction area that requires core monitoring, reflecting that the angle at which the camera needs to be adjusted is greater.

[0039] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the construction monitoring camera group is: according to the construction shape descriptor, the shape matching of the construction shape descriptors between different cameras is analyzed to obtain the shape matching degree of the construction core area between different reference cameras; according to the shape matching degree of the construction core area, all cameras are divided into several construction monitoring camera groups. The specific process is as follows: It should be noted that the construction shape descriptor in this embodiment is a chain code sequence that can describe the shape of the area. If the construction shape descriptors between different cameras are closer, it means that the regional morphology of the core area of ​​the construction site is more similar, which reflects that the construction area focused on by the corresponding camera belongs to the same site area.

[0040] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the shape matching degree of the construction core area is: taking any one camera as the target camera, and taking cameras other than the target camera as the reference camera; for any reference camera, taking the DTW distance of the construction shape descriptor between the target camera and the reference camera as the shape matching degree of the construction core area between the target camera and the reference camera.

[0041] It is particularly noted that the acquisition of the DTW distance is a well-known content of the DTW (Dynamic Time Warping) dynamic time warping algorithm, and will not be described in detail in this embodiment.

[0042] It should be noted that if the shape matching degree of the construction core area is greater, it means that the regional morphology of the core area of ​​the construction site is more similar, reflecting that the construction area focused on by the corresponding camera belongs to the same site area.

[0043] Furthermore, the shape matching degree of the construction core area is used as the distance metric, and the number of clusters is preset. , according to the distance metric and , cluster the cameras to obtain several clusters; each cluster is used as a construction monitoring camera group.

[0044] It is particularly noted that, in this embodiment, This example is described as an example, and this embodiment is not specifically limited. It can be determined according to the specific implementation situation. In addition, the process of clustering according to the distance metric and the number of clusters is a well-known content of the K-means clustering algorithm, and will not be repeated in this embodiment.

[0045] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the offset of the construction core area is: for any construction monitoring camera group, analyze the proximity of the core area area of ​​the construction site between different cameras in the construction monitoring camera group, and divide the construction monitoring camera group into several camera groups of the same construction range; for any camera in any camera group of the same construction range, analyze the offset direction and offset distance of the camera center in continuous monitoring video frames to obtain the centrality of the camera's construction monitoring screen; analyze the area distribution of the core area of ​​the camera's construction site to obtain the distribution breadth of the camera's monitoring center; combine the construction monitoring screen centrality with the distribution breadth of the monitoring center to obtain the offset of the camera's construction core area. The specific process is as follows: 1. Obtain camera groups of the same construction scope.

[0046] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the same type of construction scope camera group is: obtaining the core area chain code sequence of each camera in the construction monitoring camera group; analyzing the data differences between the core area chain code sequences, and dividing the construction monitoring camera group into several same type of construction scope camera groups. The specific process is as follows: The chain code sequence of the core area of ​​the largest construction site of each camera in the surveillance camera group is used as the core area chain code sequence; the difference in the number of chain codes contained in different core area chain code sequences is used as the distance metric, and the number of reference clusters is preset. , according to the distance metric and , cluster all cameras in the surveillance camera group to obtain several clusters; each cluster is used as a camera group of the same construction range. This example is described as an example, and this embodiment is not specifically limited. It may depend on the specific implementation situation.

[0047] 2. Obtain the centrality of the construction monitoring screen.

[0048] It should be noted that if the camera's imaging center is offset, the direction of the offset and the distance of the offset jointly characterize different offset modes; therefore, the offset direction and distance of the camera's imaging center in continuous monitoring video frames can be analyzed to obtain the camera's construction monitoring screen centrality; where the greater the construction monitoring screen centrality, the greater the degree to which the camera's imaging center is in the center of the core construction area of ​​concern, reflecting that the camera's monitoring focus is more in the local construction area that needs special attention.

[0049] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the centrality of the construction monitoring screen is: according to the coordinate movement trend of the center of the core area of ​​the construction site in the continuous monitoring video frame of the camera, the construction camera center offset vector of the camera is obtained; and the distance transformation between the construction camera center offset vector and the center of the core area of ​​the construction site is compared to obtain the centrality of the construction monitoring screen of the camera. The specific process is as follows: Get the center point coordinates of each construction site core area of ​​the camera; perform vector projection based on all center point coordinates to obtain projection values ​​of different two-dimensional vectors, and use the two-dimensional vector with the largest projection value as the construction camera center offset vector; use the arc tangent value of the ratio between the second element and the first element in the construction camera center offset vector as the first degree value, and compare the first degree value with The ratio of is taken as the centrality of the construction monitoring screen. Each two-dimensional vector contains two element values, and each two-dimensional vector corresponds to a projection value.

[0050] It is particularly noted that the process of obtaining a two-dimensional vector and a projection value according to a coordinate point is a well-known content of a PCA (Principal Component Analysis) algorithm, and will not be described in detail in this embodiment.

[0051] It should be noted that if the centrality of the construction monitoring picture is greater, it means that the camera's camera center is more in the center of the core construction area, which reflects that the camera's monitoring focus is more in the local construction area that needs special attention.

[0052] 3. Obtain the distribution breadth of monitoring centers.

[0053] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the monitoring center distribution breadth is: analyzing the overall fluctuation of the camera's construction site core area in all monitoring video frames to obtain the camera's monitoring center distribution breadth. The specific process is as follows: The average of the core area of ​​all construction sites of the camera is taken as the distribution breadth of the monitoring center of the camera.

[0054] 4. Obtain the offset of the core construction area.

[0055] The normalized value of the product of the centrality of the construction monitoring picture of the camera and the distribution breadth of the monitoring center is used as the construction core area offset of the camera.

[0056] It should be noted that if the offset of the construction core area is larger, it means that the camera's camera center is offset more from the construction area that needs core monitoring, which reflects that the angle that the camera needs to be adjusted is larger.

[0057] So far, the construction core area offset of different cameras is obtained through the above method.

[0058] The camera angle adjustment module 104 adaptively adjusts the twisting angle of the camera according to the offset of the construction core area.

[0059] Taking any camera as an example, the angle adjustment value of the camera can be calculated by the following formula: In the formula, Indicates the angle adjustment value of the camera; Indicates the offset of the construction core area of ​​the camera; Represents the inverse sine function.

[0060] Furthermore, the camera is adjusted counterclockwise in the horizontal direction. After that, the automatic adjustment of each camera angle is completed.

[0061] At this point, this embodiment is completed.

[0062] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An automated video monitoring system applied to a construction site, characterized in that: The system includes the following modules: The monitoring video frame acquisition module is used to obtain a number of monitoring video frames from different cameras, wherein the monitoring video frames contain a construction shape descriptor; obtaining the construction shape descriptor includes: retrieving monitoring video frames from all cameras in the same construction site within the past hour from the construction video database; taking any two adjacent monitoring video frames as an example, taking the first monitoring video frame as a reference monitoring video frame, and taking the second monitoring video frame as a target monitoring video frame; in the frame difference map after subtracting the two monitoring video frames, presetting a grayscale change threshold, retaining the pixel points whose grayscale value change is greater than, and obtaining all the retained pixel points; taking the coordinate positions occupied by all the retained pixel points as the changed pixel points; obtaining the chain code sequence of all the changed pixel points, and taking the chain code sequence as the construction shape descriptor of the target monitoring video frame; obtaining the construction shape descriptors of all monitoring video frames; The construction site core area acquisition module is used to analyze the changes in image information between adjacent surveillance video frames of different cameras based on the construction shape descriptor, divide the surveillance video frames into regions, and obtain several construction site core areas for each camera; The construction core area offset acquisition module is used to analyze the similarity of the shape changes of the construction site core area between different cameras based on the construction shape descriptor, and divide all cameras into several construction monitoring camera groups; in the same construction monitoring camera group, the movement mode of the construction site core area with similar area is analyzed to obtain the construction core area offset of different cameras; The method for obtaining the construction monitoring camera group is as follows: according to the construction shape descriptor, analyzing the shape matching of the construction shape descriptors between different cameras, and obtaining the shape matching degree of the construction core area between different comparison cameras; according to the shape matching degree of the construction core area, dividing all cameras into a number of construction monitoring camera groups; The method for obtaining the shape matching degree of the construction core area is: Any camera is taken as the target camera, and the cameras other than the target camera are taken as the reference cameras. For any reference camera, the DTW distance of the construction shape descriptor between the target camera and the reference camera is taken as the shape matching degree of the construction core area between the target camera and the reference camera. The camera angle adjustment module is used to adaptively adjust the twisting angle of the camera according to the offset of the core construction area.

2. The automated video monitoring system for a construction site according to claim 1, characterized in that: The method for obtaining the core area of ​​the construction site is as follows: For any camera, based on the construction shape descriptor, the difference in distance between the image information variation range and the camera center in adjacent surveillance video frames under the camera is analyzed to obtain the construction information fluctuation of the surveillance video frame; according to the construction information fluctuation, the core area of ​​the construction site is divided from the surveillance video frame.

3. The automated video monitoring system for a construction site according to claim 2, characterized in that: The method for obtaining the construction information fluctuation degree is: According to the construction shape descriptor, the construction information change area in each monitoring video frame is obtained; the distance between the internal center of the construction information change area and the camera center is analyzed to obtain the construction information fluctuation degree of the monitoring video frame.

4. The automated video monitoring system for a construction site according to claim 1, characterized in that: The method for obtaining the offset of the construction core area is: For any construction monitoring camera group, analyze the proximity of the core area of ​​the construction site between different cameras in the construction monitoring camera group, and divide the construction monitoring camera group into several camera groups with the same construction range; For any camera in any camera group of the same construction range, analyze the offset direction and offset distance of the camera center in continuous monitoring video frames to obtain the construction monitoring screen centrality of the camera; Analyze the area distribution of the core area of ​​the construction site of the camera to obtain the distribution breadth of the camera's monitoring center; The centrality of the construction monitoring screen is combined with the distribution breadth of the monitoring center to obtain the camera's construction core area offset.

5. The automated video monitoring system for a construction site according to claim 4, characterized in that: The method for obtaining the camera groups of the same construction range is as follows: Obtain the core area chain code sequence of each camera in the construction monitoring camera group; analyze the data differences between the core area chain code sequences, and divide the construction monitoring camera group into several camera groups of the same construction range.

6. The automatic video monitoring system applied to a construction site according to claim 4, characterized in that: The method for obtaining the centrality of the construction monitoring screen is: According to the coordinate movement trend of the center of the core area of ​​the construction site in the continuous monitoring video frame of the camera, the construction camera center offset vector of the camera is obtained; by comparing the distance transformation between the construction camera center offset vector and the center of the core area of ​​the construction site, the centrality of the camera's construction monitoring screen is obtained.

7. The automated video monitoring system for a construction site according to claim 4, characterized in that: The method for obtaining the distribution breadth of the monitoring center is: The overall fluctuation of the camera's core area of ​​the construction site in all surveillance video frames is analyzed to obtain the distribution breadth of the camera's monitoring center.

8. The automated video monitoring system for a construction site according to claim 1, characterized in that: After calculating the deviation of the construction core area, it also includes: The deviation of the construction core area is normalized.

Images