KR102024695B1 - System and method for modeling surveillance camera layout - Google Patents

Abstract

The present invention discloses a surveillance camera batch modeling system and method. An embodiment of the present invention, a range setting unit for setting an analysis range in the map data of a predetermined area, a grid generation unit for forming a grid including a plurality of unit grids each having a specific numerical information in the analysis range, the plurality A map generator for weighting each of the plurality of unit grids according to the numerical information of each of the unit grids and forming a map capable of classifying regions according to the size of the weights, and arranged to correspond to a specific position in the map It provides a surveillance cameras batch modeling system including a model generator for generating one or more batch modeling information of one or more surveillance cameras. According to the present invention, it is possible to model the optimal arrangement of the surveillance camera additionally installed in the region based on various data of the specific region.

Description

SYSTEM AND METHOD FOR MODELING SURVEILLANCE CAMERA LAYOUT}

The present invention relates to a system and method for modeling an arrangement of a surveillance camera additionally installed in a corresponding region based on various data of a specific region.

The demand for security and surveillance technology using cameras is increasing. Camera-based surveillance provides a potential solution for preventing and reducing crime and violations. Accordingly, the use of camera surveillance networks is increasing, and camera or video surveillance is regarded as a powerful system used to maintain the order and security of public spaces.

On the other hand, camera placement in a camera surveillance network has a big impact on the quality and efficiency of surveillance. Proper placement is related to camera properties such as range, field of view, resolution, cost of use, type of camera (visible light, infrared light, etc.). The most important of these is the deployment environment. Indeed, the effectiveness of video surveillance systems is known from overseas studies to rely heavily on the physical placement of cameras.

An example of a representative camera surveillance network is a surveillance camera (CCTV). CCTV application types are generally divided into CCTVs for collecting traffic information and breaking traffic laws, CCTVs for detection of wanted vehicles, CCTVs for illegally dumping garbage, and CCTVs for crime prevention. Traffic-related CCTVs do not photograph or record specific vehicles or individuals in order to identify traffic flows and traffic volume, but traffic information collecting CCTVs that monitor real-time monitoring, traffic regulations to control speed violations, signal violations, and bus lane violations. It is divided into CCTV for violation control and CCTV for parking control for unmanned parking. Arrangement detection is part of the Automatic Vehicle Number Identification system, which can be installed on major roads with high vehicle movements to automatically read the arranging vehicle number and quickly detect criminals. When a vehicle in transit passes through the system, a high-precision infrared camera quickly sorts out a wanted vehicle and a wanted vehicle, and displays it automatically at a checkpoint terminal in front of it. Although CCTV for garbage dumping control is being installed for garbage dumping control at the borough level, the issue of effectiveness continues to be raised. Security CCTV is operated by local governments nationwide.

As shown in FIG. 1, according to the 2016 Administrative Statistics Annual Report, the total number of CCTV installations and operations in domestic public institutions was 65,530 in 2014 and 73,9232 in 2015, except for the installation of private places. It is an increase of 8,4202 units. Among them, 34,758 were CCTVs installed for crime prevention, an increase of 49,320 from last year. CCTV has a great effect on crime prevention and criminal arrest. For example, the National Security Agency and the National Police Agency reported that the number of five violent crimes occurred within 100m radius of the school zone in the first half of 2016, and the number of crimes was reduced from 1655 in the first half of 2013 before the CCTV installation to 1091 in the first half of 2015. This is why CCTV is continuously expanding. Nationwide, local governments are expanding the number of CCTV installations to prevent blind spots. In smart cities, various projects are being developed in the direction of increasing effectiveness through the integrated operation of CCTV, which is functionally separated.

In this regard, the expansion of CCTV brings additional problems such as cost and personnel in the areas of operation and management. Therefore, in consideration of the limited budget and operating cost, it is necessary to select the optimal CCTV installation location in the physical environment within a specific space.

In general, the method of disposing a sensor such as a camera is a research topic attracting interest from various research groups around the world. The main purpose of camera placement is to define the viewpoint position to select the most informative scene of the scene of interest. Camera placement is associated with a variety of issues such as surveillance, tracking and scene monitoring, three-dimensional reconstruction and robotic sensors and cultural property or museum surveillance.

The most intuitive way to position the camera is to evenly place the camera network in the region of interest. This method is useful for maximizing coverage in uncomplicated areas. However, in most sites the actual physical environment is much more complex and difficult. In practice, camera placement issues can be approached from different perspectives, depending on the purpose of the system. Thus, the approach of camera placement analysis presented in various studies so far can be classified into two main categories: the goal-based approach and the perspective-based approach.

A goal-based approach is to find the camera placement according to the object of interest. Objects can be objects, events or phenomena that can evolve in space and time. The goal-based approach focuses primarily on the characteristics of the object rather than on the level of coverage.

In general, this method suggests an automatic sensor placement method based on 3D CAD data files. The purpose of this method was developed in the field of model-based robot vision by finding the best viewing position of the vision sensor using appropriate parameters such as position, orientation and optical settings. Each time point must meet predefined constraints primarily related to the physical and optical properties of the sensor. The main goal is to monitor and analyze the important functions of characterizing objects by moving one sensor (using a robot) from one location to another. Similar studies have been conducted regarding the three-dimensional reconstruction of objects and spaces. In Border's paper, the authors introduce analytical formulas to monitor the locomotive's movement path through scenes of interest. This method focuses on maximizing the observability of the target and capturing its movement. The authors specify that only a minimum set of a priori knowledge of the target is required and no prior knowledge of the actual environment in which the event occurs. Based on the results of these studies, we developed a solution that defines the layout of surveillance camera networks by combining fixed cameras and mobile cameras. The authors suggest using a fixed camera to observe the entire scene to extract the movement of the target, such as a person. The acquired field knowledge is then used to determine the location of the mobile camera according to ground coverage and resolution quality criteria.

The view-based method aims to optimize the coverage by assuming that the physical characteristics of the environment where the camera deployment will occur are known in advance. The definition of coverage varies, but the general definition is suggested by the quality of surveillance that can be provided by the sensor network and how well the area of interest is monitored.

One of the well-studied topics related to the perspective-based approach is the research related to the positioning of museum surveillance cameras. This is a study to determine the number of cameras that monitor every point in a museum exhibition room, a polygonal area, and finds the optimal number and location of cameras as observers in the view to satisfy or maximize the range.

In the same vein, other studies have proposed many optimization methods for sensor placement. Genetic algorithm methodologies have been proposed to optimize sensor placement, and approaches based on parallel evolutionary optimization techniques for selecting and aligning candidate sites for sensor networks have also been developed. Another study aims to find the optimal configuration of the network by deploying sensors based on the Simulated Annealing technique. All three methods apply to very simple environments that do not reflect the complexity and diversity of the actual viewing environment.

The present invention has been made to solve the above problems, the main technical problem to be achieved by the present invention is to provide a system and method for modeling the arrangement of the surveillance cameras are additionally installed in the area based on a variety of data of a particular area. will be.

The technical problem to be achieved by the present invention is not limited to the technical problem mentioned above, and other technical problems not mentioned above may be clearly understood by those skilled in the art from the following description. There will be.

In order to solve the above technical problem, an embodiment of the present invention provides a range setting unit for setting an analysis range in map data of a predetermined region, and forming a grid including a plurality of unit grids each having numerical information in the analysis range. A grid generation unit and a map generation unit which weights each of the plurality of unit grids according to numerical information of each of the plurality of unit grids, and forms a map capable of area division according to the size of the weights. In addition, the present invention provides a surveillance camera arrangement modeling system including a model generator for generating one or more arrangement modeling information of at least one surveillance camera arranged to correspond to a specific position in the map.

In one embodiment of the present invention, the map data includes terrain information, coordinate information, building information, building shape information, surveillance camera installation location information and surveillance camera installation number information of the predetermined region, and the range setting unit is The analysis range may be set using terrain information, building information, and coordinate information included in map data.

In one embodiment of the present invention, the plurality of grids are arranged in a shape having an interval of 1m to 10m, the numerical information is any one of the corresponding floating population numerical data, crime rate numerical data and traffic numerical value data, the weight The size of may be proportional to the size of the data value of the numerical information.

In one embodiment of the present invention, the grid generation unit forms a grid including a plurality of unit grids each having numerical information within the analysis range, extracts the building shape information and the planar area included in the analysis range, The map generation unit targets a plurality of unit grids formed in an area other than the planar area of the analysis range, and is formed in an area other than the planar area according to numerical information of each of the plurality of unit grids formed in the area other than the planar area. A weight may be assigned to each of the plurality of unit grids, and a map capable of distinguishing regions according to the magnitude of the weight may be formed.

In one embodiment of the present invention, the map generation unit, weighting each of the plurality of unit grids according to the numerical information of each of the plurality of unit grids, each of the plurality of unit grids according to the size of the weight It is possible to form heat maps in the form of different colors.

In one embodiment of the present invention, the one or more batch modeling information formed through the model generation unit includes the data of the number of one or more surveillance cameras are arranged to correspond to a specific position in the map, and among the one or more batch modeling information The apparatus may further include a model recommending unit configured to provide modeling information having the smallest number data value of the surveillance camera included in each of the one or more batch modeling information.

In one embodiment of the present invention, the model generation unit, the number information of one or more surveillance cameras arranged to correspond to a specific position in the map, the arrangementable area information, the effective viewing angle information, the effective viewing distance information and the effective surveillance area information Based on the plurality of arrangement modeling information of one or more surveillance cameras arranged to correspond to a specific position in the map can be generated.

In one embodiment of the present invention, the one or more surveillance cameras arranged to correspond to a specific position in the map each has a horizontal viewing angle of 0 ° to 360 °, a vertical viewing angle of 90 ° to -90 °, and a preset coverage range. And generated by the model generator based on the number information of the surveillance cameras arranged in the Voronoi region VOR (Si) calculated by Equation 1 below among one or more surveillance cameras arranged to correspond to a specific position in the map. The apparatus may further include a model recommendation unit that provides specific batch modeling information among the one or more batch modeling information.

[Equation 1]

,

,

Here, S, Si and Sj are a set of buildings corresponding to the specific position, and (Si∪Sj) ∈S, Si = {s1, s2, ..., sn}, Sj = {s1, s2, .. , sm}, Dist (a, b) represents the Euclidean distance between any point a and b in the map data, and point #p represents any point belonging to space E containing a specific location in the map.

In addition, in order to solve the above technical problem, another embodiment of the present invention, in the surveillance camera arrangement modeling method using the surveillance camera arrangement modeling system, the surveillance camera arrangement modeling system to set the analysis range to the map data of a predetermined area A range setting step, a grid generation step of forming a grid including a plurality of unit grids each having numerical information in the analysis range by the surveillance camera batch modeling system, and having the surveillance camera batch modeling system each of the plurality of unit grids A map generation step of weighting each of the plurality of unit grids according to the numerical information and forming a map capable of area division according to the magnitude of the weight, and the surveillance camera arrangement modeling system corresponds to a specific position in the map One or more placed to It provides a surveillance cameras batch modeling method comprising the step of generating a model for generating at least one batch modeling information of the surveillance cameras.

In another embodiment of the present invention, the map data includes terrain information, coordinate information, building information, building shape information, surveillance camera installation location information, and surveillance camera installation number information of the predetermined area, and the range setting step includes The monitoring camera arrangement modeling system may be configured to set the analysis range by using terrain information, building information, and coordinate information included in the map data.

In another embodiment of the present invention, the plurality of grids are arranged in a form having an interval of 1m to 10m, the numerical information is any one of the corresponding floating population numerical data, crime rate numerical data and traffic volume numerical data, the weight The size of may be proportional to the size of the data value of the numerical information.

In another embodiment of the present invention, the grid generation step, the monitoring camera layout modeling system forms a grid including a plurality of unit grids each having numerical information in the analysis range, but the building shape information included in the analysis range And extracting a planar region, wherein the map generation step includes a plurality of unit grids formed in a region other than the planar region of the analysis range by the surveillance camera layout modeling system. The method may include weighting each of the plurality of unit grids formed in regions other than the planar region according to the numerical information of each of the plurality of unit grids formed, and forming a map capable of distinguishing regions according to the size of the weights.

In another embodiment of the present invention, the map generation step, the surveillance camera batch modeling system weights each of the plurality of unit grids according to the numerical information of each of the plurality of unit grids, the size of the weight In some embodiments, each of the plurality of unit grids may form a heat map having a different color.

In another embodiment of the present invention, the one or more batch modeling information formed through the model generation step includes the data of the number of one or more surveillance cameras are arranged to correspond to a specific position in the map, the surveillance camera batch modeling system The method may further include a model recommendation step of deriving modeling information having the smallest number data value of the surveillance camera included in each of the one or more batch modeling information.

In another embodiment of the present invention, the model generation step, the number information of one or more surveillance cameras arranged so that the surveillance camera arrangement modeling system corresponds to a specific position in the map, the position information possible to place, the effective viewing angle information, the effective The method may include generating a plurality of arrangement modeling information of one or more surveillance cameras arranged to correspond to a specific position in the map based on the viewing distance information and the effective surveillance region information.

In another embodiment of the present invention, at least one surveillance camera arranged to correspond to a specific position in the map has a horizontal viewing angle of 0 ° to 360 °, a vertical viewing angle of 90 ° to -90 °, and a preset coverage range. Among the one or more surveillance cameras arranged to correspond to a specific position in the map, the surveillance camera layout modeling system uses information on the number of surveillance cameras arranged in the Voronoi region VOR (Si) calculated by Equation 2 below. The method may further include a model recommendation step of providing specific batch modeling information among one or more batch modeling information generated through the model generation step.

[Equation 2]

,

,

Here, S, Si and Sj are a set of buildings corresponding to the specific position, and (Si∪Sj) ∈S, Si = {s1, s2, ..., sn}, Sj = {s1, s2, .. , sm}, Dist (a, b) represents the Euclidean distance between any point a and b in the map data, and point p represents any point belonging to space E containing a specific location in the map.

According to the present invention, it is possible to model the optimal arrangement of the surveillance camera additionally installed in the region based on various data of the specific region.

The effects of the present invention are not limited to the above-described effects, but should be understood to include all the effects deduced from the configuration of the invention described in the detailed description or claims of the present invention.

1 is a graph showing the number of conventional surveillance cameras installed by year.
2 and 3 are diagrams for explaining the concept of coverage used in the present specification.
4 is a diagram illustrating a viewshed analysis applicable to various embodiments of the present disclosure.
5 and 6 are diagrams for explaining the concept of coverage analysis applicable to various embodiments of the present invention.
7 is a view showing Korean and US public data relating to crime occurrence.
8 is a diagram illustrating an example of coverage analysis according to various experimental examples of the present invention.
9 is a diagram illustrating a concept of coverage analysis to which weights applicable to various embodiments of the present invention are applied.
10 is a diagram illustrating a system according to an experimental example of the present invention.
11 is a diagram illustrating the architecture of a system according to an experimental example of the present invention.
12 to 18 illustrate screens generated through procedures performed according to various experimental examples of the present invention.
19 to 27 are diagrams for explaining the Voronoi diagram analysis applicable to various embodiments of the present invention.
28 to 30 are diagrams for explaining the coverage level evaluation according to various experimental examples of the present invention.
31 illustrates the main steps for implementing a Voronoi approach applicable to various embodiments of the invention.
32 to 41 are diagrams illustrating results of a surveillance camera batch modeling test according to various experimental examples of the present invention.
42 is a diagram showing the configuration of a surveillance camera arrangement modeling system according to an embodiment of the present invention.
43 is a diagram illustrating a procedure of a surveillance camera arrangement modeling method according to another embodiment of the present invention.

Hereinafter, with reference to the accompanying drawings will be described in detail the present invention. However, the present invention may be embodied in many different forms and, therefore, is not limited to the experimental examples and examples described herein. In addition, the accompanying drawings are intended to facilitate understanding of the embodiments disclosed herein, but are not limited to the technical spirit disclosed herein by the accompanying drawings, all changes included in the spirit and scope of the present invention. It is to be understood to include water, equivalents and substitutes. And the part not related to the description in order to clearly describe the present invention in the drawings are omitted, the size, shape, shape of each component shown in the drawings may be variously modified, the same / similar parts for the entire specification Identical / similar reference numerals are used.

The suffixes "step" and "unit" for components used in the following description are given or mixed in consideration of ease of specification, and do not have distinct meanings or roles from each other. In addition, in describing the experimental example and the embodiments disclosed herein, when it is determined that the detailed description of the related known technology may obscure the gist of the embodiments disclosed herein, the detailed description thereof is omitted.

Throughout the specification, when a part is said to be "connected (connected, contacted or coupled) with another part, it is not only when it is" directly connected (connected, contacted or coupled) ", but also in between. This includes cases in which "indirectly connected (connecting, contacting or coupling)" therebetween. In addition, when a part is said to "include (or prepare)" a component, it is not to exclude other components, but to "include (or prepare)" other components, unless specifically stated otherwise. That means you can.

The terminology used herein is for the purpose of describing particular example embodiments and examples only and is not intended to be limiting of the invention. The singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. As used herein, the terms "comprise" or "have" are intended to indicate that there is a feature, number, step, action, component, part, or combination thereof described on the specification, and one or more other features. It is to be understood that the present invention does not exclude the possibility of the presence or the addition of numbers, steps, operations, components, components, or a combination thereof.

In addition, terms including ordinal numbers, such as first and second, as used herein may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as the second component, and similarly, the second component may also be referred to as the first component.

Before explaining the surveillance camera arrangement modeling system provided by an embodiment of the present invention and the surveillance camera arrangement modeling method provided by another embodiment of the present invention, an analysis applicable to various embodiments of the present invention through various experimental examples of the present invention. Describe the method and its concepts first.

2 and 3 are diagrams for explaining the concept of coverage (coverage, coverage) described herein.

The concept of coverage reflects how well the surveillance area is monitored by a set of sensors (cameras). An example is an attempt in the field of heritage, where each point in the museum's exhibition room is defined to define the required number of cameras covered by at least one observer. In museum exhibits, coverage is defined by the direct visibility between surveillance cameras (CCTVs) and targets. Another example of a coverage problem is extending the life of the sensor by reducing energy consumption. This approach attempts to identify and turn off extra sensors that are monitoring the same area when not in use.

A general definition of the coverage problem has been proposed in the field of sensor networks, such as IoT, where for a group of sensor groups deployed in a target area, every point of interest is covered by at least k sensors, where k is the number of sensors that simultaneously monitor the target. define. As shown in Fig. 2, the white circle corresponds to the area covered by one sensor, the dotted line area corresponds to the area covered by two sensors, the gray area corresponds to the area covered by three sensors, and the dashed area is not covered. Corresponds to the area.

Coverage is primarily related to the sensing range of each sensor associated with the monitored phenomenon and the obstacles present in the target area. Therefore, developing a model to calculate coverage requires considering the nature and obstacles of the application. In the case of camera network deployment, the line of sight may be considered a suitable method for modeling the range of the camera network. The gaze-based model calculates visibility based on the visual line from the observation point O (or the observer) to another point T in the target area. If the visual line intersects the obstacle before reaching T, T is hidden and not visible from O. Otherwise, T is shown at O as shown in FIG.

In the three-dimensional GIS, viewshed analysis is used when the visibility analysis process is applied to all points in the target area. This analysis is derived from the movement of the gaze across the entire surface. Viewshed analysis may be performed based on one or more viewpoints. The results of the field of view analysis consist of determining the set of points on the surface visible in the set of viewpoints.

4 is a diagram illustrating a viewshed analysis used in the present invention.

The observation point corresponding to the position of the camera is represented by 2D coordinates (x, y) and corresponding elevation (z). The spatial position of the camera can be determined using a Digital Elevation Model that represents the environment in which the camera is placed. ArcGIS uses eye-based viewshed analysis for camera elevation (SPOT), vertical offsets (OFFSETA, OFFSETB), horizontal scan angles (AZIMUTH1, AZIMUTH2), vertical scan angles (VERT1, VERT2), and scan distance (RADIUS1, RADIUS2). ) As a variable. 4 shows nine variables for the viewshed analysis described in the ArcGIS manual.

The following are descriptions of the nine variables used to control Viewshed analysis in the ArcGIS manual.

SPOT: Corresponds to the ground level of the observation point (camera).

OF1 'and' OF2 ': These variables define the vertical elevation to add to the ground elevation of the observers OF1' (OFFSETA) and subjects 'OF2' (OFFSETB).

AZ1 'and' AZ2 ': These two values allow you to define the range of horizontal angles that characterize the observer's (camera's) scan. AZ1 (AZIMUTH1) corresponds to the start angle of the scan range and AZ2 (AZIMUTH2) corresponds to the end angle of the scan range. The values of AZ1 and AZ2 can range from 0 ° to 360 °, defined by the north direction.

V1 'and V2': These two elements define the observer's vertical scan range. V1 (VERT1) defines the upper limit of the vertical angle and V2 (VERT2) defines the lower limit of the vertical angle. V1 'and' V2 'can be set from -90 ° to 90 °. R1' and 'R2': This parameter determines the coverage distance of the observer (camera). R1 (RADIUS1) corresponds to the starting distance for viewing the object.

Based on the nine variables presented above, the gaze is constructed for each point of the target area. The viewshed for the environment in which the camera will be placed is then calculated as:

Oi (xi, yi, zi) is the observation point and i is the exponent of each observation point, i = 1, 2, ..., n. z i = SPOTi + OF1i. Tj (xj, yj, zj) is the target point where j is an indicator of each target, j = 1, 2, ..., m. Hj 'is the elevation of the surface corresponding to target \ Tj.

OTij (xj-xi, yj-yi, -zj-zi) 시 is the gaze composed by the observer Oi with index i and the target Tj with index j. Gaze distance It is represented by OT ij ||.

Viewshed analysis is performed as follows. If one of the following conditions is met, the target Tj is not visible from the observer Oi.

(1) The field of view " OTij " between " Oi " and the target " Tj " is obscured by one or more obstacles.

(2) The target Tj is outside the range of distances defined by R1i and R2i.

(| OTij || <R1i) or (|| OTij ||> R2i).

(3) The target Tj is outside the horizontal angle range defined by AZ1i and AZ2i.

(Arctan ((yj-yi) / (xj-xi)) <AZ1i) and (Arctan (yj-yi) / (xj-xi))> AZ2i).

(4) The target Tj is outside the vertical angle range defined by V1i and V2i.

(Arcsin ((zjzi) / || OTij ||) <V1i) and (Arcsin ((zjzi) / || OTij ||)> V2i).

(5) If all of the above conditions (1) to (4) are not met, the target Tj can be viewed from the observer Oi.

Although the principle of gaze analysis is based on a geometric approach, the implementation of Viewshed in GIS software is usually performed in a raster form on the Digital Elevation Model. This is because of the amount of computation necessary to construct a line of sight for each vertex of the vector data representing the spatial environment. However, it is also possible to use a triangular observation network (TIN) as a digital model to generate view area analysis based on the field of view. The construction of the line of sight in a 3D GIS is based on discrete calculations because it is necessary to specify the three-dimensional objects that will make up these lines. As with vector data, the number of vertices of the 3D object used in the viewshed analysis should be reasonable considering the amount of computation when calculating all the views.

According to the present invention, it is possible to design and design a grid coverage analysis method in consideration of the calculation speed of the system. For coverage analysis and comparison of CCTV arrangements, it is realistic to approach AOI (Area Of Interest) area rather than CCTV POI (Point Of Interest) location, which divides the surface of the target area into grids at regular intervals and analyzes the coverage. The scheme is a method that can be effectively applied to derive this AOI domain.

In addition, the actual CCTV layout should consider the social environment of the place as well as the physical environment. In other words, it is required to consider the location of CCTV that is practically needed in the field based on the crime rate and the distribution of the floating population. To this end, various experimental examples of the present invention set coverage of places in a target area based on public data such as crime rate or floating population, and apply coverage analysis.

Coverage analysis allows the piles to be distributed on the ground at regular intervals and to determine the performance of the overall layout. There are many ways to judge performance. First of all, if there is no major visible space in the graph interpretation, if there is a wider graph that fully covers the previous graph, the layout is better. This is also associated with the mean value, where a high standard deviation may be preferred if a major observation area is established. This is the case when coverage is concentrated in the main viewable area.

FIG. 5 is a conceptual diagram of distributing grids at regular intervals in a target area and analyzing coverage. FIG. For coverage analysis, it is desirable to analyze the building area from the surface area. To this end, in various experimental examples of the present invention, surface data was taken from national spatial information, and building area information and floor number attribute information were obtained from integrated building information data, and surface modeling was performed in a three-dimensional GIS.

For coverage analysis, it is desirable to set the total coverage area by extracting the grid except the building area from the grid uniformly distributed in the region of interest. Therefore, the phase must be identified to see if the grid is located outside the polygon, which is the building area. To this end, a crossing number (cn) method was used, and a figure for explaining this is illustrated in FIG. 6.

This method counts the number of times the ray starting at point P crosses the edge of the polygon boundary that separates the inside from the outside. If this number is even, the point is outside. When the number of crossings is odd, the point is inside. Each time a ray crosses a polygon edge, its internal parity changes. After all, any ray should end outside and outside the bounded polygon. So if the point is inside, the order of intersection must be in> out> ...> in> out and odd. Likewise, if the point is outside, there are even intersections in the sequence. out> in> ...> in> out.

When implementing the cn technique, you must ensure that only intersections that change in-out parity are calculated. In particular, in special cases where light rays pass through vertices, they must be handled correctly. In addition, you must determine whether the point on the boundary of the polygon is internal or external. The standard rule says that the point at the left or bottom edge is inside and the point at the right or top edge is outside. This allows two separate polygons to share a common boundary segment, with one point of that segment on one polygon or another, but not both at the same time. This avoids many problems that can occur especially in computer graphic displays.

The linear crossing number algorithm selects horizontal rays that extend to the right of P and are parallel to the positive x axis. This particular ray makes it easy to calculate the intersection with the polygon edges. It is easier to determine when such an intersection is impossible. To compute the total intersection point cn, the algorithm iterates over all the edges of the polygon, tests each intersection, and increments cn whenever there is an intersection. In addition, cross-testing has to deal with special cases and edges.

The grids of the coverage analysis targets are equally weighted in the physical environment.However, the layout of CCTV should be considered in consideration of the social environment of the place as well as the visibility analysis. A coverage analysis method is required to verify the effectiveness of CCTV deployment.

In general, in the case of crime prevention CCTVs within the analysis area, the variables that can be set to classify the risk by location are as follows.

Variables that can be derived from drawing analysis are the location of hazardous facilities, the distribution of places where people are more likely to stay considering the floating population data, road class (width), and crime rate. These individual variables are reclassified into four categories: crime factor (criminal incidence rate), traffic factor (floating population), hazard factor (entertainment facilities, etc.), road specification (road width), and then “monitored” for the risk level classification by place. As the probability (visibility) increases, crime will decrease. ”Among the road factors, the wider the width of the road, the more traffic there is and the more likely it is to be monitored. Since the opinions to the contrary and the contrary opinions conflict, the above matters were not considered in various experimental examples of the present invention.

The higher the incidence rate of crime, the higher the risk of crime on the street. In overseas, the spatial data of such crime incidence are disclosed, but in Korea, it is still impossible to disclose it. In this regard, crime occurrence-related Korea (left of FIG. 7) and US public data (right of FIG. 7) are shown in FIG. 7.

In the coverage analysis of CCTV, the coverage of the camera is subdivided and analyzed according to the recognition degree according to the distance between the camera and the subject as well as the effective distance of the camera. According to an exemplary embodiment of the present invention, four levels of camera coverage may be classified according to distance, and based on weights of identification (Poor), identification (good), recognition, and detection according to the degree of recognition and interpretation of people and objects from a close distance. In this regard, the CCTV deployment can be planned and implemented through camera coverage analysis, and an example of coverage analysis of CCTV is shown in FIG. 8.

Including the harmful factors in the spatial analysis does not generalize that the actual number of crimes increases as the number of hazardous facilities in the area increases, but indirectly decreases the probability of crimes as the number of potential criminals decreases as they move away from the hazards. It is based on logic and can be discussed. Therefore, in various experimental examples of the present invention, the weight of the space was applied by applying a traffic factor based on the floating population data. In the past, the floating population survey data was based on field surveys, but in the smart city environment, the trend is being applied based on mobile call data of users, and the disclosure of floating population data for various business analysis is centered on local governments and SME support organizations. It is gradually being made public. The conceptual diagram of the coverage weight to which the floating population data is applied is shown in FIG. 9.

Based on these concepts, the description of the simulation screen of the system to which the CCTV coverage analysis method is applied on the 3D GIS according to an experimental example of the present invention is as follows, and the related figure is illustrated in FIG. 10.

Referring to FIG. 10, according to an experimental example of the present invention, a grid having a predetermined interval is arranged in a space excluding a building surface in an analysis target area, and weights are applied for each location based on floating population data. The dark blue grid area shown in FIG. 10 is a high floating population area and the red area is a coverage area of CCTV. The coverage of CCTV consisted of three stages, establishing a close distance for identification, a medium distance for general recognition, and a long distance for recognition of hard or motion. The coverage of the camera was implemented except that the view face was obscured by the view surface and the building by Viewshed. The accuracy of coverage analysis can be realized as the grid gap gets smaller, but the minimum distance is set to 1m considering the calculation speed of the system for the area of the earth.

Coverage analysis is defined as a comparative analysis of the efficiency of coverage for a plurality of plans for camera position, and can be applied to surveillance camera systems. In this case, a batch support system is available which compares the proposals and improves the efficiency of the coverage while changing the position of the camera by user input in real time.

The developed system for CCTV coverage analysis according to the experimental example of the present invention used a three-stage architecture similar to the general standard web GIS, the structure shown in FIG.

Referring to Figure 11, the remote application server is responsible for storing a large amount of GIS data. Databases that interact with clients through a web server are also stored on the remote application server. The database stores the generated tables, and the interface required for communication between the database and the web server is provided by code running on the web server. The code needed to perform the processing to perform the coverage analysis is executed on the client to improve the speed to enable real-time coverage analysis according to the camera position movement. This allows planners to intuitively see changes in coverage while changing the location of CCTV installations. The web server handles requests from clients. All HTTP requests are managed by a web server, which hosts the cesium JavaScript library along with other web pages belonging to the web site. The required JavaScript file is sent to the client on every request. Apache Tomcat also hosts Geoserver, a Java-based software server that allows users to view and edit geospatial data. Geoserver provides various web services as requested by clients such as WMS, WMTS or TMS. In this case, it is used to provide satellite images and maps using tiled WMS draped in the cesium virtual world on the client side. It is also used to provide terrain data using WMS. On the client side, users only need to use a web browser and an Internet connection to access and analyze all the data. The architecture of Web GIS is cross-platform, giving users the freedom to use any operating system and web browser that can run WebGL. This ensures versatility of the system.

The system developed according to one experimental example of the present invention uses EPSG: 4326, WGS 84 coordinate system. For this reason, when the shape information of the building stored in the Ministry of Land, Building Information is called by Geojson, an error of several hundred meters may occur due to the difference in the coordinate system, and the related figure is illustrated in FIG. 12.

In order to apply the building data used in Korea to the EPSG: 4326, WGS 84 coordinate system, a correction reference value for the position error between the map and the building is required for each region (longitude). In order to solve the positional error between the map and the building when loading the building shape data on the map and finally create the geojson file and upload it on the cesium map, the data of the test bed, Daejeon, Table 1 and Figure 13 apply.

Table 1

Set up a regional range for coverage analysis based on the terrain and building data of the target area. In various experimental examples of the present invention, a user may directly set an analysis area boundary in the form of a polygon based on a map. Once the boundary of the target area is established, a grid for coverage analysis is created. The grid was created and calculated only within the interface. At this time, the interval of the grid can be arbitrarily set by the user so that the environment can be set in consideration of the calculation speed compared to the analysis area. The proper grid spacing for deriving efficiency should be derived through regression analysis on the basis of various data. However, in various experimental examples of the present invention, considering the limitations of such data, it is generally smaller than the facade length of one building. Can be set in 10m increments. As the spacing of grids gets smaller, the number of grids to be analyzed increases exponentially, which can lead to a decrease in computational speed.However, in the development system, performance has been improved to secure real-time coverage analysis even at small intervals in the same area. , The main menu of the system according to an experimental example of the present invention is shown in FIG.

The setting of the area to be analyzed is the basic condition for coverage analysis. The development system according to an experimental example of the present invention allows a user to directly input an analysis area on a map. It automatically creates a grid within the target area entered as a polyline to support the generality of the target area and to freely set the target area. In the case of terrain, mesh data in grid unit is imported, but the area to be analyzed exists in the form of a point in the degree of freedom to correspond to the setting of various areas.

After the boundary of the analysis target area is set and the grid is formed at regular intervals inside, the area of the coverage target except for the building part is set as shown in FIG. 15. Accordingly, the building group plane shape information of the corresponding area is extracted from the building integration information data, and a grid that is a calculation target is formed only in an area other than the plane.

The location of surveillance cameras according to CCTV coverage analysis in the target area will be different depending on the purpose of CCTV. To this end, the effectiveness of coverage analysis should be secured by applying weights appropriate to the purpose of the place as well as the coverage of the simple physical environment. In the case of crime-proof CCTV, weights are given for each physical location based on crime rate and floating population data. In the case of traffic enforcement and traffic information collection, weights can be assigned based on road traffic information and traffic accident occurrence data for each place. In addition, it is possible to set weights for analysis areas based on various public data. Various experimental examples of the present invention are weighted based on crime rate and floating population survey data for each location based on crime prevention CCTV and visualized it on a map through a heat map. Heat maps are a combination of heat, which means heat, and map, which means map. The characteristic of heat map is that various information that can be expressed in color is output as visual graphic in the form of heat distribution on a certain image. The floating population data has been created based on the number of people observed at each survey point, but recently, big data techniques are being developed to investigate and analyze based on location information and time information provided by a user's smartphone. Various experimental examples of the present invention were based on floating population open data provided by public and private sites. Table 2 below is an example of floating population disclosure data.

TABLE 2

Collecting data that fits the surveyed purpose by location coordinates, assigning weights for coverage analysis for each location, and visually expressing it on the system intuitively guides important locations to users during the batch editing of CCTV. In addition, various experimental examples of the present invention provide functions for basic spatial analysis of distance measurement, area calculation, and slope calculation function in three-dimensional space, thereby improving convenience and use of the system user as shown in FIG. 16.

Publicly installed cameras provide basic information on the installation location, purpose of installation, and camera specifications through the opening of public data. The purpose of installation is mainly traffic control, traffic information collection, crime prevention, facility management, disaster, and multi-purpose cameras are installed for multiple purposes. The installation location provides the latitude and longitude information so that the location can be mapped on the GIS. However, since only representative coordinates are provided for a plurality of cameras, in order to map the positions of individual cameras, an actual installation position must be input through field survey or equipment manager check. The direction of the traffic camera is based on the right and left direction of most cameras other than the traffic, 360 degree coverage is possible. The following is an example of CCTV public data provided by the city and districts. Although the installation height of CCTV is not provided, it is generally installed at the height of one floor of a building, and based on this, a system was developed according to one experimental example of the present invention.

Table 3-Examples of CCTV-related public data

In order to compare the optimal solution for the additional installation of new CCTV based on the existing CCTV, the system was developed so that the user can input the camera position on the system and intuitively know the coverage change accordingly. The coverage analysis has a general coverage analysis that evaluates only in a physical environment without weights and a weighted coverage analysis function that applies weights according to the use. The analysis results are displayed in a heat map format on a 3D screen as shown in FIG. 17. To intuitively check the coverage change according to the camera position change.

Various experimental examples of the present invention can provide a system for deriving an optimal solution among a plurality of proposed proposals by storing a plurality of CCTV arrangements of one project and allowing comparative analysis of coverage. To this end, the number of CCTVs, the general coverage comparison analysis, and the weighted coverage comparison analysis result can be visually seen as shown in FIG. 18 to support decision making for optimal camera placement.

Since the comparative analysis of coverage should be done within the same analysis target boundary, there is one environment setting value for each project, and the combination of various camera layouts is stored as a list based on one boundary surface and grid interval setting value.

Various experimental examples of the present invention provide a system for implementing a coverage analysis method for selecting an optimal camera position through a comparative analysis of a plan and a method for implementing a CCTV layout using a Voronoi diagram as a method of performing a recommendation function for a camera layout in a system. Presents additionally.

The concept of the Voronoi diagram has been used in several studies to deploy sensor networks and surveillance cameras. The diagram of the Voronoi site set is the result of dividing the space into regions. Each region corresponding to site "s" consists of all points closer to "s" than other sites. The definition of the diagram is

In the space "E" ("R2" or "R3"), a set of points called "sites" or "constructors" represents "S = {s1, s2, ..., sn"}. Also consider sisi and sj as two sites belonging to S. The dominance of the si for sj (DOM) can be formulated as follows, so that a subset of the space E s is at least close to s j and at least si.

Where Dist '(a, b) is the Euclidean distance between point a and point b and point' p 'is a point belonging to space E. As before, the Voronoi region generated by the site si is the result of the intersection of all of Si's superiority over the remaining sites in S.

Thus, the Voronoi diagram ("DIAVOR") generated from the site set S = {s1, s2, ..., sn \}} is a collection of all Voronoi areas that divide the entire space.

19 shows a Voronoi diagram occurring at a set of sites. In the Voronoi diagram, each point on the edge is equidistant from the two sites whose areas share the edge. Each vertex is also at the same distance from the three closest sites that produce the Voronoi region. In addition, the principles of the Voronoi diagram can be extended to various types of sites such as lines, polygons, and weighted features (points, lines, polygons) instead of common points. FIG. 20 shows a Voronoi diagram generated from weighted lines and polygons (B, D) with unweighted lines and shapes like polygons (A, C) and (A, C). The weights used correspond to the line length of B and the polygonal area of D. The gray scale color corresponds to the distance (or weight for B and D) between each point and the nearest generator feature.

The Voronoi approach is for CCTV deployments that belong to the Vista-based method family. In these environments, the most important factors to consider in the deployment of the camera network are paths, building facades and doorways, and the number of cameras must be optimized while achieving the highest possible coverage range. In general, in a surveillance situation, the roof of a building is not as important as the facade or entrance. Therefore, the main purpose of this approach is to reach almost 100% coverage, focusing on the path, building facades and entrances, and the minimum number of cameras possible. In addition, deploying a surveillance camera network requires a digital elevation model (DEM) that represents the ground elevation of the target area and 3D vector data that models buildings and their characteristics, as well as objects such as trees and street furniture that are useful for calculating visibility. Do. The data needed to assess coverage comes primarily from topographic surveys of sites of interest.

The Voronoi approach uses the camera's position as a generator point that serves to segment the monitored space, with the goal of finding the optimal placement of the surveillance camera network based on the Voronoi diagram generated in the space of the building located in the environment. Conventional surveillance camera placement modeling methods differ in the algorithms used to modify the camera's position or add new cameras to improve coverage, and do not provide an effective way of defining the initial position of the camera as close as possible to the optimal position. The Voronoi approach provides a way to overcome this weakness. The principle first uses the building as the generator object of the Voronoi diagram and ensures that a CCTV camera is placed at the edge of Voronoi. This type of deployment should minimize the number of cameras for maximum coverage, provide better visibility of the building's facades and entrances, and maximize roads.

Points belonging to the Voronoi edge are known to have properties that are equidistant to the constructor object that shares this edge. Therefore, when the camera is placed on the edge of Voronoi, it is as close as possible to the two corresponding buildings (creator objects) at the same time. Also, when the camera is placed much closer to the building, it is automatically away from the building it is facing. This can increase the number of cameras needed to cover the entire environment. Figure 21 shows the advantage of reducing the number of cameras by placing cameras at the Voronoi edge. The total number of cameras required to cover the entire area depends on the parameters of the camera to be installed.

For better observability, objects in the environment must occupy as much area as possible in the image or video. To achieve this goal, we need to minimize the foreshortening effect. This effect affects the amount of information that can be extracted from the observations. Shortening effects occur when the angle between the camera's point of view and the subject becomes smaller. This reduces the projection of the object in the camera's image plan and makes the monitored object appear shorter in the image. 22 shows the effect of the shortening effect.

When monitoring the built-in environment and considering the minimum number of cameras possible, placing the camera in the building closest to the camera facing face-to-face increases the shortening effect when the camera is connected to the furthest building. The shortening effect occurs because the distance from the nearest building is farther away. This is especially true if the angle between the camera's field of view and the front of the furthest building to be monitored is small. Therefore, even within the Voronoi region, placing the camera at the Voronoi edge optimizes the shortening effect on both sides, allowing better observation of the building's exterior walls and entrances.

In general, roads are constructed to ensure access to buildings in urban environments. Thus, the road is almost at the same distance from the building. Thus, the Voronoi boundary created at the location of the building is very similar to the road structure constructed in the urban environment. Therefore, placing cameras at the corners of Voronoi allows better observation and application of the roads while minimizing the number of cameras. 23 is a conceptual diagram of a camera arrangement of the Voronoi edge to better observe and apply the road.

The next step is to find the best position for each camera to get the maximum possible range while minimizing the number of cameras. For this purpose, the best location of the Voronoi edge is defined by three constraints: (1) camera parameters, (2) location of the building, and (3) minimizing the shortening effect. In the various experimental examples of the present invention, two assumptions are considered. The first is that the camera should be mounted on the shore. The second assumption is that all cameras have the following parameters.

-OFFSETB = 0 ms (no artificial object in the field).

AZIMUTH1 = 0 ° (starting horizontal angle).

AZIMUTH2 = 360 ° (end horizontal angle of the scan range).

VERT1 = 90 ° (upper limit of vertical angle).

VERT2 = -90 ° (lower limit of vertical angle).

RADIUS1 = 0 (starting distance visible).

RADIUS2 = constant (distance beyond which this point is not clearly visible. This parameter represents the camera coverage range).

There are two parameters: SPOT and OFFSETA cameras, which can take various values.

SPOT corresponds to the altitude above the ground.

OFFSETA is related to the height of the building to be monitored by each camera.

To create a Voronoi diagram, the outside plane of the building is used as the constructor object. This shape can be generated through the existing plane of the building integration information and the existing model or terrain survey of the three-dimensional building. Once the Voronoi diagram is generated, each edge is split at one of two Voronoi vertices based on the camera range as follows:

1. If the length of the Voronoi edge is less than the camera's range, this edge will not be split.

2. If the length of the Voronoi edge is larger than the camera's range, this edge will be divided into the same segments. The number of these new segments is calculated according to the following procedure.

-Ns: The number of new segments created from the Voronoi Edge split.

-Ceil: A mathematical function that calculates the integer part of a number (e.g. Ceil (1.45) = 1).

-Mod: is a mathematical function that calculates the remainder of the division

-LVE: original length of the Voronoi edge

-R Cam: range of camera

After that, the length of each segment generated is as follows.

Where LSpVE is the length of the new segment created by dividing the original Voronoi edge.

After dividing the Voronoi edge into equal subdivisions, the camera is located at the midpoint of each new segment. Placing the camera in this way increases the building's viewability. Indeed, when the camera is placed on a Voronoi vertex, narrow angles can occur to monitor the building (and building entrance), one of the main causes of the shortened problem. 24 is a view illustrating a camera disposed in the middle of a Voronoi section in relation to a surrounding building.

Dividing the Voronoi edge by camera range comes from the idea that each camera should be visible from two adjacent cameras (front and back). Thus, the distance between two consecutive cameras is always less than the camera range. Thus, if a camera fails, a blind spot is not created along the Voronoi edge, and the camera can be used until the failed camera is repaired. 25 shows a service area due to a camera failure.

To ensure efficient surveillance, cameras should be installed at a distance from which the front and entrance of the building can be fully observed. Therefore, the maximum permissible distance between the building and the entrance must be determined to ensure maximum coverage of the front and entrance. FIG. 26 shows the method used to calculate this distance according to the camera range and the distance between two adjacent cameras (this distance should not exceed the camera range).

Where R is the camera range and Dmax is the maximum allowable distance between the building and the Voronoi boundary. The distance Dmax is as follows.

D max in the above equation is an approximation based on the 2D assumption. Calculations based on 3D assumptions give nearly the same values as a camera about 10 meters high.

If one of the buildings is far from the Voronoi edge at a distance greater than the aforementioned maximum distance, you should duplicate this Voronoi edge. The camera placed at the edge of this replica must meet two conditions: The distance between the duplicated Voronoi edge and the building must not be greater than the distance Dmax mentioned in the equation above. Cameras placed on overlapping Voronoi edges should ensure full coverage of the Voronoi edges. If the building is far from the edge of Voronoi at a given distance, the opposite building will be far from the edge of Voronoi at about the same distance. 27 shows the overlap between the Voronoi edge and the number of cameras.

The location of the camera network at this stage is based on the Voronoi diagram generated from the location of the building. The camera's position is used to evaluate and calculate the range through a hybrid approach that combines raster and 3D vector analysis of visibility.

After defining the position of the camera, the level of coverage is evaluated based on the viewshed calculation. The proposed approach combines two methods for raster-based and three-dimensional vector-based view visual calculations.

The raster-based viewshed calculation method allows you to evaluate a range of interests in a raster format based on the Digital Surface Model (DSM). DSM is the result of combining ground elevation and building height. To create such a DSM, a Raster-type Digital Elevation Model (DEM) representing the ground elevation must be created. Raster DEMs are typically extracted from the Triangular Irregular Network (TIN), which represents the elevation of the ground. The location and boundaries of the building are then generated from existing 3D models or terrain surveys. The building boundaries in vector format are then converted to raster format using the same resolution of the original DEM. The resulting pixel value of the raster corresponds to the roof height of each building. Next, a raster file representing the roof height of the building is combined with a raster DEM corresponding to the ground level. Next, to perform this combination, for each pixel in the resulting DSM, the pixel value is taken from the raster DEM if the corresponding value in the building's roof raster file is null. Otherwise, the pixel value is taken from that pixel in the building's roof raster file.

Combining elevation information from raster DEMs and 3D building models provides a rasterized digital surface model of the area of interest. This DSM is essential for evaluating the range of camera networks through the Bornoi approach.

After creating the DSM, the height value of each camera corresponds to OFFSETA. This offset value is chosen according to the heights of the two buildings, which are considered the basis for generating the edge of the Voronoi where the camera is placed. The camera height is mainly related to the camera poles in which the camera is placed. Two situations can be expected: The first case is a small building around the camera pole. In this situation, the camera height value corresponds to the tallest building. The advantage of this standard is that it offers the possibility of monitoring the roof in addition to the entrance and facade. Otherwise, if it is not possible or unprofitable to construct a column that is the same height as a tall building, you should choose the standard height of the column on which to place the camera. However, the roof of the building may not be visible from the camera. The roof of the building does not have a significant effect on the results according to the various experimental examples of the present invention since it is of no significance for surveillance.

The basic parameters of the camera to be placed are integrated into the attribute table of the point feature class that represents the two-dimensional placement of the camera network. 28 shows an attribute table of a camera class. In this example, all cameras have a constant range of 100m. The parameters OFFSETB, AZIMUTH1, AZIMUTH2, VERT1, VERT2, RADIUS1, and RADIUS2 are also constant. However, “SPOT” and “OFFSETA” may differ depending on the position of each camera.

Viewshed is computed when a feature class is created that represents a raster DSM that combines ground and building heights and a camera with location and attributes. The methodology presented above is then applied to evaluate each pixel regardless of whether it is visible from one or more cameras in the deployed network. The attribute table of the camera class is used to provide the camera parameters needed to calculate the viewshed.

According to the 3D vector-based analysis of visibility, the 3D vector model allows you to perform visibility analysis on elements that cannot be evaluated based on raster DSM. Raster DSMs only allow viewshed calculations of objects and features on surfaces. The roof is therefore the only component of a building that can be in a raster DSM. Other vertical components of the building, such as facades, doors, and windows, cannot be modeled in raster DSMs. For this reason, 3D vector-based analysis of visibility must complete the coverage assessment of the camera network. This method of visibility analysis consists of constructing a line of sight between each camera and the object of interest to be monitored at this location. To do this, you need to establish a many-to-many relationship between the camera class and the class representing the object to monitor. The camera class contains important information about the camera's properties and the three-dimensional position needed to construct the spatial layout of the deployed network. The Entrance class represents the main entrance to the building (doors and windows). This many-to-many relationship means that a camera can monitor zero, one, or multiple doors, and the doors must be monitored by one or more cameras. Without this relationship, the software creates a line of sight between each camera and every vertex of the 3D model, resulting in time-consuming and difficult-to-interpret results. Figure 29 shows the UML notation for the many-to-many relationship between the Camera and Entrance classes.

The constructed gaze represents a straight line connecting the observation point (camera) and the target feature. If the camera and the target feature are sampling or polygonal features by applying the sampling distance, multiple lines can be constructed. Three-dimensional vector-based analysis of visibility consists of combining a 3D model representing an object in the monitored environment with a line of sight. The goal of coverage analysis is to determine if these objects intersect the line of sight between the camera and the object. As shown in FIG. 30, three cases may occur. The first is that if there is more than one object in the environment that intersects all the gaze between the camera and the object, the object is not completely visible in the camera. Second, if the environment contains one or more objects that intersect some of the eyes, rather than all of the eyes between the camera and the object, the object is partially visible in the camera. Finally, if no object intersects the field of view between the camera and the target, the target is completely visible from the camera.

31 summarizes the main steps in the implementation of the Bornoi approach for deploying surveillance camera networks in accordance with various experimental examples of the present invention.

Using the developed system, we analyzed the public data of the new city of Doan, a test bed, and analyzed the optimal installation area for the installation of a new CCTV based on the currently installed CCTV. Specifically, the area to be analyzed is the northern part of Doan New Town and the surrounding area that span Guam-dong, Bongmyeong-dong, and Sangdong-dong of Yuseong-gu, Daejeon. The information of CCTV installed in three legal buildings is released by three institutions of Daejeon City Hall, Daejeon City Facility Management Corporation, and Daejeon Yuseong-gu Office. Among them, Daejeon City Hall is for traffic information collection and traffic enforcement. Since only the camera information is provided for management, it referred to the data of Yuseong-gu Office, which installed cameras for the purpose of crime prevention, child protection and facility management and released related information. The published data is as of 2015, and 46 CCTVs in Guam-dong, 21 places in Bongmyeong-dong and 11 in 11 places in Daegu-dong are installed. Among them, coverage was analyzed based on the added CCTV through field survey and CCTV located in the analysis area, and the optimal area of newly established CCTV (U-pole) was derived.

The area to be analyzed is the northern part of Doan New Town, which is the subject of the 3rd-century Upole study. The grid is formed on the screen by inputting the analysis target area directly on the map and setting the interval of the analysis grid in the system. In the case of the target area, the grid is arranged on the plane for convenience as it is a new town area where the elevation of the terrain data is not large. The development system has a profile function for terrain data, so it is possible to generate grids along the topographical surface, but in this case, it may be difficult to intuitively cover coverage when analyzing the area of the district unit due to the increase of the calculation amount. However, in most urban areas, which are the target areas of analysis, the coverage analysis is possible only by the grid layout on the plane, so the grid is arranged on the plane even when the test bed is applied using the development system. 32 is a grid screen generated inside the boundary line and the boundary line of the analysis target region.

The grid for coverage analysis should be calculated without the grid located in the plane of the building. To this end, the building information of the target area is imported by using the GIS integrated building information search service provided by the National Spatial Information Portal. Including building figures and property values using spatial-based building integration information built by integrating building spatial information of continuous numerical maps (Summer Topographic Map 2.0 building layer) and building accounting property information of building administration system (Seum). Feature sets were inquired and displayed on 3D GIS using spatial data and building height information, as shown in FIG. 33.

When the final analysis target grid is set except the grid included in the building using the spatial query, the step for coverage analysis is started based on this. When the boundary and spacing of the grid and the change of the building data information change, the analysis setting conditions change, so the coverage analysis should be carried out with a new project, and the result screen is shown in FIG. 34.

Next, place weights appropriate for the purpose of analysis are applied to the area of analysis indicated by the grid. In various experimental examples of the present invention, weights were applied using floating population data that are publicly and privately disclosed and provided. The floating population is the average value of daily data for 10 days at any time point, and the dataset is shown in Table 4 below.

Table 4

Heatmaps are generated based on the maximum and minimum values to weight the floating population dataset. In the figure shown in FIG. 35, the area marked with red is the high floating population and the area marked with blue is low. The coverage analysis should be conducted considering the physical and social environments, so that the weight is applied to each grid based on the generated heat map. In addition to the floating population, weighting methods considering crime rate data, road width, or road phase relationship may be considered, and weighting factors considering institutionally important areas such as child protection zones may be considered. Only floating populations were able to collect information in the bed. It can be seen that the floating population in the general commercial area around the test bed and the neighboring commercial area in the test bed is high.

The weight is applied to each grid based on the heat map generated using the floating population data. The screens on the right of the figure shown in FIG. 36 are processes in which weights are applied to the grid. Coverage efficiency in the physical environment is based on Cov (Coverage ratio) = coverage area / total area (%), and if weighted, Wcov (Weighted Coverage ratio) = weighted coverage area / weighted area (%) do. Although the maximum value is 100%, the efficiency is different depending on the purpose of the installation even if CCTV is installed in the same location according to the weight position.

The screen shown in FIG. 37 shows a weighted grid applied to the analysis target area in the test bed. The dark grid is a high weighted area considering the floating population.

Next, install the existing CCTV in the area and establish a plurality of installation alternatives while setting the location of the new CCTV. As described above, the existing CCTV information was used by public CCTV data installed and released by the Yuseong-gu office in Daejeon. Table 5 below is an example of publicly available data and shows the purpose of installation, the latitude and longitude of the installation location, information on shooting direction, and the number of camera pixels. The purpose of installation is related to the weight of the grid, and most of the security cameras, child protection and facility management cameras are used. For shooting information, most cameras, except for traffic enforcement cameras, support 360-degrees, so 360-degree coverage analysis is performed. The number of camera pixels is related to the coverage distance of the camera, and the lower the number of pixels, the lower the coverage distance should be set. The coverage distance of the general camera is set to 100m and the recognizable distance is set to 50m, and the coverage analysis is applied by applying the weight to each distance of the camera even when the coverage analysis is performed.

Table 5

The screen shown in FIG. 38 is a screen map of locations of CCTVs in an analysis target area currently installed on a weighted grid layer. The red grid is the area where the camera is installed.

There are nine public CCTVs located in the test bed analysis area, and most of them are adjacent to the floating population. For new CCTVs, planners will preferentially place CCTVs in grid areas that are intuitively weighted while looking at grid density in the system. On the weighted grid, 17 candidate regions are expected, and users can enter multiple plans for deploying CCTVs in these regions. The part indicated in red on the screen shown in FIG. 39 is a location input by the user as the location of the new CCTV.

As the number of new CCTV installations increases, the number of installation alternatives increases, so user input based on weight mapping is easy to use when the number of cameras is small, but in many cases, the system needs to suggest recommendations for the convenience of users. Do. To this end, in various experimental examples of the present invention, the Voronoi method was conceived as a method of proposal plan recommended by the system. However, the system design is designed to derive the optimal installation area through the final coverage analysis by comparing both user input and system recommendation. Voronoi should be understood as a methodology for suggesting alternatives rather than optimal ones.

The final coverage analysis should compare the high and low coverage areas, as well as the efficiency issues of ensuring maximum area with minimal cameras. Therefore, comparing the coverage area per unit by dividing the coverage area by the number of installed cameras can also be a criterion to derive the optimal plan among various plans. This is summarized as follows, and the related figure is illustrated in FIG. 40.

CCTV Num = Number of CCTV Installed

Coverage ratio (Cov) = coverage area / total area (%)

Weighted coverage ratio (Wcov) = weighted coverage area / weighted total area (%)

CCTV installation efficiency (effective coverage per unit) = Wcov / CCTV Num

The test bed simulation is to derive the optimal layout plan for two newly installed CCTVs. Therefore, the significance of the number of installations is low, so the coverage of the plans is compared and analyzed only by the weighted coverage area. The screen shown in FIG. 41 is an example of coverage analysis for a plurality of arrangements. Finally, it was found that the layout plan in the block area of CGV located on the north side of the new urban design had high coverage.

As described above, in various experimental examples of the present invention, a system for performing a coverage analysis function for CCTV optimal deployment implemented in a Web GIS environment was implemented as a basis. In the case of VMS, the same coverage analysis function is used. Using the developed system, two optimal places for placement of Upole were derived by analyzing alternatives for optimal CCTV placement (two places in Yupole) based on the floating population data in the new urban area of Doan-si, Daejeon. The developed system was implemented using the topographical information of VWORLD and the shape and attribute information of the building information of the Ministry of Land, Infrastructure, and Transport. In the developed system, not only the coverage analysis in the physical environment but also the weighting of physical places suitable for the purpose of CCTV based on the floating population data, the effectiveness of the coverage analysis was improved and collected for various purposes such as crime rate data and traffic data. The public information available will be used to assess the optimal deployment plan for the purpose of CCTV. In addition, in the simulation according to various experimental examples of the present invention, constants for weighting were constantly set on the basis of the maximum value and the minimum value for the simulation of two additionally constructed CCTVs, but the increase in the number of CCTVs and the density of installation places in the future Considering this, we develop a system that is highly applicable to the field by regression analysis of setting values for constants for weighting through on-site evaluation, feedback, and input and output of various similar case values. It is necessary to increase the field applicability of the system.

In addition, various experimental examples of the present invention proposed a principle of a camera surveillance network deployment method using a Bornoi diagram. This approach will provide system users with a recommendation for placement, aiming to provide optimal deployment of the monitoring and surveillance network while optimizing the number of cameras to be deployed, providing user convenience for comparison analysis of the proposal. The idea of this approach is to create a Voronoi diagram from the outline of the building and use the Voronoi edges to place surveillance cameras. Then, to assess the quality of the network coverage, a raster-based viewshed calculation using a raster digital surface model of the analyte area and visibility dependent on a three-dimensional model including the main entrance of the building to be monitored The combination of 3D vector-based analysis and raster and vector visibility analysis will enhance the effectiveness of evaluating the expected coverage when deploying surveillance camera networks. In addition to optimizing the number of cameras deployed, this allows the selection of camera positions that provide full monitoring of the main entrance of the building.

The CCTV coverage analysis approach according to various experimental examples of the present invention is useful for monitoring other elements of the urban landscape such as traffic movement and pedestrian monitoring. In particular, the Bornoi approach is a suitable method for deploying cameras to monitor areas of high security interest. This can also be very effective for the placement of mobile cameras that can be mounted on a UAV. The configuration of these networks can vary depending on the dynamic nature of the environment, the number of UAVs available, and the importance of a particular target.

In the improvement of CCTV / VMS optimal installation support system function according to various experimental examples of the present invention, the system speed is improved by the system development using the point grid to improve the coverage analysis function and the building data modeling using the spatial information of the Ministry of Land, Infrastructure and Transport The system can be used. For the application of big data, we developed a weighted support function using floating population data, and developed a program to run the support system based on WebGIS. Accordingly, by developing programs applicable in areas other than testbeds, we have implemented a framework that can further develop and spread services in various fields in the future.

In addition, various experimental examples of the present invention proposed an optimal arrangement of new CCTV based on the currently installed CCTV. To this end, the location of new CCTVs was compared and analyzed by simulation using the system using integrated building information, public CCTV public information, and floating population data.

Based on the analysis method applicable to the various embodiments of the present invention described above, the concept thereof, and various experimental examples of the present invention, hereinafter, a surveillance camera arrangement modeling system according to an embodiment of the present invention and other aspects of the present invention The surveillance camera arrangement modeling method according to the embodiment will be described. However, descriptions overlapping with the contents described with reference to FIGS. 2 to 41 will be omitted.

42 is a diagram illustrating a configuration of a surveillance camera arrangement modeling system (hereinafter, referred to as a "surveillance camera arrangement modeling system 10") according to an embodiment of the present invention.

The surveillance camera arrangement modeling system 10 includes a range setting unit 11 for setting an analysis range in map data of a predetermined region, and a grid generation unit for forming a grid including a plurality of unit grids each having numerical information within the analysis range. (12) a map generator 13 for assigning a weight to each of the plurality of unit grids according to the numerical information of each of the plurality of unit grids, and forming a map capable of area division according to the size of the weights; And a model generator 14 generating one or more batch modeling information of one or more surveillance cameras arranged to correspond to a specific position in the map.

Here, the map data includes terrain information, coordinate information, building information, building shape information, surveillance camera installation location information, and surveillance camera installation number information of the predetermined region. In this case, the range setting unit 11 may set the analysis range using the terrain information, the building information, and the coordinate information included in the map data.

The analysis range means the same range as that for the coverage analysis described above with reference to FIGS. 2 to 41.

In addition, the plurality of grids are arranged in a shape having an interval of 1m to 10m, the numerical information is any one of the corresponding floating population numerical data, crime rate numerical data and traffic volume numerical data, the magnitude of the weight of the numerical information It can be proportional to the size of the data value.

The weight also has the same meaning as the weight described above with reference to FIGS. 2 to 41.

In addition, the grid generation unit 12 may form a grid including a plurality of unit grids having numerical information in the analysis range, respectively, and may extract the building shape information and the planar area included in the analysis range.

At this time, the map generation unit 13 targets a plurality of unit grids formed in an area other than the planar area among the analysis ranges, and the plane according to the numerical information of each of the plurality of unit grids formed in the area other than the planar area. A weight may be given to each of a plurality of unit grids formed in regions other than the region, and a map capable of distinguishing regions according to the size of the weight may be formed.

In addition, the map generation unit 13 weights each of the plurality of unit grids according to the numerical information of each of the plurality of unit grids, and the plurality of unit grids each have a different color according to the magnitude of the weight. It is also possible to form a prominent heat map.

In addition, the one or more batch modeling information formed through the model generator 14 may include data of the number of one or more surveillance cameras arranged to correspond to a specific position in the map, and the surveillance camera batch modeling system 10 may include The model recommendation unit 15 may further include modeling information that provides modeling information having the smallest number data value of the surveillance camera included in each of the one or more batch modeling information.

In addition, the model generator 14 may map the map based on the number information of the one or more surveillance cameras arranged to correspond to a specific position in the map, the position information of the positionable area, the effective viewing angle information, the effective viewing distance information, and the effective surveillance area information. A plurality of batch modeling information of one or more surveillance cameras arranged to correspond to a specific position in the apparatus may be generated.

That is, the model generator 14 may generate a plurality of camera layout modeling information corresponding to a specific location of the predetermined region through the coverage analysis described above with reference to FIGS. 2 to 41.

As described above with reference to FIGS. 2 to 41, one or more surveillance cameras arranged to correspond to a specific position in the map may have a horizontal viewing angle of 0 ° to 360 °, a vertical viewing angle of 90 ° to -90 °, and It may have a set coverage range.

In addition, the above-described model recommendation unit 15 is a surveillance camera disposed in the Voronoi region VOR (Si) calculated by Equation 1 and Equation 2 below among one or more surveillance cameras arranged to correspond to a specific position in the map. Using the number information of the, may provide specific batch modeling information from one or more batch modeling information generated by the model generator 14.

(수식 1)

(Formula 1)

(수식 2)

(Formula 2)

Here, S, Si and Sj are a set of buildings corresponding to the specific position, and (Si∪Sj) ∈S, Si = {s1, s2, ..., sn}, Sj = {s1, s2, .. , sm}, Dist (a, b) represents the Euclidean distance between any point a and b in the map data, and point #p represents any point belonging to space E containing a specific location in the map. Si is composed of all points closer to Si than Sj. The building corresponding to the specific location may be a building formed at a specific location. E may be an entire space or a predetermined space in the map.

In addition, the model recommendation unit 15 may use a specific arrangement among one or more arrangement modeling information generated by the model generation unit 14 using the number of cameras arranged in the edge region among the Voronoi region VOR (Si). Modeling information can be provided. The edge area refers to an area within a predetermined interval with respect to the Voronoi area outline.

In order to help understanding the surveillance camera layout model generation and recommendation according to various embodiments of the present disclosure, the model generator 14 generates three layout modeling information, first information, second information, and third information. The first information to the third information have one, four, and four numerical value information which are information on the number of surveillance cameras arranged in the Voronoi region VOR (Si), respectively, and the second information and the third information are each boroughs. As an example, the number information of the surveillance cameras arranged in the edge area of the noi area has two numerical value information and three numerical value information.

In the above example, the model recommendation unit 15 may provide the second information or the third information by using the number information of the surveillance cameras disposed in the Voronoi region VOR (Si), and the model recommendation unit 15 may be provided. The third information may be provided using information on the number of cameras disposed in the edge region among the Voronoi region VOR (Si).

Unlike the above example, when the first information to the third information, which is the surveillance camera arrangement modeling information, respectively have two, three, and four numerical value information as the number of surveillance camera data, the model recommendation unit 15 is a surveillance camera. The first information having the smallest count data value may be provided.

That is, the model recommendation unit 15 may extract the optimal modeling information from the one or more batch modeling information generated by the model generator 14 by analyzing the Voronoi diagram described above with reference to FIGS. 2 to 41. The optimal modeling information may be modeling information corresponding to a first rank by evaluating, quantifying, or indexing numerical value information included in each modeling information.

In this regard, using information on the present specification may mean setting the ranking by evaluating, quantifying, or indexing content included in each information.

43 is a diagram illustrating a procedure of a surveillance camera batch modeling method (hereinafter, referred to as a surveillance camera batch modeling method) according to another embodiment of the present invention.

The surveillance camera batch modeling method is a method using the surveillance camera batch modeling system 10 illustrated in FIG. 42, in which the surveillance camera batch modeling system 10 sets an analysis range on map data of a predetermined region (S21). The grid generation step (S22) of forming a grid comprising a plurality of unit grids each having numerical information within the analysis range, the surveillance camera arrangement modeling system 10, the surveillance camera arrangement modeling system 10 is the plurality of units A map generation step (S23) of assigning a weight to each of the plurality of unit grids according to the numerical information of each grid, and forming a map capable of area classification according to the size of the weight, and the surveillance camera arrangement modeling system 10 ) Is a batch modeling information of one or more surveillance cameras arranged to correspond to a specific position in the map The model generation step (S24) of generating the above is included.

Here, the map data may include terrain information, coordinate information, building information, building shape information, surveillance camera installation location information, and surveillance camera installation number information of the predetermined region. At this time, the range setting step (S21) may be the step of setting the analysis range by the surveillance camera arrangement modeling system 10 using the terrain information, building information and coordinate information included in the map data.

The analysis range means the same range as that for the coverage analysis described above with reference to FIGS. 2 to 41.

In addition, the plurality of grids are arranged in a shape having an interval of 1m to 10m, the numerical information is any one of the corresponding floating population numerical data, crime rate numerical data and traffic volume numerical data, the magnitude of the weight of the numerical information It can be proportional to the size of the data value.

The weight also has the same meaning as the weight described above with reference to FIGS. 2 to 41.

In addition, in the grid generation step (S22), the surveillance camera layout modeling system 10 forms a grid including a plurality of unit grids each having numerical information within the analysis range, but the building shape information and the planar area included in the analysis range. It may further include the step of extracting.

At this time, the map generation step (S23) is a plurality of unit grids formed in the area other than the planar area, the surveillance camera batch modeling system 10 targets a plurality of unit grids formed in the area other than the planar area of the analysis range The method may include weighting each of a plurality of unit grids formed in regions other than the planar region according to the numerical information, and forming a map capable of distinguishing regions according to the size of the weight.

In addition, the map generation step (S23), the surveillance camera batch modeling system 10 weights each of the plurality of unit grids according to the numerical information of each of the plurality of unit grids, and according to the magnitude of the weight Each of the plurality of unit grids may form a heat map having a different color.

In addition, the one or more batch modeling information formed through the model generation step (S24) may include the data of the number of one or more surveillance cameras are arranged to correspond to a specific position in the map, respectively, wherein the surveillance camera batch modeling method The surveillance camera batch modeling system 10 may further include a model recommendation step S25 of deriving modeling information having the smallest number data value of the surveillance camera included in each of the one or more batch modeling information among the one or more batch modeling information. Can be.

In addition, the model generation step (S24), the number of one or more surveillance cameras arranged so that the surveillance camera arrangement modeling system 10 corresponds to a specific position in the map, the location information available, the effective viewing angle information, the effective viewing distance information And generating a plurality of layout modeling information of one or more surveillance cameras arranged to correspond to a specific location in the map based on the effective surveillance area information.

That is, the model generation step S14 may be a step of generating a plurality of camera arrangement modeling information corresponding to a specific location of the predetermined region through the coverage analysis described above with reference to FIGS. 2 to 41.

As described above with reference to FIGS. 2 to 41, one or more surveillance cameras arranged to correspond to a specific position in the map may have a horizontal viewing angle of 0 ° to 360 °, a vertical viewing angle of 90 ° to -90 °, and a preset view angle. It may have a coverage range.

In addition, the above-described model recommendation step (S25) is a Voronoi region (VOR) of one or more surveillance cameras arranged so that the surveillance camera arrangement modeling system 10 corresponds to a specific position in the map calculated by Equation 3 and Equation 4 below. (Si) may be a step of providing specific batch modeling information among the one or more batch modeling information generated through the model generation step S24 based on the number information of the surveillance cameras arranged in (Si).

(수식 3)

(Formula 3)

(수식 4)

(Formula 4)

Here, S, Si and Sj are a set of buildings corresponding to the specific position, and (Si∪Sj) ∈S, Si = {s1, s2, ..., sn}, Sj = {s1, s2, .. , sm}, Dist (a, b) represents the Euclidean distance between any point a and b in the map data, and point #p represents any point belonging to space E containing a specific location in the map. Si is composed of all points closer to Si than Sj. The building corresponding to the specific location may be a building formed at a specific location. E may be an entire space or a predetermined space in the map.

Specific batch modeling information among one or more batch modeling information generated through model generation step S24 using the number information of cameras disposed in the edge region among the Voronoi region VOR (Si) through the model recommendation step S25. May be provided. The edge area refers to an area within a predetermined interval with respect to the Voronoi area outline.

In order to help understanding the surveillance camera layout model generation and recommendation according to various embodiments of the present disclosure, three batch modeling information, first information, second information, and third information, are generated through the model generation step S24. Each of the first to third information has one, four, and four numerical values, which are information on the number of surveillance cameras disposed in the Voronoi region VOR (Si), respectively, and the second and third information, respectively. As an example, the number information of the surveillance cameras disposed in the edge region of the Voronoi region has two numerical value information and three numerical value information.

In the above example, the second information or the third information may be provided according to the number information of the surveillance cameras disposed in the Voronoi region VOR (Si) through the model recommendation step S25, and the Voronoi region VOR ( Among the Si)), the third information may be provided according to the number information of the cameras arranged in the edge region.

Unlike the above example, in the case where the first information to the third information, which are the surveillance camera arrangement modeling information, have two, three, and four numerical values as the number of monitoring camera data, the monitoring is performed through the model recommendation step S25. First information with the smallest camera number data value may be provided.

That is, the model recommendation step 15 may extract the optimal modeling information from the one or more batch modeling information generated according to the model generation step S14 through analysis of the Voronoi diagram described above with reference to FIGS. 2 to 41. The optimal modeling information may be modeling information corresponding to the first rank by evaluating, quantifying, or indexing numerical value information included in each modeling information.

As described above, the range setting step S21 may include all procedures performed by the range setting unit 11, and likewise, the grid generation step S22, the map generation step S23, and the model generation step ( S24) and the model recommendation step S25 may include all procedures performed by the grid generator 12, the map generator 13, the model generator 14, and the model recommender 15, respectively.

The description of the present invention described above is for illustrative purposes, and it will be understood by those skilled in the art that the present invention may be easily modified in other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are exemplary in all respects and not restrictive. The scope of the present invention is represented by the following claims, and it should be construed that all changes or modifications derived from the meaning and scope of the claims and their equivalents are included in the scope of the present invention.

Claims (16)

A range setting unit for setting an analysis range in map data of a predetermined region;
A grid generator configured to form a grid including a plurality of unit grids each having numerical information within the analysis range;
A map generator which weights each of the plurality of unit grids according to the numerical information of each of the plurality of unit grids, and forms a map capable of distinguishing regions according to the size of the weights; And
It includes a model generator for generating one or more batch modeling information of one or more surveillance cameras arranged to correspond to a specific position in the map,
The map data includes terrain information, coordinate information, building information, building shape information, surveillance camera installation location information, and surveillance camera installation number information of the predetermined region.
The range setting unit sets the analysis range by using terrain information, building information, and coordinate information included in the map data.
The grid generation unit forms a grid including a plurality of unit grids each having numerical information within the analysis range, and extracts the building shape information and the planar area included in the analysis range,
The map generation unit targets a plurality of unit grids formed in an area other than the planar area of the analysis range, and is formed in an area other than the planar area according to numerical information of each of the plurality of unit grids formed in the area other than the planar area. And a weighting factor for each of the plurality of unit grids, and a map capable of distinguishing areas according to the magnitudes of the weights.
delete The method of claim 1,
The plurality of grids are arranged in a shape having an interval of 1m to 10m,
The numerical information is any one of the corresponding floating population numerical data, crime rate numerical data and traffic volume numerical data,
And the magnitude of the weight is in proportion to the magnitude of the data value of the numerical information.
delete The method of claim 1,
The map generator,
A weight is assigned to each of the plurality of unit grids according to the numerical information of each of the plurality of unit grids, and each of the plurality of unit grids forms a heat map having a different color according to the magnitude of the weight. Surveillance camera batch modeling system.
The method of claim 1,
The one or more batch modeling information formed through the model generator includes data of the number of one or more surveillance cameras arranged to correspond to a specific location in the map,
And a model recommendation unit for providing modeling information having the smallest number data value of the surveillance camera included in each of the one or more batch modeling information among the one or more batch modeling information.
The method of claim 1,
The model generator,
One or more arrangements corresponding to a specific position in the map based on the number information of the one or more surveillance cameras arranged to correspond to the specific position in the map, the deployable region information, the effective viewing angle information, the effective viewing distance information, and the effective surveillance region information. Surveillance camera batch modeling system, characterized in that for generating a plurality of batch modeling information of the surveillance camera.
The method of claim 1,
One or more surveillance cameras arranged to correspond to specific locations in the map each have a horizontal viewing angle of 0 ° to 360 °, a vertical viewing angle of 90 ° to -90 °, and a preset coverage range.
Of the one or more surveillance cameras arranged to correspond to a specific position in the map, using the model information using the number information of the surveillance cameras arranged in the Voronoi region (VOR (Si)) calculated by Equation 1 below, generated through the model generator Surveillance camera batch modeling system further comprises a model recommending unit for providing specific batch modeling information from one or more batch modeling information.
[Equation 1]

,

Here, S, Si and Sj are a set of buildings corresponding to the specific position, and (Si∪Sj) ∈S, Si = {s1, s2, ..., sn}, Sj = {s1, s2, .. , sm}, Dist (a, b) represents the Euclidean distance between any point a and b in the map data, and point p represents any point belonging to space E containing a specific location in the map.
In the surveillance camera batch modeling method using the surveillance camera batch modeling system,
A range setting step of setting, by the surveillance camera arrangement modeling system, an analysis range on map data of a predetermined region;
A grid generation step of forming, by the surveillance camera arrangement modeling system, a grid including a plurality of unit grids each having numerical information within the analysis range;
A map generation step of the surveillance camera arrangement modeling system weighting each of the plurality of unit grids according to numerical information of each of the plurality of unit grids, and forming a map capable of area division according to the size of the weights; And
And generating at least one batch modeling information of at least one surveillance camera arranged by the surveillance camera batch modeling system to correspond to a specific position in the map.
The map data includes terrain information, coordinate information, building information, building shape information, surveillance camera installation location information, and surveillance camera installation number information of the predetermined region.
The range setting step is a step of setting the analysis range by the surveillance camera arrangement modeling system using the terrain information, the building information and the coordinate information included in the map data,
In the generating grid step, the surveillance camera arrangement modeling system may form a grid including a plurality of unit grids having numerical information in the analysis range, respectively, and extract the building shape information and the planar area included in the analysis range. Including,
In the map generation step, the surveillance camera arrangement modeling system targets a plurality of unit grids formed in an area other than the planar area among the analysis ranges, and according to numerical information of each of the plurality of unit grids formed in the area other than the planar area. And weighting each of a plurality of unit grids formed in an area other than the planar area, and forming a map capable of classifying areas according to the size of the weight.
delete The method of claim 9,
The plurality of grids are arranged in a shape having an interval of 1m to 10m,
The numerical information is any one of the corresponding floating population numerical data, crime rate numerical data and traffic volume numerical data,
And the magnitude of the weight is in proportion to the magnitude of the data value of the numerical information.
delete The method of claim 9,
The map generation step,
The surveillance camera arrangement modeling system weights each of the plurality of unit grids according to numerical information of each of the plurality of unit grids, and each of the plurality of unit grids has a different color according to the magnitude of the weight. Surveillance camera arrangement modeling method, characterized in that the step of forming a heat map.
The method of claim 9,
The one or more batch modeling information formed through the model generation step includes data of the number of one or more surveillance cameras arranged to correspond to a specific position in the map,
The surveillance camera batch modeling system further comprises a model recommendation step of deriving the modeling information having the smallest number data value of the surveillance camera included in each of the one or more batch modeling information of the one or more batch modeling information; Camera placement modeling method.
The method of claim 9,
The model generation step,
The specific location in the map based on the number information, the deployable area information, the effective viewing angle information, the effective viewing distance information, and the effective surveillance area information of one or more surveillance cameras arranged such that the surveillance camera arrangement modeling system corresponds to the specific location in the map. And generating a plurality of batch modeling information of at least one surveillance camera arranged to correspond to the surveillance camera batch modeling method.
The method of claim 9,
One or more surveillance cameras arranged to correspond to specific locations in the map each have a horizontal viewing angle of 0 ° to 360 °, a vertical viewing angle of 90 ° to -90 °, and a preset coverage range.
Among the one or more surveillance cameras arranged to correspond to a specific position in the map, the surveillance camera arrangement modeling system uses information on the number of surveillance cameras arranged in the Voronoi region VOR (Si) calculated by Equation 2 below. And a model recommendation step of providing specific batch modeling information among the one or more batch modeling information generated through the model generation step.
[Equation 2]

,

Here, S, Si and Sj are a set of buildings corresponding to the specific position, and (Si∪Sj) ∈S, Si = {s1, s2, ..., sn}, Sj = {s1, s2, .. , sm}, Dist (a, b) represents the Euclidean distance between any point a and b in the map data, and point p represents any point belonging to space E containing a specific location in the map.

Images