JP7305965B2 - Video surveillance system parameter setting method, device and video surveillance system - Google Patents

Description

The present invention relates to the field of information technology, and more particularly to a method, apparatus and video surveillance system for configuring parameters for a video surveillance system.

In recent years, with the increase in the number of vehicles, traffic problems such as traffic congestion, traffic accidents, air pollution, etc. become more serious. In order to monitor and mitigate such traffic problems, the installation of large-scale surveillance cameras allows video surveillance systems to provide real-time detection and detection of traffic events based on a large number of video sources. can report.

Such a video surveillance system usually supports detection of multiple traffic events such as illegal parking, vehicle flow, traffic congestion, etc., and can facilitate operations such as traffic guidance.

The inventors of the present invention have discovered the following. That is, conventional video surveillance systems require the user to manually set (provide) configuration parameters to the video surveillance system before the video surveillance system can detect and report otherwise the desired detection results. may not be obtained. Among them, the setting parameters may include, for example, the detection area in the image, the analysis parameters for image analysis, etc. The analysis parameters are, for example, the number of roads, the driving direction of the vehicle, the threshold for analysis, and other parameters. It can be.

Manually setting configuration parameters in a video surveillance system by a user has at least the following problems.

1. When the number of surveillance cameras is relatively large, such as 1000, it is a difficult task for the user to set all the setting parameters in the video surveillance system;
2. Once the camera position, observation angle or field of view changes, the user needs to manually reset the relevant parameters;
3. It is difficult to support movable cameras, such as PTZ (Pan/Tilt/Zoom, PTZ) cameras. This is because the movable camera can freely change the viewing angle or field of view, and each time the change requires the user to manually reset the relevant parameters.

Among them, regarding the above-mentioned problem 3, there are the following technical proposals in the prior art. That is, the correspondence relationship between the predetermined state of the PTZ camera and the setting parameters is set in advance, and when the PTZ camera is in a certain state, the setting parameter corresponding to the predetermined state is called to control the video surveillance system. Make settings for However, this technical solution can only support some types of cameras and is difficult to apply to all types of cameras. In addition, since the number of predetermined states that can be supported by such a technical solution is limited, it is difficult to perform effective parameter settings when the camera is in a state other than the predetermined state. Therefore, such technical solutions are greatly limited when used.

Embodiments of the present invention provide a method, apparatus and video surveillance system for setting parameters for a video surveillance system, generating the setting parameters based on images obtained by cameras of the video surveillance system; By setting (providing) the setting parameters to the video surveillance system, effective and quick parameter setting for the video surveillance system can be performed, thus manpower can be saved and applicability is wide.

According to a first aspect of an embodiment of the present invention, there is provided an apparatus for setting parameters for a video surveillance system, comprising:
a generating unit for generating configuration parameters based on images obtained by cameras of said video surveillance system; and a configuration unit for setting said configuration parameters to said video surveillance system.

According to a second aspect of an embodiment of the present invention, there is provided a method of setting parameters for a video surveillance system, comprising:
generating configuration parameters based on images obtained by cameras of the video surveillance system; and setting the configuration parameters to the video surveillance system.

According to a third aspect of an embodiment of the present invention there is provided a video surveillance system comprising a camera, an analysis device and a device for setting parameters for a video surveillance system as described in the first aspect above. wherein the device for setting parameters for the video surveillance system generates setting parameters based on the images obtained by the camera, and sets (provides) the setting parameters to the analysis device; The analysis device analyzes the images obtained by the camera based on the set parameters.

A beneficial effect of the present invention is: generating configuration parameters based on images obtained by cameras of a video surveillance system, and setting the configuration parameters to the video surveillance system; It can effectively and quickly set parameters for the video surveillance system, thus saving manpower and having wide applicability.

1 is a diagram showing a parameter setting device for a video surveillance system according to Embodiment 1 of the present invention; FIG. FIG. 4 is a diagram showing a generation unit in Example 1 of the present invention; FIG. 10 is a diagram showing a division result of image division by the scene analysis unit in Example 1 of the present invention; FIG. 4 is a diagram showing the position of the region of the first parameter on the image in Example 1 of the present invention; FIG. 4 is a diagram showing a parameter generation unit in Example 1 of the present invention; FIG. 10 is a diagram showing a parameter setting method for a video surveillance system according to Embodiment 2 of the present invention; FIG. 10 is a diagram showing a method of generating setting parameters in Example 2 of the present invention; FIG. 10 is a diagram showing a method of generating a parameter different from the first parameter in Example 2 of the present invention; FIG. 10 is a diagram showing a video surveillance system in Example 3 of the present invention; FIG. 10 is a diagram showing a monitoring flow by the video monitoring system in Example 3 of the present invention; FIG. 10 is a configuration diagram of an electronic device according to Example 4 of the present invention;

Preferred embodiments for carrying out the present invention will now be described in detail with reference to the accompanying drawings.

Embodiment 1 of the present invention provides a parameter setting device for a video surveillance system.

FIG. 1 is a diagram showing a parameter setting device for the video surveillance system of this embodiment. As shown in FIG. 1, a parameter setting device 100 for a video surveillance system includes a generating unit 101 and a setting unit 102 .

In this embodiment, the generation unit 101 generates configuration parameters based on images obtained (eg, captured) by cameras of the video surveillance system, and the configuration unit 102 generates the configuration parameters generated by the generation unit 101. Used to set configuration parameters to the video surveillance system.

According to this embodiment, a configuration parameter is generated based on the images obtained by the cameras of the video surveillance system, and the configuration parameters are set in the video surveillance system to effect a video surveillance system. And it can quickly set parameters, which saves manpower and has wide applicability.

In this embodiment, the cameras of the video surveillance system may be stationary cameras (fixed cameras), for example stationary cameras mounted on both sides of the road, or the cameras may be movable The movable camera may be a camera installed on a moving object, such as a camera installed in a car, a camera mounted on a drone, or a mobile terminal carried by a pedestrian. Alternatively, the movable camera may be a camera with a variable viewing angle, such as a PTZ (Pan/Tilt/Zoom, PTZ) camera.

In this embodiment, the image obtained by the camera may be an image obtained by photographing a scene to be monitored, and the scene to be monitored may be, for example, a road, a forest, a building, or the like. can be In the following description, an image obtained by photographing a road will be described as an example, but the present invention is not limited to this, and the image is photographed for another scene. It may be obtained by

FIG. 2 is a diagram showing the generation unit 101 in this embodiment. The generation unit 101 may include a scene analysis unit 201, as shown in FIG.

In this embodiment, the scene analysis unit 201 is used to divide the image into different regions. The scene analysis unit 201 can divide the image into different regions by analyzing the image based on the model obtained by deep learning method.

For example, by adopting a sample image with a scene similar to the image and performing offline training based on deep learning, a pixel (pixel) level deep learning model can be obtained, among which, in the sample image, each Pixels may be labeled as lanes, roadside areas, trees, sky, and so on. A scene analysis unit 201 can use the pixel-level deep learning model obtained by the training to perform analysis on the image obtained by the camera to divide the image into different regions. .

FIG. 3 is a diagram showing the division result of image division by the scene analysis unit 201 in this embodiment. As shown in FIG. 3, the first segmented region 301 represents the region occupied by the road, the second segmented region 302 represents the tree or green belt region, and the third segmented region 303 represents the vehicle. represents the area occupied by , the fourth area 304 resulting from the division represents the area occupied by buildings, and the fifth area 305 resulting from the division represents the area occupied by the sky.

In this embodiment, the scene analysis unit 201 may select one model out of at least two models based on features of the image, and divide the image into different regions according to the selected model. It is possible to improve the accuracy of the segmentation by adaptively selecting the model to perform the segmentation. Among them, the features of the image may be, for example, features such as tone, brightness, etc. of the image. The at least two models may be different models obtained by training on the same scene, e.g. a model for roads in the case of normal daytime sunshine and a model for the lack of sunshine ( For example, it may be a model for roads in cloudy weather, rain, etc.).

For example, the scene analysis unit 201 selects a model for the road in the case of lack of sunlight when the average color tone of the image is relatively low, and divides the image into different regions using the selected model. can do.

In this embodiment, the model used by the scene analysis unit 201 to perform segmentation may have different identification accuracies (identification accuracy) for different locations in the image, e.g. It may have higher identification accuracy for areas that are important to the video surveillance system and lower identification accuracy for areas that are less important to the video surveillance system. This makes it possible to make the results of the division more to the point and improve the efficiency of the division.

For example, the model used by the scene analysis unit 201 may discriminate with relatively high accuracy for areas occupied by and near roads or green belts, for areas occupied by trees, etc. A relatively low accuracy identification may be made. Note that the model can be obtained by biased training on different regions.

In this embodiment, scene analysis unit 201 performs segmentation on at least two frames of the image, and generates the segmentation results based on segmentation results on the at least two frames of the image. can do. This makes it possible to further improve the accuracy of division. For example, the scene analysis unit 201 performs a process of dividing a plurality of frames of images obtained by a camera within a predetermined period (for example, 2 seconds) into different regions, and then divides the plurality of frames of images into different regions. It is possible to obtain the image segmentation result within the predetermined period by statistically dividing the segmentation result for the .

Note that the model and method used by the scene analysis unit 201 to divide the image are not limited to this, and other models or other methods may be adopted.

In this embodiment, the segmentation result obtained by segmenting the image by the scene analysis unit 201 may be used as the first parameter in the setting parameters. A first parameter may be used to indicate the area in the image to be detected by the video surveillance system.

FIG. 4 is a diagram showing the position of the area of the first parameter on the image. As shown in FIG. 4, in the image 400 obtained by the camera, the area 401 is the area occupied by the road, the area 402 is the roadside area, and the area 403 is the area occupied by the sky.

In this embodiment, the area 401 may be used for vehicle flow statistics, detection of events such as reverse driving, among which the number of roads shown in the area 401 is 4, and the area 402 is It may be used for detection of events such as illegal parking, road encroachment, etc. Region 403 is used for detection of events such as yellow sand, PM2.5 (fine particulate matter), smoke generation, etc. May be used.

In this embodiment, the generation unit 101 may further include a parameter generation unit 202, as shown in FIG. A parameter generation unit 202 can generate a parameter different from the first parameter in the setting parameters based on the image and the segmentation result of the image segmentation by the scene analysis unit 201 .

FIG. 5 is a diagram showing a parameter generation unit in this embodiment. The parameter generation unit 202 may include an intermediate processing unit 501, as shown in FIG.

In this embodiment, the intermediate processing unit 501 may perform processing on the image in each region segmented by the scene analysis unit 201, such as edge detection, vehicle detection, (vehicle detection), vehicle tracking, trajectory generation, brightness detection, and the like. In addition, since the conventional technique can be referred to for the specific method of each process, the detailed description thereof will be omitted in the present embodiment.

In this embodiment, the processing performed in each area by the intermediate processing unit 501 may correspond only to that area. For example, brightness detection may be performed in the fifth region 305 of FIG. 3, and vehicle detection and/or vehicle tracking and/or trajectory generation may be performed in the first region 301 of FIG. In other words, the intermediate processing unit 501 may only perform processing corresponding to that region in each region, ie, it need not perform all processing in all regions of the image. This can reduce the amount of calculation for intermediate processing.

In this embodiment, at least part of the processing result of the intermediate processing unit 501 may be used as the second parameter among the setting parameters. For example, the result of luminance detection includes visibility information of the road surface, so the result of luminance detection for the fifth region 305 is taken as the second parameter, reflecting the visibility value of the road surface. Also, for example, the result of trajectory generation, that is, the trajectory includes road direction information, so the result of trajectory generation for the first region 301 is the second parameter. , may reflect the direction of the road.

In this embodiment, the parameter generation unit 202 may further include a fusion unit 502, as shown in FIG. The fusion unit 502 can generate a third parameter among setting parameters based on the processing result of the intermediate processing unit 501 and the segmentation result of segmentation of the image by the scene analysis unit 201 .

For example, the fusion unit 502 can perform fusion on the vehicle detection result of the intermediate processing unit 501 and the road area segmented by the scene analysis unit 201 to calculate the number of roads, and for example, The fusion unit 502 compares the width of the road area divided by the scene analysis unit 201 with the width of the vehicle in the vehicle detection result of the intermediate processing unit 501, and calculates the number of roads in the road area. , and the number of roads is taken as the third parameter. Further, for example, the edge detection result of the intermediate processing unit 501 may be used to segment connected regions in an image, so that the edge detection result and the road region segmented by the scene analysis unit 201 You can also get the number of roads by fusing against .

In this embodiment, as described above, the generation unit 101 can generate configuration parameters based on the images obtained by the cameras of the video surveillance system, and the configuration parameters generated by the generation unit 101 include: A first parameter (e.g. the area to be detected by the video surveillance system in an image, etc.) and/or a second parameter (e.g. result of luminance detection, driving trajectory, etc.) and/or a third parameter (eg, number of roads, etc.).

In this embodiment, the configuration unit 102 can generate a configuration file based on the configuration parameters generated by the generation unit 101, and the configuration file is, for example, a configuration table that stores (records) the configuration parameters. There may be and the configuration table can be used to set the configuration parameters to the video surveillance system.

According to this embodiment, a configuration parameter is generated based on the images obtained by the cameras of the video surveillance system, and the configuration parameters are set in the video surveillance system to effect a video surveillance system. In addition, the parameters can be quickly set, manpower can be saved, and since the parameters are set automatically, the parameters can be quickly set even if there is a change in the state of the camera. Therefore, the parameter setting device for video surveillance system according to this embodiment can support multiple types of cameras.

Embodiment 2 of the present invention provides a parameter setting method for a video surveillance system, which corresponds to the parameter setting device 100 for video surveillance system in Embodiment 1.

FIG. 6 is a diagram showing a parameter setting method for the video surveillance system in this embodiment. As shown in FIG. 6, the method includes the following steps.

Step 601: Generate configuration parameters based on the images obtained by the cameras of the video surveillance system.

Step 602: Set the configuration parameters to the video surveillance system.

FIG. 7 is a diagram showing a method of generating setting parameters in this embodiment. As shown in FIG. 7, the configuration parameter generation method includes the following steps.

Step 701: Divide the image into different regions, wherein the division result of dividing the image is taken as the first parameter among the setting parameters.

In step 701, the image can be segmented into different regions by analyzing the image based on the model obtained by the deep learning method.

In step 701 of the present embodiment, furthermore, at least one of the following methods can be used to further improve the segmentation accuracy: at least two regions based on the image features; selecting one of the models, and segmenting the image into different regions based on the selected model; such that the models used for segmentation have different identification accuracies for different locations of the image. and performing segmentation on at least two frames of the image, and generating the segmentation result based on segmentation results on at least two frames of the image.

In this embodiment, as shown in FIG. 7, the configuration parameter generation method may further include the following steps.

Step 702: Generate a parameter different from the first parameter among the set parameters according to the image and the segmentation result for the image.

FIG. 8 is a diagram showing a method of generating a parameter different from the first parameter in step 702 in this embodiment. As shown in FIG. 8, the method includes the following steps.

Step 801: For each divided region, perform intermediate processing corresponding to the region on the image, and set at least a part of the result of the intermediate processing as the second parameter among the setting parameters.

In this embodiment, as shown in FIG. 8, the method may further include the following steps.

Step 802: Generate a third parameter among the setting parameters based on the result of the intermediate processing and the segmentation result for the image.

According to this embodiment, a configuration parameter is generated based on the images obtained by the cameras of the video surveillance system, and the configuration parameters are set in the video surveillance system to effect a video surveillance system. In addition, the parameters can be quickly set to save manpower, and since the parameters are automatically set, the parameters can be quickly set regardless of changes in the camera status. Therefore, the parameter setting method for the video surveillance system according to this embodiment can support multiple kinds of cameras.

Embodiment 3 of the present invention provides a video surveillance system, wherein said video surveillance system comprises a parameter setting device for video surveillance system according to embodiment 1.

FIG. 9 is a diagram showing a video surveillance system in embodiment 3 of the present invention. As shown in Figure 9, the video surveillance system 900 includes a camera 901, an analysis device 902, and a parameter setting device 903 for the video surveillance system.

Among them, the parameter setting device 903 for the video surveillance system generates setting parameters based on the image obtained by the camera 901, and sets (provides) the setting parameters to the analysis device 902, and the analysis device 902 can analyze the images obtained by the camera 901 based on the set parameters.

In this embodiment, the camera 901 can be a fixed camera or a movable camera, wherein the movable camera is, for example, a camera with a variable viewing angle, such as a PTZ (Pan/Tilt/Zoom, PTZ ) may be a camera or the like.

In this embodiment, the analysis device 902 is configured based on the set parameters to perform analysis on the images obtained by the one or more cameras 901, such as, for example, event detection. It can be. For example, when the analysis device 902 detects a particular event, it can send an alarm signal and/or a screenshot or video of the event to the control server. Analyzer 902 may be implemented in hardware, software, or a combination of software and hardware. In this embodiment, as shown in FIG. 9, the analysis device 902 may be configured in an analysis server 902A, and the number of analysis devices 902 in the analysis server 902A equals the number of surveillance cameras 901 to be analyzed. may be set based on As for the description of the analysis device 902 and the analysis server 902A, the conventional technology can be referred to, so detailed description thereof will be omitted in this embodiment.

In this embodiment, the working principle of the parameter setting device 903 for video surveillance system can refer to the description in the first embodiment. As shown in Figure 9, the parameter setting device 903 for the video surveillance system may be configured in a management server 903A.

As shown in FIG. 9, the management server 903A may further have a management device 904, in which the management device 904 controls the hardware in the video surveillance system 900 and analyzes the alarms sent by the analysis device 902. A screenshot or video of the signal and/or the event may be received. For the description of the management device 904 and the management server 903A, the conventional technology can be referred to, so detailed description thereof will be omitted in this embodiment.

In this embodiment, the video surveillance system 900 may have only one management server 903A.

In this embodiment, as shown in FIG. 9, the video surveillance system 900 may be configured with only one parameter setting device 903 for the video surveillance system, whereby the device 903 can Based on the image of 901 , parameter settings can be made for the analysis device 902 corresponding to the camera 901 . Therefore, costs can be reduced.

However, the present invention is not limited to this, and the video surveillance system 900 may be configured with multiple parameter setting devices 903 for the video surveillance system. Also, a plurality of parameter setting devices 903 for the video surveillance system may be configured in the management server 903A, or may be configured in each analysis server 902A.

As shown in FIG. 9, the video surveillance system 900 may further comprise a detection device 905, which is used to detect changes in the state of the camera, wherein the state of the camera is detected. A change in refers to a change in at least one of the position of the camera, the viewing angle of the camera, and the viewing range of the camera. In this embodiment, the detection device 905 may detect a change in camera state based on a sensor and/or may detect a change in camera state based on image processing results. .

In this embodiment, the parameter setting device 903 for video surveillance system regenerates the setting parameters when the detection device 905 detects a change in the state of the camera 901, and the regenerated setting parameters are It can be set in the analyzer 902 .

As shown in FIG. 9, the video surveillance system 900 may further include a display device 906 , which can display the analysis results of the analysis device 902 .

In this embodiment, the physical installation method of each component in the video surveillance system 900 is not limited to FIG. 9, and other installation methods may be adopted.

FIG. 10 is a diagram showing a monitoring flow by the video monitoring system 900 in this embodiment. As shown in FIG. 10, the monitoring flow may include the following steps.

Step 1001: Acquire an image with a camera.

Step 1002: Determine whether initialization is required. If yes, step 1004 is performed; if no, step 1003 is performed. Initialization refers to setting parameters by the parameter setting device 903 for the video surveillance system. In this step, it is possible to determine if the video surveillance system 900 needs to be initialized by determining if the relevant configuration parameters have already been set; i.e., it needs to be initialized, and if it is set, it determines "no".

Step 1003: Determine if there is a change in the state of the camera. For example, the detection device 905 may make the determination. If yes, step 1004 is performed; if no, step 1007 is performed.

Step 1004: Generate and set configuration parameters. Step 1004 includes steps 1005 and 1006 of generating configuration parameters. Among them, the parameters generated and set in step 1004 are, for example, the area in the image to be detected by the video surveillance system, and/or the result of luminance detection, and/or the driving trajectory, and/or , the number of roads, etc. Note that steps 1005 and 1006 shown in FIG. 10 only show an example of generating setting parameters, but the description in the first embodiment can be referred to for other examples.

Step 1005: Divide the image into different regions. This step corresponds to the function of the scene analysis unit 201 in the first embodiment.

Step 1006: Based on the image and the segmentation result in step 1005, a parameter different from the first parameter among the setting parameters is generated. This step corresponds to the function of the parameter generation unit 202 in the first embodiment.

Step 1007: Perform analysis on the image acquired by the camera according to the set parameters. The steps can be performed by the analyzer 902, for example.

The video surveillance system 900 in this embodiment may be used for traffic monitoring, or may be used for other fields such as forest monitoring, natural disaster monitoring, building monitoring, and so on.

According to this embodiment, configuration parameters are generated based on the images obtained by the cameras of the video surveillance system, and the configuration parameters are set in the video surveillance system, thereby enabling and validating the video surveillance system. The parameters can be quickly set to save manpower, and since the parameters are automatically set, the parameters can be quickly set even if the camera status changes. Thus, the video surveillance system according to this embodiment can support multiple types of cameras.

Embodiment 4 of the present invention provides an electronic device, which includes a parameter setting device for a video surveillance system as described in embodiment 1.

In this embodiment, the electronic device may be the management server 903A described in embodiment 3, whereby the function of the parameter setting device for the video surveillance system can be configured in the management server 903A. Also, in this embodiment, the electronic device may be the analysis server 902A described in embodiment 3, whereby the function of the parameter setting device for the video surveillance system can be configured in the analysis server 902A. .

FIG. 11 is a configuration diagram of an electronic device according to Example 4 of the present invention. As shown in FIG. 11, the electronic device 1100 may include a central processing unit (CPU) 1101 and a memory 1102 , the memory 1102 being connected to the central processing unit 1101 . The storage unit 1102 can store various data, and can also store a parameter setting program for the video surveillance system, and execute the program under the control of the central processing unit 1101. You can also run

In one implementation, the functionality of a parameter setter for a video surveillance system can be integrated into central processing unit 1101 .

Wherein, the central processing unit 1101 may be configured to: generate configuration parameters based on the images obtained by the cameras of the video surveillance system; set to

Wherein, the central processing unit 1101 may further be configured to: divide the image into different regions, and take the division result of the image as the first parameter among the configuration parameters.

Wherein the central processing unit 1101 may be further configured to: select one model from at least two models according to the features of the image; divide the image into different regions based on .

Wherein, the central processing unit 1101 may further be set so that the model used for segmentation has different identification accuracy for different positions of the image.

Wherein the central processing unit 1101 may be further configured to: perform segmentation for at least two frames of the image; The division result is generated based on the division result of .

Wherein, the central processing unit 1101 may further be configured to set a parameter different from the first parameter among the configuration parameters based on the image and the segmentation result for the image. Generate.

Wherein, the central processing unit 1101 may be further configured as follows: in each divided region, perform intermediate processing on the image corresponding to the region; At least part of the result of is set as the second parameter among the setting parameters.

Wherein, the central processing unit 1101 may be further configured to generate a third parameter among the configuration parameters based on the intermediate processing result and the image segmentation result.

In addition, as shown in FIG. 11, the electronic device 1100 may further include an input/output unit 1103, a display unit 1104, etc., among which the functions of these parts are similar to the prior art, so here, A detailed description thereof will be omitted. Note that electronic device 1100 does not necessarily include all the components in FIG. In addition, the electronic device 1100 may further include components not shown in FIG. 11, for which reference can be made to the prior art.

An embodiment of the present invention further provides a computer-readable program, wherein the program, when executed in a parameter setting device or an electronic device for a video surveillance system, reads: A parameter setting device or an electronic device (computer) can be caused to execute the parameter setting method for the video surveillance system described in the second embodiment.

An embodiment of the present invention further provides a storage medium storing a computer-readable program, wherein the storage medium stores the computer-readable program described above, wherein the computer-readable program is for a video surveillance system. can be caused to execute the parameter setting method for the video surveillance system described in the second embodiment.

Also, the apparatus, method, etc. according to the embodiments of the present invention may be implemented by software, hardware, or a combination of hardware and software. The invention also relates to such a computer readable program, i.e., said program, when executed by a logic component, is capable of causing said logic component to implement the device or component described above, or A component may implement the method or steps thereof described above. Furthermore, the present invention also relates to a storage medium storing the above program, such as a hard disk, magnetic disk, optical disk, DVD, fresh memory, and the like.

In addition, additional remarks are disclosed as follows with respect to the above-described examples and the like.

(Appendix 1)
A device for setting parameters for a video surveillance system, comprising:
An apparatus comprising: a generating unit for generating configuration parameters based on images obtained by cameras of said video surveillance system; and a configuration unit for setting said configuration parameters to said video surveillance system.

(Appendix 2)
A device according to Appendix 1, wherein
The generating unit is
comprising a scene analysis unit for segmenting the image into different regions;
The apparatus, wherein the segmentation result of the image by the scene analysis unit is the first parameter among the setting parameters.

(Appendix 3)
The device of Appendix 2, wherein
The apparatus of claim 1, wherein the scene analysis unit selects one model from among at least two models based on features of the image and divides the image into different regions based on the selected model.

(Appendix 4)
The device of Appendix 2, wherein
The apparatus of claim 1, wherein the models used by the scene analysis unit to perform the segmentation have different identification accuracies for different locations of the image.

(Appendix 5)
The device of Appendix 2, wherein
The apparatus, wherein the scene analysis unit performs segmentation on at least two frames of the image and generates the segmentation result based on segmentation results on at least two frames of the image.

(Appendix 6)
The device of Appendix 2, wherein
The generating unit further comprises:
a parameter generation unit that generates a parameter different from the first parameter among the setting parameters based on the image and a segmentation result of the image by the scene analysis unit.

(Appendix 7)
The device of clause 6, wherein
The parameter generation unit is
performing processing corresponding to each region on the image in each region divided by the scene analysis unit;
An apparatus, wherein at least part of the processing result of the intermediate processing unit is a second parameter among the configuration parameters.

(Appendix 8)
The device of clause 7, wherein
The parameter generation unit further comprises:
An apparatus, comprising: a fusion unit for generating a third parameter among said set parameters based on a processing result of said intermediate processing unit and a segmentation result of said image by said scene analysis unit.

(Appendix 9)
A video surveillance system,
comprising a camera, an analysis device and a device for setting parameters for a video surveillance system according to any one of Appendices 1 to 8;
a device for setting parameters for the video surveillance system, generating setting parameters based on the images obtained by the cameras, setting the setting parameters to the analysis device;
The video surveillance system, wherein the analysis device analyzes the images obtained by the camera based on the set parameters.

(Appendix 10)
The video surveillance system of Clause 9, wherein
The video surveillance system further comprises:
a detection device for detecting a change in state of the camera;
A device for setting parameters for the video surveillance system generates the setting parameters and sets the setting parameters to the analysis device when a change in the state of the camera is detected by the detection device. Monitoring system.

(Appendix 11)
A method of setting parameters for a video surveillance system, comprising:
A method comprising: generating configuration parameters based on images obtained by cameras of said video surveillance system; and configuring said configuration parameters in said video surveillance system.

(Appendix 12)
11. The method of Supplementary Note 11,
Generating the configuration parameters includes:
dividing the image into different regions;
A method, wherein a segmentation result for the image is a first parameter among the set parameters.

(Appendix 13)
12. The method of Appendix 12, wherein
Dividing the image into different regions includes:
A method comprising: selecting one of at least two models based on features of said image; and dividing said image into different regions based on said selected model.

(Appendix 14)
12. The method of Appendix 12, wherein
A method according to claim 1, wherein models used for segmentation have different identification accuracies for different positions of the image.

(Appendix 15)
12. The method of Appendix 12, wherein
Dividing the image into different regions includes:
A method comprising: segmenting at least two frames of said image; and generating said segmentation results based on segmentation results for at least two frames of said image.

(Appendix 16)
12. The method of Appendix 12, wherein
Generating the configuration parameters further includes:
generating a parameter different from the first parameter among the set parameters based on the image and a segmentation result for the image.

(Appendix 17)
16. The method of Supplementary Note 16,
Generating a parameter different from the first parameter in the configuration parameters includes:
In each divided region, performing intermediate processing corresponding to the region on the image,
The method, wherein at least part of the result of the intermediate processing is a second parameter among the configuration parameters.

(Appendix 18)
17. The method of Supplementary Note 17,
Generating a parameter different from the first parameter in the configuration parameters further includes:
generating a third parameter among the set parameters based on a result of the intermediate processing and a segmentation result for the image.

Although the preferred embodiment of the present invention has been described above, the present invention is not limited to this embodiment, and all modifications to the present invention fall within the technical scope of the present invention as long as they do not depart from the gist of the present invention.

Claims (10)

A device for setting parameters for a video surveillance system, comprising:
a generation unit for generating configuration parameters based on images obtained by cameras of said video surveillance system; and a configuration unit for setting said configuration parameters to said video surveillance system ,
the generation unit includes a scene analysis unit that divides the image into different regions;
A result of segmentation of the image by the scene analysis unit is set as a first parameter among the setting parameters,
wherein the scene analysis unit selects one model from among at least two models based on features of the image and divides the image into different regions based on the selected model;
the models used by the scene analysis unit to perform segmentation have different identification accuracies for different locations of the image;
The apparatus, wherein the scene analysis unit performs segmentation on at least two frames of the image and generates a segmentation result of the image based on segmentation results on at least two frames of the image. 2. The device of claim 1 , wherein
The generating unit is
The apparatus further comprising a parameter generation unit for generating a parameter different from the first parameter among the set parameters based on the image and the segmentation result of the image by the scene analysis unit. 3. The apparatus of claim 2 , wherein
The parameter generation unit is
An intermediate processing unit that performs processing corresponding to each region on the image in each region divided by the scene analysis unit;
An apparatus, wherein at least part of the processing result of the intermediate processing unit is a second parameter among the configuration parameters. 4. A device according to claim 3 , wherein
The parameter generation unit is
The apparatus further comprising a fusion unit for generating a third parameter among the set parameters based on a processing result of the intermediate processing unit and a segmentation result of the image by the scene analysis unit. A video surveillance system,
comprising a camera, an analysis device and a device for setting parameters for a video surveillance system according to any one of claims 1 to 4 ,
a device for setting parameters for the video surveillance system, generating setting parameters based on the images obtained by the cameras, setting the setting parameters to the analysis device;
The video surveillance system, wherein the analysis device analyzes the images obtained by the camera based on the set parameters. A video surveillance system according to claim 5 , wherein
The video surveillance system comprises:
further comprising a detection device for detecting a change in state of the camera;
A device for setting parameters for the video surveillance system generates the setting parameters and sets the setting parameters to the analysis device when a change in the state of the camera is detected by the detection device. Monitoring system. A method of setting parameters for a video surveillance system, comprising:
generating configuration parameters based on images obtained by cameras of the video surveillance system; and setting the configuration parameters to the video surveillance system ;
Generating the configuration parameters includes selecting one of at least two models based on features of the image and dividing the image into different regions based on the selected model. ,
The image segmentation result is set as a first parameter in the setting parameters,
the model used for segmentation has different identification accuracies for different positions of the image;
The method of claim 1, wherein the segmentation is performed on at least two frames of the image, and segmentation results of the image are generated based on segmentation results on at least two frames of the image. 8. The method of claim 7 , wherein
Generating the configuration parameters includes:
The method further comprising generating a parameter different from the first parameter among the set parameters based on the image and a segmentation result of the image. 9. The method of claim 8 , wherein
Generating a parameter different from the first parameter in the configuration parameters includes:
In each divided region, performing intermediate processing corresponding to the region on the image,
A method, wherein at least part of the result of the intermediate processing is used as a second parameter among the setting parameters. 10. The method of claim 9 , wherein
Generating a parameter different from the first parameter in the configuration parameters includes:
The method further comprising generating a third parameter among the set parameters based on the result of the intermediate processing and the segmentation result of the image.

Images