Security monitoring method and device

Abstract

The invention relates to a security monitoring method and device. The method comprises the following steps: acquiring an environment monitoring image, determining at least one monitoring area in the environment monitoring image, verifying a target object detected in each monitoring area according to a preset monitoring type, and adjusting a monitoring strategy of the monitoring area according to a verification result, and monitoring the target object of each preset monitoring type according to the adjusted monitoring strategy. In the method, the target object detected in the monitoring area can be verified based on the preset monitoring type, so that the monitoring strategy corresponding to the monitoring area is adjusted according to the verification result, respective monitoring strategies are made for different monitoring areas, fine-grained division of the monitoring strategies is realized, and the accuracy of the monitoring strategies is improved. The monitoring flexibility is improved, and the monitoring accuracy is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a security monitoring method and device. Background Art

[0002] With the continuous development of science and technology, electronic security technology applied in various fields has developed rapidly.

[0003] Taking the security monitoring system used indoors and outdoors as an example, the cameras installed in the security monitoring system can collect environmental images within the coverage area and identify preset types of monitoring objects to provide early warning functions.

[0004] However, the security monitoring system in the related art has low monitoring accuracy. 。 Summary of the invention

[0005] Based on this, a security monitoring method and device are provided to improve the accuracy of security monitoring.

[0006] In a first aspect, the present application provides a security monitoring method, comprising:

[0007] Acquire an environmental monitoring image, and determine at least one monitoring area in the environmental monitoring image;

[0008] According to the preset monitoring type, verify the target objects detected in each monitoring area;

[0009] Adjust the monitoring strategy of the monitoring area based on the verification results;

[0010] The target objects of each preset monitoring type are monitored according to the adjusted monitoring strategy.

[0011] In one embodiment, adjusting the monitoring strategy of the monitoring area according to the verification result includes:

[0012] Determine the content of policy adjustments based on the verification results;

[0013] Adjust the monitoring strategy of the monitoring area according to the strategy adjustment content.

[0014] In one embodiment, adjusting the monitoring strategy of the monitoring area according to the strategy adjustment content includes:

[0015] The monitoring area is sent to the user end, and the operation instruction of the user end for the target object detected in the monitoring area is received; the operation instruction is used to represent the content of the strategy adjustment;

[0016] Adjust the monitoring strategy of the monitoring area according to the operating instructions.

[0017] In one embodiment, the monitoring strategy includes a preset monitoring type and a confidence threshold corresponding to the preset monitoring type; adjusting the monitoring strategy of the monitoring area according to the operation instruction includes:

[0018] If the operation instruction is a confidence adjustment instruction, the first object type indicated by the confidence adjustment instruction and the adjusted confidence carried by the confidence adjustment instruction are obtained, and the confidence threshold corresponding to the first object type in the monitoring strategy is replaced by the adjusted confidence;

[0019] If the operation instruction is an instruction to abandon monitoring, the second object type indicated in the instruction to abandon monitoring is obtained, and the second object type in the monitoring strategy is deleted.

[0020] In one embodiment, determining at least one monitoring area in an environmental monitoring image includes:

[0021] For the monitoring images in the environmental monitoring images, the target detection model is used to detect the target objects of the preset monitoring type in the monitoring images to obtain the target distribution map of the monitoring images;

[0022] The monitoring area is determined based on the target distribution map of the monitoring image in the environmental monitoring image.

[0023] In one embodiment, determining the monitoring area according to the target distribution map of each monitoring image in the environmental monitoring image includes:

[0024] The target distribution map of each monitoring image in the environmental monitoring image is divided into regions using a preset division template;

[0025] A monitoring area is determined from the multiple candidate areas obtained by division.

[0026] In one embodiment, determining a monitoring area from the multiple candidate areas obtained by division includes:

[0027] For each target distribution map, obtaining detection information of the target object detected in each candidate area;

[0028] According to the detection information of the target objects in the candidate areas in each target distribution map, a monitoring area is determined among the multiple candidate areas.

[0029] In one embodiment, determining a monitoring area from among multiple candidate areas based on detection information of target objects in the candidate areas in each target distribution map includes:

[0030] For each target distribution map, obtain the location and type of the target object in each candidate area;

[0031] A monitoring area is determined among the multiple candidate areas according to the location and type of the target object in each candidate area.

[0032] In one embodiment, determining a monitoring area from among multiple candidate areas according to the location and type of the target object in each candidate area includes:

[0033] Compare the locations and types of target objects in the same candidate area between target distribution maps;

[0034] The same candidate area in which the position and type of the target object in the target distribution map of a preset number of consecutive frames are matched is determined as the monitoring area.

[0035] In a second aspect, the present application also provides a security monitoring device, comprising:

[0036] An area determination module is used to obtain an environmental monitoring image and determine at least one monitoring area in the environmental monitoring image;

[0037] The target verification module is used to verify the target objects detected in each monitoring area according to the preset monitoring type;

[0038] A strategy adjustment module is used to adjust the monitoring strategy of the monitoring area according to the verification results;

[0039] The monitoring execution module is used to monitor the target objects of each preset monitoring type according to the adjusted monitoring strategy.

[0040] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any of the above methods when executing the computer program.

[0041] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of any of the above methods when the computer program is executed by a processor.

[0042] In a fifth aspect, the present application also provides a computer program product, including a computer program, which implements the steps of any of the above methods when executed by a processor.

[0043] In the above security monitoring method and device, by acquiring environmental monitoring images and determining at least one monitoring area in the environmental monitoring images, the target objects detected in each monitoring area are verified according to the preset monitoring type, and the monitoring strategy of the monitoring area is adjusted according to the verification result, and then the target objects of each preset monitoring type are monitored according to the adjusted monitoring strategy. In the above method, the target objects detected in the monitoring area can be verified based on the preset monitoring type, and the monitoring strategy of the corresponding monitoring area can be adjusted according to the verification result, so that different monitoring strategies can be formulated for different monitoring areas, thereby realizing the fine-grained division of monitoring strategies, and improving the monitoring flexibility while improving the monitoring accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is an internal structure diagram of a computer device in one embodiment;

[0045] Figure 2 A schematic diagram of a process flow of a security monitoring method in an embodiment;

[0046] Figure 3 A schematic diagram of a process for adjusting a monitoring strategy in an embodiment;

[0047] Figure 4 A schematic diagram of a flow chart for adjusting a monitoring strategy in another embodiment;

[0048] Figure 5 A schematic diagram of a flow chart for adjusting a monitoring strategy in another embodiment;

[0049] Figure 6 A schematic diagram of a process for determining a monitoring area in an embodiment;

[0050] Figure 7 is a schematic diagram of a target distribution diagram in one embodiment;

[0051] Figure 8 A schematic diagram of a process for determining a monitoring area in another embodiment;

[0052] Fig. 9 A schematic diagram of a process for determining a monitoring area according to detection information in one embodiment;

[0053] Fig.10 A schematic diagram of a process for determining a monitoring area according to the location and type of a target object in one embodiment;

[0054] Fig.11 A schematic diagram of a flow chart of determining a monitoring area according to the location and type of a target object in another embodiment;

[0055] Fig.12 is a schematic diagram of a target distribution diagram in another embodiment;

[0056] Fig.13 FIG. 4 is a structural block diagram of a security monitoring device in an embodiment. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0058] The security monitoring method provided in the embodiment of the present application can be applied to Figure 1In the computer device shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a security monitoring method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0059] Those skilled in the art will understand that Figure 1 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0060] In one embodiment, Figure 2 As shown, a security monitoring method is provided, which is applied to Figure 1 The computer device in the example is used to illustrate, including the following steps:

[0061] S210: Acquire an environmental monitoring image, and determine at least one monitoring area in the environmental monitoring image.

[0062] The monitoring area is an image area in the environment monitoring image, which may be one or more. The monitoring area may be an image area at a preset position or an image area determined based on the target detection result.

[0063] Optionally, the computer device may communicate with the camera, or the camera may be integrated into the computer device to obtain the environmental monitoring image captured by the camera, and perform target detection on the monitoring image in the environmental monitoring image, and then determine the monitoring area according to the target detection result. For example, the image area where the preset type of target is detected is used as the monitoring area, or the area where the largest number of targets are detected is used as the monitoring area.

[0064] Optionally, the above method can be specifically applied to a microcontroller unit (MCU) in a computer device. The computer device can also be equipped with a target detection device such as an infrared sensor or a radar sensor. The computer device can obtain environmental monitoring images when the target detection device is triggered.

[0065] S220: Verify the target objects detected in each monitoring area according to the preset monitoring type.

[0066] The preset monitoring type is a type of target object to be monitored that is preset. For example, the preset monitoring type may include a human figure, a face, a package, an animal, or a vehicle, etc. In this embodiment, the preset monitoring type is not specifically limited.

[0067] Optionally, after determining the monitoring areas, the computer device may further obtain target detection results for each monitoring area, and verify the target objects detected in each monitoring area according to a preset monitoring type to obtain a verification result of whether the type of the target objects detected in each monitoring area meets the preset monitoring type.

[0068] Optionally, the computer device may pre-store reference objects of the same type uploaded when setting the preset monitoring type. When verifying the target object detected in each monitoring area, the computer device may obtain the image similarity between the area of ​​the target object detected in the monitoring area and the corresponding reference object image of the same type, and determine the verification result based on the image similarity. If the obtained image similarity is greater than the similarity threshold, the verification result is that the type of the detected target object meets the preset monitoring type; conversely, if the obtained image similarity is less than or equal to the similarity threshold, the verification result is that the type of the detected target object does not meet the preset monitoring type. For example, the preset monitoring type is humanoid, the computer device obtains the target object detected in the monitoring area, and obtains the image similarity between the target object and the pre-stored humanoid reference image. If the image similarity is 60%, which is greater than the similarity threshold of 50%, the computer device determines that the verification result is that the target object detected in the monitoring area meets the preset monitoring type.

[0069] Optionally, the computer device may pre-store the misjudged object image corresponding to the preset monitoring type uploaded when the preset monitoring type is set. When verifying the target object detected in each monitoring area, the computer device may obtain the image similarity between the area where the target object is detected in the monitoring area and the corresponding misjudged object image, and determine the verification result according to the image similarity. If the obtained image similarity is greater than the similarity threshold, the verification result is that the type of the detected target object does not conform to the preset monitoring type; conversely, if the obtained image similarity is less than or equal to the similarity threshold, the verification result is that the type of the detected target object conforms to the preset monitoring type. For example, the preset monitoring type is a human figure, and the misjudged object image is a tree trunk image. The computer device obtains the target object detected in the monitoring area, and obtains the image similarity between the target object and the pre-stored tree trunk image. If the image similarity is 65%, which is greater than the similarity threshold of 50%, the computer device determines that the verification result is that the target object detected in the monitoring area does not conform to the preset monitoring type.

[0070] S230. Adjust the monitoring strategy of the monitoring area according to the verification result.

[0071] Among them, the monitoring strategy is used to characterize the monitoring means to be performed on the monitoring area. For example, it can include the confidence thresholds used for different types of target objects during target detection, whether to perform an alarm warning in the form of sound and light after detecting a target object of a preset monitoring type, or to perform a video recording and send the recorded video to the client for feedback warning, and whether to perform target detection of the preset monitoring type for the corresponding monitoring area, etc.

[0072] Optionally, the computer device may adjust the confidence threshold and early warning mode of the preset monitoring type in the monitoring strategy based on the verification result, and may also adjust whether to perform target detection of the preset monitoring type in the monitoring area. For example, if the verification result is that the target object detected in the monitoring area does not conform to the preset monitoring type, the computer device may increase the confidence threshold of the preset monitoring type in the corresponding monitoring area, or no longer perform detection of the target object of the preset monitoring type in the monitoring area.

[0073] S240: Monitor the target objects of each preset monitoring type according to the adjusted monitoring strategy.

[0074] Optionally, after obtaining the adjusted monitoring strategy, the computer device can detect the target object of the preset monitoring type in the corresponding monitoring area according to the adjusted monitoring strategy. For example, in the adjusted monitoring strategy of a certain monitoring area, the confidence threshold of the preset monitoring type is adjusted from 50% to 65%. If the computer device outputs that the confidence of a certain target object in the detection area is a human figure is 55%, and after judging the confidence threshold of 65%, it is determined that the target object is not a human figure, the computer device can not issue an early warning for the target object in the monitoring area, or not detect the human-shaped target object.

[0075] In this embodiment, the computer device acquires an environmental monitoring image and determines at least one monitoring area in the environmental monitoring image, so as to verify the target objects detected in each monitoring area according to the preset monitoring type, and adjust the monitoring strategy of the monitoring area according to the verification result, and then monitor the target objects of each preset monitoring type according to the adjusted monitoring strategy. In the above method, the target objects detected in the monitoring area can be verified based on the preset monitoring type, so as to adjust the monitoring strategy of the corresponding monitoring area according to the verification result, so as to formulate respective monitoring strategies for different monitoring areas, realize the fine-grained division of monitoring strategies, and improve the monitoring flexibility while improving the monitoring accuracy.

[0076] In practical applications, the policy content that needs to be adjusted can be determined based on the verification results. Based on this, in one embodiment, Figure 3 As shown, the above S230, adjusting the monitoring strategy of the monitoring area according to the verification result, includes:

[0077] S310: Determine the policy adjustment content according to the verification result.

[0078] The verification result is used to characterize whether there is a false detection or missed detection in the target detection process. The policy adjustment content is used to characterize the policy content that needs to be adjusted. For example, the policy adjustment content may include increasing / decreasing the confidence threshold of the preset monitoring type, or giving up monitoring.

[0079] Optionally, after obtaining the verification result for the target object in the monitoring area, the computer device may determine different policy adjustment contents based on different verification results.

[0080] Optionally, when the verification result is that the target object detected in the monitoring area does not conform to the preset monitoring type, it indicates that there is a false detection in the target detection process, and the computer device can determine that the corresponding strategy adjustment content is to increase the confidence threshold of the preset monitoring type. Among them, the false detection in the target detection process indicates that the confidence threshold of the preset monitoring type used in the currently implemented monitoring strategy is low, which causes the false detection. The confidence threshold of the preset monitoring type can be increased to improve the false detection.

[0081] Optionally, when the verification result shows that the target objects detected in the monitoring area meet the preset monitoring type but are not all detected, it indicates that there are missed detections in the target detection process, and the computer device can determine that the corresponding strategy adjustment content is to reduce the confidence threshold of the preset monitoring type. Among them, the missed detection in the target detection process indicates that the confidence threshold of the preset monitoring type used in the monitoring strategy currently implemented in the detection area is high, causing the missed detection, and the confidence threshold of the preset monitoring type can be reduced to improve the missed detection.

[0082] Optionally, when the verification result is that the target object detected in the monitoring area does not conform to the preset monitoring type, and the confidence threshold of the preset monitoring type is equal to the preset minimum confidence, the computer device can determine that the corresponding policy adjustment content is to abandon monitoring. Among them, there is a false detection in the target detection process, but the confidence threshold of the preset monitoring type is equal to the preset minimum confidence, indicating that there is an object in the monitoring area that is very similar to the target object of the preset monitoring type, and the monitoring of the monitoring area can be abandoned to reduce resource waste.

[0083] S320: Adjust the monitoring strategy of the monitoring area according to the strategy adjustment content.

[0084] Optionally, after determining the policy adjustment content corresponding to the verification result, the computer device may make adaptive adjustments to the monitoring policy of the monitoring area with reference to the policy adjustment content. For example, when the policy adjustment content is to increase the confidence threshold of the preset monitoring type, the computer device increases the confidence threshold of the preset monitoring type in the monitoring policy of the monitoring area; when the policy adjustment content is to reduce the confidence threshold of the preset monitoring type, the computer device reduces the confidence threshold of the preset monitoring type in the monitoring policy of the monitoring area; when the policy adjustment content is to abandon monitoring, the computer device does not execute the monitoring policy of the monitoring area.

[0085] In this embodiment, the computer device determines the content of the policy adjustment according to the verification result, and then adjusts the monitoring policy of the monitoring area according to the content of the policy adjustment. In the above method, the verification result can be used to reflect the actual monitoring situation of the monitoring area, and the content of the policy adjustment can be accurately determined based on the verification result, and then the monitoring policy of the monitoring area can be adjusted in a targeted manner, thereby improving the accuracy of the policy adjustment.

[0086] The above policy adjustment content can be characterized based on the operation instructions fed back by the user end. Figure 4 As shown, the above S310, determining the policy adjustment content according to the verification result, includes:

[0087] S410: Send the monitoring area to the user terminal, and receive an operation instruction from the user terminal for the target object detected in the monitoring area.

[0088] The operation instruction is used to represent the policy adjustment content. For example, the operation instruction may carry the policy adjustment content.

[0089] It should be noted that the computer device and the user terminal communicate with each other, and the computer device can send the determined monitoring area (i.e., the detection area image) to the user terminal. The monitoring area includes the target object marked by the target detection and the type of the corresponding target object. The user can determine whether there is a false detection or missed detection verification result in the target detection process based on the monitoring area displayed by the user terminal, and operate the target object detected in the monitoring area based on the verification result to generate an operation instruction carrying the corresponding policy adjustment content.

[0090] Optionally, if the user selects a target object in the monitoring area and modifies the confidence threshold of the preset monitoring type corresponding to the target object, the user end generates a confidence instruction carrying the preset monitoring type and the modified confidence threshold; if the user selects a target object in the monitoring area and deletes or ignores the preset monitoring type corresponding to the target object, the user end generates an instruction to abandon monitoring carrying the preset monitoring type; optionally, if the user chooses to continue monitoring, the user end generates an instruction to continue monitoring.

[0091] Optionally, the user may also choose to ignore this monitoring, and the user end generates an instruction to ignore monitoring.

[0092] S420. Adjust the monitoring strategy of the monitoring area according to the operation instruction.

[0093] Optionally, after receiving the operation instruction fed back by the user terminal, the computer device may adjust the monitoring strategy of the corresponding monitoring area according to the information carried in the operation instruction.

[0094] In an optional embodiment, the monitoring strategy includes a preset monitoring type and a confidence threshold corresponding to the preset monitoring type, such as Figure 5 As shown, the above S420, adjusting the monitoring strategy of the monitoring area according to the operation instruction, includes:

[0095] S510: If the operation instruction is a confidence adjustment instruction, obtain the first object type indicated by the confidence adjustment instruction and the adjusted confidence carried therein, and use the adjusted confidence to replace the confidence threshold corresponding to the first object type in the monitoring strategy.

[0096] The confidence adjustment instruction is used to adjust the confidence threshold of the target object.

[0097] Optionally, after receiving the confidence adjustment instruction fed back by the user end, the computer device obtains the first object type indicated by the confidence adjustment instruction, and obtains the confidence carried by the confidence adjustment instruction as the adjusted confidence. When adjusting the monitoring strategy of the corresponding monitoring area, the computer device directly uses the adjusted confidence to replace the confidence threshold corresponding to the target object of the first object type in the monitoring strategy of the corresponding monitoring area. For example, for monitoring area A, the corresponding monitoring strategy includes a preset monitoring type including a human figure, and the corresponding confidence threshold is 50%. When receiving the confidence adjustment instruction of the monitoring area A, the first object type indicated by the confidence adjustment instruction is a human figure, and the adjusted confidence carried is 60%. The computer device directly replaces the preset monitoring type human figure confidence threshold of 50% in the monitoring strategy of the monitoring area A with the adjusted confidence of 60%. After modification, the monitoring strategy corresponding to the monitoring area A includes a corresponding preset monitoring type human figure confidence threshold of 60%.

[0098] S520: If the operation instruction is an instruction to abandon monitoring, obtain the second object type indicated in the instruction to abandon monitoring, and delete the second object type in the preset monitoring types in the monitoring strategy.

[0099] The abandon monitoring instruction is used to abandon the detection of the target object in the monitoring area.

[0100] Optionally, after receiving the abandon monitoring instruction fed back by the user terminal, the computer device obtains the second object type indicated by the abandon monitoring instruction. When adjusting the monitoring strategy of the corresponding monitoring area, the computer device directly deletes the second object type in the preset monitoring type in the monitoring strategy of the corresponding monitoring area. For example, for monitoring area B, the preset monitoring types in the corresponding monitoring strategy include human figures and packages. When receiving the abandon monitoring instruction of the monitoring area B, the second object type indicated by the abandon monitoring instruction is a human figure. The computer device directly deletes the human figure in the preset monitoring type in the monitoring area B. After the modification, the preset monitoring types in the monitoring strategy corresponding to the monitoring area B only include packages.

[0101] In this embodiment, the computer device sends the monitoring area to the user terminal, receives the operation instructions of the user terminal for the target object detected in the monitoring area, and adjusts the monitoring strategy of the monitoring area according to the operation instructions. Among them, the operation instructions are used to characterize the verification results; the monitoring strategy includes a preset monitoring type and a confidence threshold corresponding to the preset monitoring type; in the case where the operation instruction is a confidence adjustment instruction, the first object type indicated by the confidence adjustment instruction and the adjusted confidence carried are obtained, and the confidence threshold corresponding to the first object type in the monitoring strategy is replaced by the adjusted confidence; in the case where the operation instruction is a command to abandon monitoring, the second object type indicated in the command to abandon monitoring is obtained, and the second object type in the monitoring strategy is deleted. The operation instructions generated based on the user operation can be used to characterize the verification result of whether the type of the target object detected in the monitoring area meets the preset monitoring type, thereby improving the accuracy of the verification result and further improving the monitoring accuracy.

[0102] The monitoring area can be determined based on detecting a target object of a preset monitoring type for each frame of the environmental monitoring image. Based on this, in one embodiment, Figure 6 As shown, in the above S210, determining at least one monitoring area in the environmental monitoring image includes:

[0103] S610: For a monitoring image in the environmental monitoring image, use a target detection model to detect a target object of a preset monitoring type on the monitoring image to obtain a target distribution map of the monitoring image.

[0104] Among them, the target detection model is used to detect the target object of the preset monitoring type in the avatar. Optionally, the target detection model can correspond to the preset monitoring type one by one, and one target detection model can be used to detect the target object of the corresponding preset monitoring type, and can be trained using a large number of target object images of the preset monitoring type. For example, the human detection target is used to detect the target object of the preset monitoring type human shape, and the package detection model is used to detect the target object of the preset monitoring type package.

[0105] Optionally, the computer device may input each monitoring image frame in the environmental monitoring image into the target detection model corresponding to each preset monitoring type, so as to detect the target object of the preset monitoring type in each monitoring image frame in the environmental monitoring image, and mark the detected target object in each monitoring image frame according to the type, so as to obtain the target distribution map corresponding to each monitoring image frame. Figure 7 As shown, the target distribution map selects target objects of different preset monitoring types with target frames of different colors, including human-shaped target frames with confidence levels of 0.61 and 0.54, and a package target frame with a confidence level of 0.64.

[0106] S620: Determine a monitoring area according to a target distribution map of the monitoring image in the environmental monitoring image.

[0107] Optionally, the computer device may obtain a target distribution map of the monitoring image in the environmental monitoring image, determine a reference area with the largest number of target objects in the target distribution map, and compare multiple time-continuous monitoring images to determine whether there is a reference area in the target distribution map of the detection image with a preset number of consecutive frames, and if so, use the reference area as the above-mentioned monitoring area. For example, if the preset number of frames is 50, if the reference area with the largest number of target objects in the target distribution map of the monitoring image of 60 consecutive frames in the environmental monitoring image is all in the lower left corner area, the computer device determines that the lower left corner area is the monitoring area.

[0108] In this embodiment, for the monitoring image in the environmental monitoring image, the computer device uses the target detection model to detect the target object of the preset monitoring type on the monitoring image, obtains the target distribution map of the monitoring image, and then determines the monitoring area according to the target distribution map of the monitoring image in the environmental monitoring image. In the above method, the target detection model is used to detect the target object of the preset monitoring type on the monitoring image, which improves the detection accuracy of the target object, obtains the target distribution map that accurately reflects the actual distribution of the target object, and then determines the monitoring area according to the target distribution map, which improves the monitoring attention to the target object of the preset monitoring type, and helps to improve the effective monitoring of the target object of the preset monitoring type.

[0109] In order to improve the regularity of the monitoring area, in one embodiment, as Figure 8 As shown, the above S620, determining the monitoring area according to the target distribution map of the monitoring image in the environmental monitoring image, includes:

[0110] S810: Use a preset division template to divide the target distribution map of each monitoring image in the environmental monitoring image into regions.

[0111] The preset partitioning template is used to determine the number, arrangement style and shape of the candidate regions obtained by the partitioning. For example, the nine-square partitioning template is used to partition the target distribution map into 9 rectangular candidate regions arranged in a 3*3 arrangement. Optionally, the computer device may use the preset partitioning template to perform region partitioning on each target distribution map to divide the target distribution map into multiple candidate regions. For example, Figure 7 As shown, the computer device can use a preset nine-grid partitioning template to divide the target distribution map into nine candidate areas. In this embodiment, the number and shape of the candidate areas obtained by the partitioning are not specifically limited, and the partitioning template can be set according to the needs.

[0112] S820: Determine a monitoring area from the multiple candidate areas obtained by division.

[0113] Optionally, after the target distribution map is divided into regions, the computer device can determine the monitoring area from the multiple candidate areas obtained by the division according to the distribution of the target objects in each target distribution map. The monitoring area can be the same or different with respect to the monitoring image in the environmental monitoring image. For example, taking the example of using a preset nine-square grid division template to divide the target distribution map into 9 candidate areas, for all monitoring images in the environmental monitoring image, the candidate area in the upper left corner can be used as the monitoring area for continuous monitoring, or the candidate area in the upper left corner can be used as the monitoring area for some monitoring images, the candidate area in the middle can be used as the monitoring area for some monitoring images, and the candidate area in the lower left corner can be used as the monitoring area for some monitoring images.

[0114] In an optional embodiment, if Fig. 9 As shown, the above S820, determining the monitoring area from the multiple candidate areas obtained by division, includes:

[0115] S910. For each target distribution map, obtain detection information of the target object detected in each candidate area.

[0116] Optionally, the detection information of the target object may include attribute information of the target object. Exemplarily, the detection information of the target object includes at least one of the type, quantity, position, size, etc. of the target object.

[0117] Optionally, after the computer device divides the target distribution image into multiple candidate areas, it obtains the detection information of the target object in each candidate area. For example, continuing the above example, after the computer device divides the target distribution image into nine candidate areas, it obtains the type, quantity and position of the target object detected in each candidate area.

[0118] S920: Determine a monitoring area from multiple candidate areas according to detection information of target objects in the candidate areas in each target distribution map.

[0119] Optionally, the probability of false detection is higher in an area with a large number of target objects, and the computer device may determine the monitoring area from the multiple candidate areas according to the number of target objects detected in the multiple candidate areas in each target distribution map. For example, the computer device may count the number of target objects in each candidate area, and determine the candidate area with the largest number of targets as the monitoring area.

[0120] In this embodiment, the computer device uses a preset division template to divide the target distribution map of each monitoring image in the environmental monitoring image into regions, so as to determine the monitoring area from the multiple candidate areas obtained by the division. Among them, for each target distribution map, the computer device obtains the detection information of the target object detected in each candidate area, so as to determine the monitoring area in multiple candidate areas according to the detection information of the target object in the candidate area in each target distribution map. In the above method, the target distribution map of each monitoring image is divided into regions using a preset division template, so as to realize the standardized division of the target distribution map, improve the regularity of the candidate areas obtained by the division, and then determine the monitoring area according to the detection information of the target object detected in the candidate area, so as to improve the effective monitoring of the target object of the preset monitoring type.

[0121] In actual applications, if a static object in the environment is mistakenly detected as a target object of a preset monitoring type, the computer device will continue to trigger monitoring, such as issuing an early warning or recording a video, which not only causes a certain waste of resources, but also fails to achieve an effective security monitoring effect. Therefore, it is necessary to focus on the area containing static objects in the environmental monitoring image. Based on this, in one embodiment, if Fig.10 As shown, the above S920, determining a monitoring area in multiple candidate areas according to the detection information of the target object in the candidate area in each target distribution map, includes:

[0122] S1010. For each target distribution map, obtain the position and type of the target object in each candidate area.

[0123] The detection information of the target object includes the location and type of the target object.

[0124] Optionally, the environmental monitoring image includes multiple monitoring image frames, each monitoring image frame corresponds to a target distribution map frame, and the computer device obtains the position and type of the target object in each candidate area in the corresponding target distribution map for each target distribution map.

[0125] S1020: Determine a monitoring area among multiple candidate areas according to the location and type of the target object in each candidate area.

[0126] Optionally, the computer device may select a candidate area including a static object from multiple candidate areas based on the position and type of the target object in each candidate area in the target distribution map corresponding to each frame of monitoring, and use the candidate area as the detection area for each frame of monitoring image in the environmental monitoring image.

[0127] In an optional embodiment, if the type and position of the target object in the same candidate area match between the target distribution maps of consecutive preset frames, the target object can be determined to be a static object, and the same candidate area is the monitoring area accordingly. Fig.11 As shown, the above S1020, determining a monitoring area in multiple candidate areas according to the location and type of the target object in each candidate area, includes:

[0128] S1110 , comparing the locations and types of target objects in the same candidate area among the target distribution maps.

[0129] The monitoring images in the environmental monitoring image increase over time, and the computer device can determine the monitoring area based on the target distribution map corresponding to the existing monitoring images. The monitoring area can change with the change of the target distribution map.

[0130] Optionally, after obtaining the position and type of the target object in each candidate area in each target distribution map, the computer device can compare the target distribution maps horizontally to compare the position and type of the target object in the same candidate area. For example, the environmental monitoring image currently includes 10 frames of target distribution maps corresponding to 10 frames of monitoring images, and the computer device compares the position and type of the target object in candidate area 1, the position and type of the target object in candidate area 2, and the position and type of the target object in candidate area 3 in these 10 frames of monitoring images...

[0131] S1120: Determine the same candidate area in which the position and type of the target object match in the target distribution map of a preset number of consecutive frames as the monitoring area.

[0132] Optionally, during the comparison process, the computer device determines whether the positions and types of target objects in the same candidate area match between adjacent target distribution maps, so as to obtain the number of consecutive frames of target distribution maps in which the positions and types of target objects in the same candidate area match. If the number of consecutive frames is greater than a preset number of frames, the same candidate area is determined to be a monitoring area. Fig.12 As shown, the target objects of the preset monitoring type detected in the first candidate area (the upper left corner area in the nine-square grid) in multiple target distribution images are displayed in one target distribution image. Among them, if the types of the target objects in the first candidate area in the target distribution images for 120 consecutive frames (greater than the preset number of frames 100) are all human-shaped and the positions match, the computer device can determine that the first candidate area is the monitoring area.

[0133] It should be noted that if the position and type of the target object in the same candidate area match between the target distribution images of a consecutive preset number of frames, indicating that the same target object appears in the monitoring images of a consecutive preset number of frames in the environmental monitoring image and its position remains unchanged, the target object is a static object, and the computer device can determine that the candidate area including the static object is the monitoring area (also referred to as a "high-frequency trigger area").

[0134] Optionally, the computer device may obtain the intersection-over-union ratio between the target frames of the detected target objects to determine whether the positions of the target objects match according to the intersection-over-union ratio. The intersection-over-union (IoU) is used to characterize the degree of overlap between the target frames of the selected target objects, which can be obtained by dividing the intersection between the two target frames by the union between the two target frames.

[0135] Optionally, the computer device may compare the intersection-and-union ratios between target objects in the same candidate area between time-continuous target distribution maps, wherein if the intersection-and-union ratio is greater than an intersection-and-union ratio threshold, it is determined that the positions of the target objects match; otherwise, if the intersection-and-union ratio is less than or equal to the intersection-and-union ratio threshold, it is determined that the positions of the target objects do not match.

[0136] In this embodiment, the computer device obtains the position and type of the target object in each candidate area for each target distribution map, and then determines the monitoring area in multiple candidate areas according to the position and type of the target object in each candidate area. Among them, the position and type of the target object in the same candidate area between each target distribution map can be compared, and the same candidate area where the position and type of the target object in the target distribution map of a preset number of consecutive frames are matched is determined as the monitoring area. In the above method, based on the comparison of the position and type of the target object in the same candidate area between the target distribution maps, the area where the static object that is mistakenly detected as the preset monitoring type is located is extracted as the monitoring area, so as to verify the monitoring area later, avoid the waste of resources caused by false detection, and reduce unnecessary video recording, thereby improving the security monitoring effect.

[0137] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0138] Based on the same inventive concept, the embodiment of the present application also provides a security monitoring device for implementing the security monitoring method involved above. The implementation solution provided by the device to solve the problem is similar to the implementation solution recorded in the above method, so the specific limitations in one or more security monitoring device embodiments provided below can refer to the limitations of the security monitoring method above, and will not be repeated here.

[0139] In one embodiment, Fig.13 As shown, a security monitoring device is provided, including: an area determination module 1301, a target verification module 1302, a strategy adjustment module 1303 and a monitoring execution module 1304, wherein:

[0140] The region determination module 1301 is used to obtain an environmental monitoring image and determine at least one monitoring region in the environmental monitoring image;

[0141] The target verification module 1302 is used to verify the target object detected in each monitoring area according to the preset monitoring type;

[0142] The strategy adjustment module 1303 is used to adjust the monitoring strategy of the monitoring area according to the verification result;

[0143] The monitoring execution module 1304 is used to monitor the target objects of each preset monitoring type according to the adjusted monitoring strategy.

[0144] In one embodiment, the policy adjustment module 1003 includes:

[0145] A content determination submodule is used to determine the strategy to adjust the content according to the verification result;

[0146] The strategy adjustment submodule is used to adjust the monitoring strategy of the monitoring area according to the strategy adjustment content.

[0147] In one embodiment, the policy adjustment submodule includes:

[0148] The information feedback unit is used to send the monitoring area to the user end and receive the operation instruction of the user end for the target object detected in the monitoring area; the operation instruction is used to represent the content of the strategy adjustment;

[0149] The instruction adjustment unit is used to adjust the monitoring strategy of the monitoring area according to the operation instruction.

[0150] In one embodiment, the monitoring strategy includes a preset monitoring type and a confidence threshold corresponding to the preset monitoring type; and the instruction adjustment unit includes:

[0151] A first adjustment subunit is used to obtain, if the operation instruction is a confidence adjustment instruction, the first object type indicated by the confidence adjustment instruction and the adjustment confidence carried therein, and replace the confidence threshold corresponding to the first object type in the monitoring strategy with the adjustment confidence;

[0152] The second adjustment subunit is used for obtaining the second object type indicated in the abandon monitoring instruction and deleting the second object type in the monitoring strategy if the operation instruction is an abandon monitoring instruction.

[0153] In one embodiment, the region determination module 1301 includes:

[0154] The target distribution submodule is used to detect the target objects of the preset monitoring type in the monitoring image of the environmental monitoring image using the target detection model to obtain the target distribution map of the monitoring image;

[0155] The monitoring area submodule is used to determine the monitoring area according to the target distribution map of the monitoring image in the environmental monitoring image.

[0156] In one embodiment, the monitoring area submodule includes:

[0157] A region division unit, used for dividing the target distribution map of each monitoring image in the environmental monitoring image into regions using a preset division template;

[0158] The area determination unit is used to determine the monitoring area from the multiple candidate areas obtained by division.

[0159] In one embodiment, the area division unit includes:

[0160] An information acquisition subunit, for acquiring detection information of target objects detected in each candidate area for each target distribution map;

[0161] The area determination subunit is used to determine the monitoring area in multiple candidate areas according to the detection information of the target objects in the candidate areas in each target distribution map.

[0162] In one embodiment, the region determination subunit is configured to:

[0163] For each target distribution map, obtain the location and type of the target object in each candidate area;

[0164] A monitoring area is determined among the multiple candidate areas according to the location and type of the target object in each candidate area.

[0165] In one of the embodiments, the region determination subunit is further configured to:

[0166] Compare the locations and types of target objects in the same candidate area between target distribution maps;

[0167] The same candidate area in which the position and type of the target object in the target distribution map of a preset number of consecutive frames are matched is determined as the monitoring area.

[0168] Each module in the above security monitoring device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each module.

[0169] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0170] Acquire an environmental monitoring image and determine at least one monitoring area in the environmental monitoring image; verify the target objects detected in each monitoring area according to a preset monitoring type; adjust the monitoring strategy of the monitoring area according to the verification result; monitor the target objects of each preset monitoring type according to the adjusted monitoring strategy.

[0171] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0172] Determine the content of the strategy adjustment based on the verification results; adjust the monitoring strategy of the monitoring area based on the content of the strategy adjustment.

[0173] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0174] The monitoring area is sent to the user end, and an operation instruction of the user end for the target object detected in the monitoring area is received; the operation instruction is used to represent the content of the strategy adjustment; and the monitoring strategy of the monitoring area is adjusted according to the operation instruction.

[0175] In one embodiment, the monitoring strategy includes a preset monitoring type and a confidence threshold corresponding to the preset monitoring type; when the processor executes the computer program, the following steps are also implemented:

[0176] If the operation instruction is a confidence adjustment instruction, then the first object type indicated by the confidence adjustment instruction and the adjusted confidence carried are obtained, and the adjusted confidence is used to replace the confidence threshold corresponding to the first object type in the monitoring strategy; if the operation instruction is an instruction to abandon monitoring, then the second object type indicated in the instruction to abandon monitoring is obtained, and the second object type in the monitoring strategy is deleted.

[0177] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0178] For the monitoring images in the environmental monitoring images, the target detection model is used to detect the target objects of the preset monitoring type in the monitoring images to obtain the target distribution map of the monitoring images; the monitoring area is determined according to the target distribution map of the monitoring images in the environmental monitoring images.

[0179] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0180] A preset partitioning template is used to perform regional partitioning on the target distribution map of each monitoring image in the environmental monitoring image; and a monitoring area is determined from multiple candidate areas obtained by the partitioning.

[0181] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0182] For each target distribution map, detection information of the target object detected in each candidate area is obtained; and a monitoring area is determined among the multiple candidate areas according to the detection information of the target object in the candidate area in each target distribution map.

[0183] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0184] For each target distribution map, the location and type of the target object in each candidate area are obtained; and according to the location and type of the target object in each candidate area, a monitoring area is determined among the multiple candidate areas.

[0185] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0186] The positions and types of target objects in the same candidate area are compared between the target distribution maps; the same candidate area in which the positions and types of target objects in the target distribution maps of consecutive preset frames match is determined as the monitoring area.

[0187] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0188] Acquire an environmental monitoring image and determine at least one monitoring area in the environmental monitoring image; verify the target objects detected in each monitoring area according to a preset monitoring type; adjust the monitoring strategy of the monitoring area according to the verification result; monitor the target objects of each preset monitoring type according to the adjusted monitoring strategy.

[0189] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0190] Determine the content of the strategy adjustment based on the verification results; adjust the monitoring strategy of the monitoring area based on the content of the strategy adjustment.

[0191] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0192] The monitoring area is sent to the user end, and an operation instruction of the user end for the target object detected in the monitoring area is received; the operation instruction is used to represent the content of the strategy adjustment; and the monitoring strategy of the monitoring area is adjusted according to the operation instruction.

[0193] In one embodiment, the monitoring strategy includes a preset monitoring type and a confidence threshold corresponding to the preset monitoring type; when the computer program is executed by the processor, the following steps are also implemented:

[0194] If the operation instruction is a confidence adjustment instruction, then the first object type indicated by the confidence adjustment instruction and the adjusted confidence carried are obtained, and the adjusted confidence is used to replace the confidence threshold corresponding to the first object type in the monitoring strategy; if the operation instruction is an instruction to abandon monitoring, then the second object type indicated in the instruction to abandon monitoring is obtained, and the second object type in the monitoring strategy is deleted.

[0195] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0196] For the monitoring images in the environmental monitoring images, the target detection model is used to detect the target objects of the preset monitoring type in the monitoring images to obtain the target distribution map of the monitoring images; the monitoring area is determined according to the target distribution map of the monitoring images in the environmental monitoring images.

[0197] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0198] A preset partitioning template is used to perform regional partitioning on the target distribution map of each monitoring image in the environmental monitoring image; and a monitoring area is determined from multiple candidate areas obtained by the partitioning.

[0199] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0200] For each target distribution map, detection information of the target object detected in each candidate area is obtained; and a monitoring area is determined among the multiple candidate areas according to the detection information of the target object in the candidate area in each target distribution map.

[0201] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0202] For each target distribution map, the location and type of the target object in each candidate area are obtained; and according to the location and type of the target object in each candidate area, a monitoring area is determined among the multiple candidate areas.

[0203] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0204] The positions and types of target objects in the same candidate area are compared between the target distribution maps; the same candidate area in which the positions and types of target objects in the target distribution maps of consecutive preset frames match is determined as the monitoring area.

[0205] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0206] Acquire an environmental monitoring image and determine at least one monitoring area in the environmental monitoring image; verify the target objects detected in each monitoring area according to a preset monitoring type; adjust the monitoring strategy of the monitoring area according to the verification result; monitor the target objects of each preset monitoring type according to the adjusted monitoring strategy.

[0207] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0208] Determine the content of the strategy adjustment based on the verification results; adjust the monitoring strategy of the monitoring area based on the content of the strategy adjustment.

[0209] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0210] The monitoring area is sent to the user end, and an operation instruction of the user end for the target object detected in the monitoring area is received; the operation instruction is used to represent the content of the strategy adjustment; and the monitoring strategy of the monitoring area is adjusted according to the operation instruction.

[0211] In one embodiment, the monitoring strategy includes a preset monitoring type and a confidence threshold corresponding to the preset monitoring type; when the computer program is executed by the processor, the following steps are also implemented:

[0212] If the operation instruction is a confidence adjustment instruction, then the first object type indicated by the confidence adjustment instruction and the adjusted confidence carried are obtained, and the adjusted confidence is used to replace the confidence threshold corresponding to the first object type in the monitoring strategy; if the operation instruction is an instruction to abandon monitoring, then the second object type indicated in the instruction to abandon monitoring is obtained, and the second object type in the monitoring strategy is deleted.

[0213] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0214] For the monitoring images in the environmental monitoring images, the target detection model is used to detect the target objects of the preset monitoring type in the monitoring images to obtain the target distribution map of the monitoring images; the monitoring area is determined according to the target distribution map of the monitoring images in the environmental monitoring images.

[0215] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0216] A preset partitioning template is used to perform regional partitioning on the target distribution map of each monitoring image in the environmental monitoring image; and a monitoring area is determined from multiple candidate areas obtained by the partitioning.

[0217] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0218] For each target distribution map, detection information of the target object detected in each candidate area is obtained; and a monitoring area is determined among the multiple candidate areas according to the detection information of the target object in the candidate area in each target distribution map.

[0219] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0220] For each target distribution map, the location and type of the target object in each candidate area are obtained; and according to the location and type of the target object in each candidate area, a monitoring area is determined among the multiple candidate areas.

[0221] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0222] The positions and types of target objects in the same candidate area are compared between the target distribution maps; the same candidate area in which the positions and types of target objects in the target distribution maps of consecutive preset frames match is determined as the monitoring area.

[0223] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0224] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0225] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A security monitoring method, characterized in that: The method comprises: Acquire an environmental monitoring image, and determine at least one monitoring area in the environmental monitoring image; According to the preset monitoring type, verify the target object detected in each of the monitoring areas; Adjust the monitoring strategy of the monitoring area according to the verification result; The target objects of each preset monitoring type are monitored according to the adjusted monitoring strategy.

2. The method according to claim 1, characterized in that The step of adjusting the monitoring strategy of the monitoring area according to the verification result includes: Determine the policy adjustment content according to the verification result; The monitoring strategy of the monitoring area is adjusted according to the strategy adjustment content.

3. The method according to claim 2, characterized in that The step of adjusting the monitoring strategy of the monitoring area according to the strategy adjustment content includes: Sending the monitoring area to a user terminal, and receiving an operation instruction from the user terminal for a target object detected in the monitoring area; the operation instruction is used to represent the policy adjustment content; The monitoring strategy of the monitoring area is adjusted according to the operating instruction.

4. The method according to claim 3, characterized in that The monitoring strategy includes the preset monitoring type and the confidence threshold corresponding to the preset monitoring type; the monitoring strategy of the monitoring area adjusted according to the operation instruction includes: If the operation instruction is a confidence adjustment instruction, obtaining the first object type indicated by the confidence adjustment instruction and the adjustment confidence carried therein, and replacing the confidence threshold corresponding to the first object type in the monitoring strategy with the adjustment confidence; If the operation instruction is an instruction to abandon monitoring, the second object type indicated in the instruction to abandon monitoring is obtained, and the second object type in the monitoring strategy is deleted.

5. The method according to any one of claims 1 to 4, characterized in that The determining at least one monitoring area in the environmental monitoring image comprises: For the monitoring image in the environmental monitoring image, using a target detection model to detect the target object of the preset monitoring type on the monitoring image to obtain a target distribution map of the monitoring image; The monitoring area is determined according to a target distribution map of the monitoring image in the environmental monitoring image.

6. The method according to claim 5, characterized in that The determining the monitoring area according to the target distribution map of each monitoring image in the environmental monitoring image includes: Using a preset division template to divide the target distribution map of each monitoring image in the environmental monitoring image into regions; The monitoring area is determined from the multiple candidate areas obtained by division.

7. The method according to claim 6, characterized in that The determining the monitoring area from the multiple candidate areas obtained by division includes: For each of the target distribution maps, obtaining detection information of the target object detected in each candidate area; The monitoring area is determined in the plurality of candidate areas according to detection information of target objects in the candidate areas in each of the target distribution maps.

8. The method according to claim 7, characterized in that Determining the monitoring area in the plurality of candidate areas according to the detection information of the target object in the candidate area in each of the target distribution maps comprises: For each of the target distribution maps, the position and type of the target object in each of the candidate areas are obtained; and according to the position and type of the target object in each of the candidate areas, the monitoring area is determined in the multiple candidate areas.

9. The method according to claim 8, characterized in that Determining the monitoring area in the plurality of candidate areas according to the location and type of the target object in each of the candidate areas comprises: comparing the locations and types of target objects within the same candidate area between the target distribution maps; The same candidate area in which the position and type of the target object match in the target distribution map of a continuous preset number of frames is determined as the monitoring area.

10. A security monitoring device, characterized in that: The device comprises: An area determination module, used to obtain an environmental monitoring image and determine at least one monitoring area in the environmental monitoring image; A target verification module, used to verify the target object detected in each monitoring area according to a preset monitoring type; A strategy adjustment module, used to adjust the monitoring strategy of the monitoring area according to the verification result; The monitoring execution module is used to monitor the target objects of each preset monitoring type according to the adjusted monitoring strategy.

Images