Monitoring device acquisition method, device and computer-readable storage medium

Abstract

An embodiment of the present invention discloses a method, a device, and a computer-readable storage medium for obtaining a monitoring device. The method includes: determining a target image tag from an image tag set, where the image tag set includes a plurality of image tags, each of the image tags is associated with a monitoring device, and the monitoring image collected by the monitoring device contains the object represented by the image tag; determining a target monitoring device according to the target image tag. Through the present invention, the problem of low efficiency in obtaining a monitoring video for a specified location in the related art is solved, and the technical effect of improving the efficiency of obtaining a monitoring video for a specified location is achieved.

Description

Technical Field

[0001] The present invention relates to the field of image processing, and in particular, to a method, an apparatus, and a computer-readable storage medium for acquiring a monitoring device. Background Art

[0002] In real life, monitoring cameras have spread all over the streets and alleys, covering most of the buildings and streets in the city. However, when an event occurs at the entrance of Building A and it is necessary to mobilize the cameras around Building A to view the video, due to problems such as the orientation of the monitoring cameras, it is often necessary to check each monitoring camera near Building A one by one to determine whether the images captured by each monitoring camera cover the entrance of Building A.

[0003] Regarding the problem of low efficiency in acquiring monitoring videos for a specified location in the related art, no effective solution has been obtained yet. Summary of the Invention

[0004] To solve the above technical problems, embodiments of the present invention are expected to provide a method, an apparatus, and a computer-readable storage medium for acquiring a monitoring device, which can solve the problem of low efficiency in acquiring monitoring videos for a specified location in the related art.

[0005] Embodiments of the present invention provide a method for acquiring a monitoring device, including: determining a target image tag from an image tag set, where the image tag set includes a plurality of image tags, and each image tag is associated with a monitoring device, and the monitoring image collected by the monitoring device includes an object represented by the image tag; determining a target monitoring device according to the target image tag.

[0006] Optionally, the method further includes: respectively extracting a first content of a first image collected by the monitoring device and a second content of a second image of an area where the monitoring device is located, where the second image is determined from street view data in a geographic system; performing feature matching on the first content and the second content, and determining an image tag of the monitoring device according to a feature matching result.

[0007] Optionally, the performing feature matching on the first content and the second content, and determining an image tag of the monitoring device according to the feature matching result includes: when the first content and the second content match successfully, obtaining geographic information of the second content; determining an image tag of the monitoring device according to the geographic information of the second content.

[0008] Optionally, the feature matching of the first content and the second content, and determining the image tag of the monitoring device according to the feature matching result further includes: identifying the text information and / or identification information included in the first content in the first image to obtain a first identification result; adding the first identification result to the image tag; or comparing the first identification result with the geographical information of the area where the monitoring device is located to obtain the geographical information corresponding to the first identification result; adding the geographical information to the image tag.

[0009] Optionally, the feature matching of the first content and the second content, and determining the image tag of the monitoring device according to the feature matching result further includes: in the case where the first content and the second content do not match successfully, identifying the text information and / or identification information included in the first content in the first image to obtain a first identification result; adding the first identification result to the image tag; or comparing the first identification result with the geographical information of the area where the monitoring device is located to obtain the geographical information corresponding to the first identification result; adding the geographical information to the image tag.

[0010] Optionally, the method further includes: obtaining the position of the object represented by the image tag in the first image; and displaying the image tag at the position in the image collected by the monitoring device.

[0011] An embodiment of the present invention further provides a monitoring device acquisition device, which is characterized in that it includes: a first acquisition module, configured to determine a target image tag from an image tag set, where the image tag set includes a plurality of image tags, each image tag is associated with a monitoring device, and the monitoring image collected by the monitoring device includes the object represented by the image tag; a second acquisition module, configured to determine a target monitoring device according to the target image tag.

[0012] Optionally, the device further includes: an extraction module, configured to extract the first content of the first image collected by the monitoring device and the second content of the second image of the area where the monitoring device is located respectively, where the second image is determined by the street view data in the geographic system; a third acquisition module, configured to perform feature matching on the first content and the second content, and determine an image tag of the monitoring device according to the feature matching result; the third acquisition module includes: a first acquisition unit, configured to acquire the geographic information of the second content when the first content and the second content match successfully; a second acquisition unit, configured to determine the image tag of the monitoring device according to the geographic information of the second content. The third acquisition module includes: a third acquisition unit, configured to identify the text information and / or identification information included in the first content in the first image, and obtain a first identification result; a first addition unit, configured to add the first identification result to the image tag; or, a fourth acquisition unit, configured to compare the first identification result with the geographic information of the area where the monitoring device is located, and obtain the geographic information corresponding to the first identification result; a second addition unit, configured to add the geographic information to the image tag; or,

[0013] The third acquisition module includes: a fifth acquisition unit, configured to identify the text information and / or identification information included in the first content in the first image, and obtain a first identification result when the first content and the second content do not match successfully; a third addition unit, configured to add the first identification result to the image tag; or, a sixth acquisition unit, configured to compare the first identification result with the geographic information of the area where the monitoring device is located, and obtain the geographic information corresponding to the first identification result; a fourth addition unit, configured to add the geographic information to the image tag;

[0014] The device further includes: a fourth acquisition module, configured to acquire the position of the object represented by the image tag in the first image; a display module, configured to display the image tag at the position in the image collected by the monitoring device.

[0015] An embodiment of the present invention provides an electronic device, including: a processor; a memory for storing processor-executable instructions; wherein, the processor is configured to execute the steps of any one of the above methods.

[0016] An embodiment of the present invention provides a computer-readable storage medium, on which instructions are stored, and when the instructions are executed by a processor, the steps of any one of the above methods are implemented.

[0017] Embodiments of the present invention provide a method, apparatus, electronic device, and computer-readable storage medium for acquiring a monitoring device, which determine a target image tag from an image tag set. The image tag set includes a plurality of image tags, and each image tag is associated with a monitoring device. The monitoring image collected by the monitoring device contains the object represented by the image tag. According to the target image tag, a target monitoring device is determined. That is to say, embodiments of the present invention can quickly locate a monitoring device through an image tag, and then acquire the monitoring image collected by the monitoring device, without viewing one by one the monitoring images collected by all monitoring devices in the target monitoring device set, thus solving the problem of low efficiency in obtaining a monitoring video for a specified location in the related art and achieving the technical effect of improving the efficiency of obtaining a monitoring video for a specified location. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings described herein are used to provide a further understanding of the present invention, form a part of this application, and the illustrative embodiments and descriptions thereof of the present invention are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0019] Figure 1 is a schematic flowchart of a method for acquiring a monitoring device provided by an embodiment of the present invention;

[0020] Figure 2 is a schematic diagram of an apparatus for acquiring a monitoring device provided by an embodiment of the present invention;

[0021] Figure 3 is a schematic diagram of another apparatus for acquiring a monitoring device provided by an embodiment of the present invention;

[0022] Figure 4 is a schematic diagram of yet another apparatus for acquiring a monitoring device provided by an embodiment of the present invention;

[0023] Figure 5 is a schematic diagram of still another apparatus for acquiring a monitoring device provided by an embodiment of the present invention;

[0024] Figure 6 is a schematic diagram of a system for acquiring a monitoring device provided by an embodiment of the present invention;

[0025] Figures 7a - 7c is a schematic diagram of a monitoring image provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0026] To enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0027] It should be noted that the terms "first", "second", etc. in the description, claims and drawings of the present invention are used to distinguish different objects, rather than to limit a specific order.

[0028] The embodiment of the present invention provides a method for obtaining a monitoring device. Figure 1 As shown in the flowchart of a method for obtaining a monitoring device provided by the embodiment of the present invention, Figure 1 The method for obtaining a monitoring device provided by the embodiment of the present application includes the following steps:

[0029] S102. Determine a target image tag from the image tag set, where the image tag set includes multiple image tags, and each image tag is associated with a monitoring device, and the monitoring image collected by the monitoring device contains the object represented by the image tag.

[0030] It should be noted that in the embodiment of the present application, determining the target image tag from the image tag set may include various implementation manners. For example, the user may directly input or select the target image tag (such as Starbucks in Hangzhou Jiguang Center), or the user may input keywords and retrieve the target image tag from the image tag set, or the user may select Starbucks in Jiguang Center on the map, and the system automatically obtains the geographical information and automatically matches the associated target image tag.

[0031] The above objects include but are not limited to: buildings, landmarks, etc. in the monitoring image.

[0032] In addition, it should be noted that the above image tag may be composed of one or more of the following information: landmark objects, building names, store names, school names, text information, icons, etc., and no limitation is made here. For example, the above image tag may be Building A, Store B in Building A, Store D in Square C, and School D on a certain road section, etc.

[0033] S104. Determine the target monitoring device according to the target image tag.

[0034] Optionally, in this embodiment of the present application, the specific implementation manner of the above S104 may include: determining in which section of the monitored image the target image tag is marked, and determining which monitoring device captured the monitored image (that is, determining the correspondence between the target image tag, the monitored video, and the monitoring device), and then screening out the target monitoring device.

[0035] For example, assume that the target image tag is "Shop B in Building A", and there are 3 monitoring devices in the target monitoring device set, namely monitoring device 1, monitoring device 2, and monitoring device 3. The image tags of monitoring device 1 include "Shop C" and "School D", the image tags of monitoring device 2 include "Shop B in Building A", and the image tags of monitoring device 3 include "Building A". Then, based on the target image tag "Shop B in Building A", monitoring device 2 can be screened out as the target monitoring device.

[0036] Another example, assume that the target image tag is "Shop B in Building A", and there are 3 monitoring devices in the target monitoring device set, namely monitoring device 1, monitoring device 2, and monitoring device 3. The image tags of monitoring device 1 include "Shop C", "School D", and "Shop B in Building A", the image tags of monitoring device 2 include "Shop B in Building A", and the image tags of monitoring device 3 include "Building A". Then, based on the target image tag "Shop B in Building A", finally, monitoring device 1 and monitoring device 2 can be screened out as the target monitoring devices.

[0037] Through the above steps S102 to S104, the target image tag is determined from the image tag set. Among them, the image tag set includes multiple image tags, and each image tag is associated with a monitoring device, and the monitored image collected by the monitoring device contains the object represented by the image tag; based on the target image tag, the target monitoring device is determined. That is to say, the embodiment of the present invention can quickly locate the monitoring device through the image tag, and then obtain the monitored image collected by the monitoring device, without viewing one by one the monitored images collected by all the monitoring devices in the target monitoring device set, thereby solving the problem of low efficiency in obtaining the monitored video for a specified location in the related art, and achieving the technical effect of improving the efficiency of obtaining the monitored video for a specified location.

[0038] In an optional implementation manner, the above method further includes:

[0039] S11, respectively extract the first content of the first image collected by the monitoring device and the second content of the second image of the area where the monitoring device is located, where the second image is determined by the street view data in the geographic system;

[0040] Optionally, the acquisition method of the second image in the above S11 may include the following steps:

[0041] S111. Obtain the longitude and latitude information of the monitoring device;

[0042] S112. With the position indicated by the longitude and latitude information as the center, determine a second image of the area where the monitoring device is located from the geographic system.

[0043] Through the above steps S111 - S112, an image of the area where the monitoring device is located in the geographic system can be effectively obtained.

[0044] Optionally, in the embodiments of the present application, a deep learning algorithm can be used to model the first content in the above first image and the second content in the second image. Among them, the first content and the second content include, but are not limited to, buildings, shops, communities, schools, rivers, etc.

[0045] In addition, it should be noted that the above geographic system includes, but is not limited to: Baidu, Amap, Google, etc. Among them, the above street view data can also be album data uploaded by users to the above geographic system.

[0046] S12. Perform feature matching on the first content and the second content, and determine an image tag for the monitoring device according to the feature matching result.

[0047] Among them, the implementation manner of the above S12 can include the following steps:

[0048] S121. When the first content and the second content match successfully, obtain the geographic information of the second content;

[0049] S122. Determine an image tag for the monitoring device according to the geographic information of the second content.

[0050] For example, if the monitoring image of Building A matches successfully with Building XX, obtain the geographic information corresponding to Building XX in the street view image, such as Hangzhou Juguang Center, and use the Hangzhou Juguang Center as the image tag.

[0051] Or, the implementation manner of the above S12 can also include the following steps:

[0052] S123. Identify the text information and / or identification information included in the first content in the first image to obtain a first recognition result;

[0053] S124. Add the first recognition result to the image tag; or,

[0054] S125. Compare the first recognition result with the geographic information of the area where the monitoring device is located to obtain the geographic information corresponding to the first recognition result; add the geographic information to the image tag.

[0055] Optionally, in this embodiment, the reason why the first recognition result is not directly added to the image label in S125 is to solve the problem of inaccurate image labels caused by inaccurate recognition results in S123, especially for icon recognition. Of course, when the recognition result in S123 is accurate, the first recognition result can also be directly added to the image label.

[0056] Alternatively, the implementation manner of the above S12 may further include the following steps:

[0057] S126, in the case where the first content and the second content do not match successfully, recognize the text information and / or identification information included in the first content in the first image to obtain a first recognition result; add the first recognition result to the image label; or,

[0058] S127, compare the first recognition result with the geographical information of the area where the monitoring device is located to obtain the geographical information corresponding to the first recognition result; add the geographical information to the image label.

[0059] For example, if the monitoring image of Building A and the street view image of Building XX do not match successfully, then it is necessary to further recognize the text information and / or identification information included in Building A (for example, recognize the text information as "Xingsheng Pharmacy"), and then add "Xingsheng Pharmacy" to the image label, or compare "Xingsheng Pharmacy" with the geographical information of the area where the monitoring device is located to obtain the geographical information as "Wuzhou International Square", and add "Wuzhou International Square" to the image label.

[0060] In an optional implementation manner, the above method further includes:

[0061] S21, obtain the position of the object represented by the image label in the first image;

[0062] S22, display the image label at this position in the image collected by the monitoring device.

[0063] Through the above S21 - S22, it is possible to display the image label of the collected image in the video stream, and further locate the monitoring image and the monitoring device that collected the monitoring image through the image label.

[0064] In an optional implementation manner, extracting the second content of the second image of the area where the monitoring device is located includes:

[0065] S31, in the case where the street view data is updated, obtain the update list of the street view data, where the update list includes historical street view data and current street view data;

[0066] S32, determine the second image through the update list.

[0067] Through the above S31 - S32, when updating street view data, a street view data update list is established. When retrieving monitoring devices, the historical data of street view data in the update list is also used to retrieve tags, further improving the accuracy of the monitoring devices.

[0068] Optionally, if different time periods correspond to different image tag sets, the time period can also be set as part of the image tag. For example, some spherical cameras can adjust their monitoring directions, so the spherical camera corresponds to different image tag sets in different time periods. Therefore, when retrieving tags, the retrieved time period can also be combined to determine the monitoring device, which can further improve the retrieval efficiency of the monitoring device.

[0069] Taking buildings and shops in the monitoring images as examples below, the embodiments of the present application will be illustrated.

[0070] Based on the street view map (corresponding to street view data), geographic information data, and the longitude and latitude of the monitoring device, this example uses an image analysis and modeling algorithm to establish a method for calibrating tags of monitoring devices based on the content of the picture. The core is to compare the buildings and shops in the monitoring image with the street view map, and combine the recognition of the shop names, building names, or logos that appear in the monitoring image to calibrate the tags of the monitoring image, which is beneficial for directly and quickly retrieving the monitoring image. The main implementation process of this method includes:

[0071] S41, Obtain the longitude and latitude data of the monitoring device, and retrieve geographic information systems such as Baidu, Amap, and Google to obtain the geographic information data of buildings, shops, etc. and their corresponding street view data around the longitude and latitude.

[0072] S42, Obtain the picture of the first frame stream of the monitoring device, model the buildings, shops, etc. (the first content) in the picture; taking the longitude and latitude of the monitoring device as the center, model the buildings, shops, etc. (the second content) in the surrounding street view map, and based on image analysis technology, compare the models in the picture pairwise, and establish a corresponding relationship between the buildings and shops in the monitoring image and the buildings and shops in the street view map according to the rule of the highest similarity. For example, monitoring map building - street view map XX Building, monitoring map shop - street view map XXX Supermarket.

[0073] S43, Identify the shop names, building names, logos, etc. that appear in the monitoring screen. For example, identify that the shop in S42 is Starbucks and further compare it with the geographic information data near the longitude and latitude of the monitoring device.

[0074] S44. Combining the analysis results of S42 and S43 above, combine the stores, buildings and geographical information data that appear in the surveillance images. For example, for the above-mentioned Starbucks store, the comparison result between the building it belongs to (the first content) and the street view map is Hangzhou Juguang Center (the feature comparison result), then the surveillance image label is calibrated as "Starbucks in Hangzhou Juguang Center".

[0075] In summary, the embodiment of the present invention does not require manual labeling, and can automatically extract various feature identifiers from the surveillance images and the street views of the geographical system through an intelligent recognition algorithm, so as to label each building and store in the surveillance images and display them on the screen. It can also quickly retrieve the positions of the surveillance devices through the labels.

[0076] The embodiment of the present invention also provides a surveillance device acquisition device. Figure 2 As shown in the schematic diagram of a surveillance device acquisition device provided by the embodiment of the present invention, Figure 2 The surveillance device acquisition device provided by the embodiment of the present application includes:

[0077] A first acquisition module 22, configured to determine a target image label from an image label set, where the image label set includes a plurality of image labels, and each of the image labels is associated with a surveillance device, and the surveillance image collected by the surveillance device includes the object represented by the image label;

[0078] It should be noted that in the embodiment of the present application, determining the target image label from the image label set may include various implementation manners. For example, the user may directly input or select the target image label (such as "Starbucks in Hangzhou Juguang Center"), or the user may input keywords and retrieve the target image label in the image label set, or the user may select "Starbucks in Hangzhou Juguang Center" on the map, and the system automatically obtains the geographical information and automatically matches the associated target image label.

[0079] The above objects include but are not limited to: buildings, landmarks, etc. in the surveillance images.

[0080] In addition, it should be noted that the above image label may be composed of one or more of the following information: landmark objects, building names, store names, school names, text information, icons, etc., and no limitation is made here. For example, the above image label may be Building A, Store B in Building A, Store D in Square C, and School D on a certain road section, etc.

[0081] A second acquisition module 24, configured to determine a target surveillance device according to the target image label.

[0082] Optionally, in this embodiment of the present application, the above-mentioned second acquisition module 24 may be used to determine in which segment of the monitored image the target image tag is marked, and determine which monitoring device captured the monitored image (that is, determine the correspondence between the target image tag, the monitored video, and the monitoring device), and then filter out the target monitoring device.

[0083] For example, assume that the target image tag is "Store B in Building A", and there are 3 monitoring devices in the target monitoring device set, namely monitoring device 1, monitoring device 2, and monitoring device 3. The image tags of monitoring device 1 include "Store C" and "School D", the image tags of monitoring device 2 include "Store B in Building A", and the image tags of monitoring device 3 include "Building A". Then, based on the target image tag "Store B in Building A", monitoring device 2 can be filtered out as the target monitoring device.

[0084] Another example, assume that the target image tag is "Store B in Building A", and there are 3 monitoring devices in the target monitoring device set, namely monitoring device 1, monitoring device 2, and monitoring device 3. The image tags of monitoring device 1 include "Store C", "School D", and "Store B in Building A", the image tags of monitoring device 2 include "Store B in Building A", and the image tags of monitoring device 3 include "Building A". Then, based on the target image tag "Store B in Building A", finally, monitoring device 1 and monitoring device 2 can be filtered out as the target monitoring devices.

[0085] Through Figure 2 the device shown, the target image tag is determined from the image tag set, where the image tag set includes multiple image tags, and each such image tag is associated with a monitoring device, and the monitored image captured by the monitoring device contains the object represented by the image tag; based on the target image tag, the target monitoring device is determined. That is to say, the embodiment of the present invention can quickly locate the monitoring device through the image tag, and then obtain the monitored image captured by the monitoring device, without viewing one by one the monitored images captured by all the monitoring devices in the target monitoring device set, thereby solving the problem of low efficiency in obtaining the monitored video for a specified location in the related art, and achieving the technical effect of improving the efficiency of obtaining the monitored video for a specified location.

[0086] Optionally, as Figure 3 shown, the above-mentioned device further includes:

[0087] An extraction module 32, configured to extract the first content of the first image captured by the monitoring device and the second content of the second image of the area where the monitoring device is located, where the second image is determined from the street view data in the geographic system;

[0088] Optionally, the above extraction module 32 is further configured to obtain the longitude and latitude information of the monitoring device; and determine a second image of the area where the monitoring device is located from the geographic system with the position indicated by the longitude and latitude information as the center.

[0089] Through the above extraction module 32, the image of the area where the monitoring device is located in the geographic system can be effectively obtained.

[0090] Optionally, in the embodiments of the present application, a deep learning algorithm may be used to model the first content in the first image and the second content in the second image, where the first content and the second content include, but are not limited to, buildings, shops, communities, schools, rivers, etc.

[0091] In addition, it should be noted that the above geographic system includes, but is not limited to: Baidu, Amap, Google, etc. Among them, the above street view data may also be album data uploaded by users to the above geographic system.

[0092] The third acquisition module 34 is configured to perform feature matching on the first content and the second content, and determine an image label of the monitoring device according to the feature matching result.

[0093] Optionally, as Figure 4 shown, the above third acquisition module 34 includes:

[0094] The first acquisition unit 42 is configured to obtain the geographic information of the second content when the first content and the second content match successfully; the second acquisition unit 44 is configured to determine an image label of the monitoring device according to the geographic information of the second content.

[0095] For example, if the monitoring image of Building A matches Building XX successfully, then obtain the geographic information corresponding to Building XX in the street view image, such as Hangzhou Jiguang Center, and use the Hangzhou Jiguang Center as the image label.

[0096] Optionally, the following units of the third acquisition module may also be used for Figure 4 equivalent replacement. Among them, the third acquisition module includes: a third acquisition unit configured to identify the text information and / or identification information included in the first content in the first image to obtain a first identification result; a first addition unit configured to add the first identification result to the image label; or, a fourth acquisition unit configured to compare the first identification result with the geographic information of the area where the monitoring device is located to obtain the geographic information corresponding to the first identification result; a second addition unit configured to add the geographic information to the image label.

[0097] Optionally, in this embodiment, the reason for not directly adding the first recognition result to the image label in the above manner is to solve the problem of inaccurate image labels caused by inaccurate recognition results, especially for icon recognition. Of course, when the recognition result is accurate, the first recognition result can also be directly added to the image label.

[0098] Alternatively, the third acquisition module includes: a fifth acquisition unit, configured to, when the first content and the second content do not match successfully, recognize the text information and / or identification information included in the first content in the first image to obtain a first recognition result; a third addition unit, configured to add the first recognition result to the image label; or, a sixth acquisition unit, configured to compare the first recognition result with the geographical information of the area where the monitoring device is located to obtain the geographical information corresponding to the first recognition result; a fourth addition unit, configured to add the geographical information to the image label.

[0099] For example, if the monitoring image of Building A and the street view image of Building XX do not match successfully, then it is necessary to further recognize the text information and / or identification information included in Building A (for example, recognize that the text information is "Xingsheng Pharmacy"), and then add "Xingsheng Pharmacy" to the image label, or compare the "Xingsheng Pharmacy" with the geographical information of the area where the monitoring device is located to obtain the geographical information as "Wuzhou International Square", and add the "Wuzhou International Square" to the image label.

[0100] In an alternative embodiment, as Figure 5 shown, the above device further includes:

[0101] A fourth acquisition module 52, configured to acquire the position of the object represented by the image label in the first image;

[0102] A display module 54, configured to display the image label at the position in the image acquired by the monitoring device.

[0103] By Figure 5 the above device, it is possible to display the image label of the acquired image in the video stream, and further locate the monitoring image and the monitoring device that acquired the monitoring image through the image label.

[0104] In an alternative embodiment, the above extraction module 32 is further configured to, when the street view data is updated, acquire an update list of the street view data, where the update list includes historical street view data and current street view data; and determine the second image through the update list.

[0105] Through the above extraction module 32, when updating the street view data, a street view data update list is established. When retrieving monitoring devices, the tags are retrieved based on the historical data of the street view data in the update list, further improving the accuracy of the monitoring devices.

[0106] Optionally, if different time periods correspond to different image tag sets, the time period can also be set as a part of the image tag. For example, some spherical cameras can adjust their monitoring directions. Then, this spherical camera corresponds to different image tag sets in different time periods. Therefore, when retrieving tags, the retrieved time period can also be combined to determine the monitoring device, which can further improve the retrieval efficiency of the monitoring device.

[0107] The embodiment of the present invention also provides a monitoring device acquisition system. Figure 6 As shown in the schematic diagram of a monitoring device acquisition system provided by the embodiment of the present invention, Figure 6 The monitoring device acquisition system provided by the embodiment of the present application includes: a camera 62 and an image calibration processing device 64 in the monitoring device. Among them, the image calibration processing device 64 includes: an input / output module 641, a data support module 642, an image analysis module 643, and a tag positioning module 644; the input / output module 641 is used to acquire the image information captured by the camera in the monitoring device; the data support module 642 is used to acquire the street view data of the monitoring device and the location information of the monitoring device; the image analysis module 643 is used to analyze specific buildings in the street view data and the image information through modeling, obtain the correspondence between the specific building in the image information (corresponding to the first content) and the specific building in the street view data (corresponding to the second content), and identify the specific building in the image information based on this correspondence to obtain an identification result; the tag positioning module 644 uses the identification result of the image analysis module 643 and the street view data and the location information of the monitoring device in the data support module 642 to generate an image tag, so as to determine the corresponding monitoring device based on the image tag.

[0108] Optionally, the input / output module 641 is further used to acquire the real-time image information of the camera irradiation direction. Among them, when the pan-tilt controls the camera to rotate, the image information captured by the rotated camera is acquired, and the image information captured before the camera rotates is updated according to the image information.

[0109] Specifically, as Figure 6 shown, the image calibration processing device 64 is composed of an input / output module 641, a data support module 642, an image analysis module 643, and a tag positioning module 644, denoted as the input / output module (P), the data support module (B), the image analysis module (A), and the tag positioning module (S).

[0110] Among them, the input / output module (P) is used for interaction. Using this module, the input of the bitstream picture of the monitoring device position (i.e., the image information obtained from the monitoring device in the embodiments of the present application) is completed. If the video picture changes after the monitoring device is controlled by the pan-tilt, new bitstream pictures are iterated in time as input pictures to complete tasks such as output of label calibration results.

[0111] The data support module (B) includes: a geographic information (G) and monitoring data (D) part. Among them, the geographic information (G) is used to obtain geographic information system data, including but not limited to geographic information such as buildings, shops, communities, rivers, schools, etc. around the monitoring device and the corresponding street view data. The monitoring data (D) is used to obtain the longitude and latitude information of the monitoring device position, the irradiation direction of the monitoring probe (if any), and obtain the real-time bitstream of the monitoring device position.

[0112] The image analysis module (A) includes: an image modeling (M), recognition analysis (R), and image comparison (C) part. The image modeling (M) is used to model objects in the first-frame bitstream picture using deep learning algorithms, including but not limited to buildings, shops, communities, schools, rivers, etc. For example, a building and a shop are obtained; similarly, taking the longitude and latitude of the monitoring device position as the center, objects in the street view map image are modeled using deep learning algorithms. The recognition analysis (R) is used to recognize the detailed features that appear in the monitoring screen, including but not limited to the names, logos, or other text information of buildings, shops, communities, schools, etc. For example, when the monitoring image captures the storefront of Luckin Coffee, the brand logo of Luckin Coffee is recognized, and the store is recorded as "Luckin Coffee". The image comparison (C) is used to pairwise compare the bitstream picture uploaded by the monitoring device position with the picture modeling results in the street view map based on the results of the image modeling (M), and select the matching object with the highest similarity to form the corresponding relationship between the monitoring image and the geographic data. For example, if the similarity between Building A in the monitoring image and Building XX in the street view map is the highest, then the corresponding relationship of Building A - Building XX is established.

[0113] For example, as Figures 7a - 7c shown, Figures 7a - 7c is a schematic diagram of a monitoring image provided by an embodiment of the present invention. A monitoring device position is set on the auxiliary road of Yuhangtang Road, Xihu District, Hangzhou, and the longitude and latitude are (120.10, 30.30) respectively. According to its longitude and latitude, it is compared with the map geographic information data, and the nearby shops and buildings include: Industrial Bank, Zhishang Zhenxuan Hotel, Wuzhou International Plaza, Wuzhou International Apartment, etc. Analyze a certain shop in the monitoring image, and the monitoring image is as Figures 7a - 7cAccording to image modeling and comparison with the surrounding street view map modeling, the building where it is located has the highest similarity of 93% with the "Wuzhou International Plaza" in the street view map. Then, it is considered that the building where the store is located is the Wuzhou International Plaza. Identify and analyze the surveillance image, as shown in Figure 7c the mark, and it is concluded that the name of the store is "Xingsheng Pharmacy". Then, mark the store as "Wuzhou International Plaza Xingsheng Pharmacy".

[0114] The label positioning module (S) is used to record the information in the surveillance image based on the comprehensive result of the image analysis module. For example, if the store captured in the surveillance image is identified as "Luckin Coffee" and its affiliated building is determined to be "Hangzhou Jiguang Center" through comparison, then label the store as "Jiguang Center Luckin Coffee". The server displays the corresponding geographical data label name "Jiguang Center Luckin Coffee" at the same position in the surveillance video according to the coordinates of the original store in the surveillance picture, and can quickly locate the corresponding surveillance device based on the label and filter out the surveillance images containing this mark.

[0115] The embodiment of the present invention also provides an electronic device, including: a processor; a memory for storing instructions executable by the processor; wherein, the processor is configured to execute the steps of any one of the above methods.

[0116] The embodiment of the present invention also provides a computer-readable storage medium, on which instructions are stored, and when the instructions are executed by the processor, the steps of any one of the above methods are implemented.

[0117] The above is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention.

Claims

1. A method for obtaining a monitoring device, characterized in that, Including: Determine a target image tag from an image tag set, where the image tag set includes multiple image tags, and each image tag is associated with a monitoring device, and the monitoring image collected by the monitoring device includes the object represented by the image tag; the object includes at least one of the following: buildings and markers in the monitoring image; Determine a target monitoring device according to the target image tag; wherein, the image tag is determined by the following method: Extract the first content of the first image collected by the monitoring device and the second content of the second image of the area where the monitoring device is located respectively, wherein the second image is determined by the street view data in the geographic system; perform feature matching on the first content and the second content; When the first content and the second content match successfully, obtain the geographic information of the second content; Determine the image tag of the monitoring device according to the geographic information of the second content; When the first content and the second content do not match successfully, identify the text information and / or identification information included in the first content in the first image to obtain a first recognition result; add the first recognition result to the image tag; or, compare the first recognition result with the geographic information of the area where the monitoring device is located to obtain the geographic information corresponding to the first recognition result; add the geographic information to the image tag; Wherein, the geographic information is the name corresponding to the second content or the first recognition result; Obtain the position of the object represented by the image tag in the first image; Display the image tag at the position in the image collected by the monitoring device.

2. The method according to claim 1, characterized in that, When the first content and the second content match successfully, the method further includes: Identify the text information and / or identification information included in the first content in the first image to obtain a first recognition result; Add the first recognition result to the image tag; or, Compare the first recognition result with the geographic information of the area where the monitoring device is located to obtain the geographic information corresponding to the first recognition result; add the geographic information to the image tag.

3. A monitoring device acquisition device, characterized in that, Including: A first acquisition module, configured to determine a target image tag from an image tag set, where the image tag set includes multiple image tags, and each image tag is associated with a monitoring device, and the monitoring image collected by the monitoring device includes the object represented by the image tag; the object includes at least one of the following: buildings and markers in the monitoring image; A second acquisition module, configured to determine a target monitoring device according to the target image tag; An extraction module, configured to extract the first content of the first image collected by the monitoring device and the second content of the second image of the area where the monitoring device is located respectively, wherein the second image is determined by the street view data in the geographic system; A third acquisition module, configured to perform feature matching on the first content and the second content; The third acquisition module includes: A first acquisition unit, configured to acquire the geographical information of the second content when the first content and the second content match successfully; a second acquisition unit, configured to determine an image tag of the monitoring device according to the geographical information of the second content; or, The third acquisition module includes: a fifth acquisition unit, configured to identify the text information and / or identification information included in the first content in the first image and obtain a first identification result when the first content and the second content do not match successfully; a third addition unit, configured to add the first identification result to the image tag; or, a sixth acquisition unit, configured to compare the first identification result with the geographical information of the area where the monitoring device is located to obtain the geographical information corresponding to the first identification result; a fourth addition unit, configured to add the geographical information to the image tag; wherein, the geographical information is a name corresponding to the second content or the first identification result; A fourth acquisition module, configured to acquire the position of the object represented by the image tag in the first image; A display module, configured to display the image tag at the position in the image acquired by the monitoring device.

4. The apparatus according to claim 3, wherein The third acquisition module includes: a third acquisition unit, configured to identify the text information and / or identification information included in the first content in the first image and obtain a first identification result; a first addition unit, configured to add the first identification result to the image tag; or, a fourth acquisition unit, configured to compare the first identification result with the geographical information of the area where the monitoring device is located to obtain the geographical information corresponding to the first identification result; a second addition unit, configured to add the geographical information to the image tag.

5. An electronic device, characterized in that, Comprising: A processor; A memory for storing instructions executable by the processor; Wherein, the processor is configured to execute the steps of any one of the methods according to claims 1-2.

6. A computer-readable storage medium having instructions stored thereon, characterized in that, When the instructions are executed by the processor, the steps of any one of the methods according to claims 1-2 are implemented.

Images