Warning method and device based on indoor scene and integrating video with Bluetooth position marker

Abstract

The invention provides an indoor scene-based video fusion Bluetooth beacon alarm method and device, and the method comprises the steps: obtaining corresponding Bluetooth beacon information and video information based on an alarm triggering behavior and the correlation between a previously configured Bluetooth beacon device and a shooting device; according to the Bluetooth beacon information and the video information, in combination with a preset weight distribution mechanism, obtaining a weighted fusion feature; and according to the weighted fusion features, performing classification by using a decision tree trained in advance to obtain a classification result, and determining whether to give an alarm or not according to the classification result. According to the invention, the Bluetooth beacon information and the video information are obtained based on the association between the Bluetooth beacon equipment and the shooting equipment under the condition that the alarm is triggered, so that the Bluetooth beacon information and the video information are subjected to multi-source fusion and behavior classification, false alarm and delay are reduced, the alarm efficiency and practicability are improved, and the user experience is improved. On the basis that the alarm accuracy is ensured, the situation that single-factor alarm is poor in accuracy and reliability is avoided.

Description

Technical Field

[0001] The present invention relates to the field of security monitoring technology, and in particular to an alarm method and device based on indoor scene video fusion Bluetooth location marker. Background Art

[0002] In most alarm scenarios, single-factor alarm is generally used, that is, a separate alarm method is used, such as Bluetooth alarm or video alarm.

[0003] However, Bluetooth alarms can usually only provide the location information of the device and cannot accurately restore the details of the event, such as human behavior, item status, etc., and cannot provide corresponding disposal measures, such as people falling and equipment damage. In addition, Bluetooth alarms are easily affected by device positioning drift, accidental device touches, etc., and the authenticity and accuracy of the alarm cannot be intuitively judged: due to algorithm and machine configuration factors, video alarms have a certain delay in identifying alarms and cannot issue real-time alarms. They are also easily affected by environmental factors, resulting in false alarms and missed alarms. Summary of the invention

[0004] The present invention provides an alarm method and device based on indoor scene video fusion Bluetooth marker, which is used to solve the defect that single-factor alarm in the prior art cannot accurately reflect the accuracy of events, and retains the occurrence process of the alarm event while ensuring timely reflection of the accuracy of events, thereby improving the accuracy of alarms.

[0005] The present invention provides an alarm method for indoor scene-based video fusion Bluetooth beacon, comprising: based on the triggering alarm behavior and the association between the previously configured Bluetooth beacon device and the shooting device, obtaining corresponding Bluetooth beacon information and video information; obtaining weighted fusion features according to the Bluetooth beacon information and the video information in combination with a preset weight distribution mechanism; according to the weighted fusion features, using a previously trained decision tree for classification to obtain a classification result, and determining whether to alarm according to the classification result.

[0006] According to the present invention, an alarm method for indoor scene-based video fusion Bluetooth beacon is provided. Based on the triggering alarm behavior and the association between the previously configured Bluetooth beacon device and the shooting device, the corresponding Bluetooth beacon information and video information are obtained, including: based on the triggering alarm behavior, if the alarm behavior is Bluetooth triggered, the corresponding Bluetooth beacon information is obtained, and based on the shooting device previously associated with the Bluetooth beacon device, the corresponding video information is obtained; if the alarm behavior is video triggered, the corresponding video information is obtained, and based on the Bluetooth beacon associated with the shooting device corresponding to the video, the corresponding Bluetooth information is obtained to obtain the Bluetooth beacon information.

[0007] According to the present invention, an alarm method for video fusion Bluetooth beacon based on indoor scenes, before obtaining corresponding Bluetooth beacon information and video information based on the triggered alarm behavior and the association between the previously configured Bluetooth beacon device and the shooting device, includes: obtaining the shooting device location information and the Bluetooth beacon device location information; determining the distance between any shooting device and any Bluetooth beacon device according to the shooting device location information and the Bluetooth beacon device location information, the distance including the horizontal distance and the vertical distance; determining whether the vertical distance is less than a preset height threshold; based on the vertical distance being less than the preset height threshold, screening the shooting devices and Bluetooth beacon devices with the corresponding vertical distance, and eliminating the shooting devices and Bluetooth beacon devices corresponding to the vertical distance greater than or equal to the preset height threshold; determining whether the horizontal distance between the screened shooting devices and Bluetooth beacon devices is less than or equal to the preset horizontal distance threshold; based on the horizontal distance between the screened shooting devices and Bluetooth beacon devices being less than or equal to the preset horizontal distance threshold, associating the shooting devices and Bluetooth beacon devices with the corresponding vertical distance.

[0008] According to an alarm method for a video-fused Bluetooth beacon based on an indoor scene provided by the present invention, a weighted fusion feature is obtained according to Bluetooth beacon information and video information in combination with a preset weight allocation mechanism, including: extracting features from Bluetooth beacon information to obtain Bluetooth features, and determining Bluetooth alarm classification results according to the Bluetooth features; extracting features from video information to obtain video features, and determining video alarm classification results according to the video features; determining the allocation weights of corresponding Bluetooth features and corresponding video features according to the Bluetooth alarm classification results and the video alarm classification results in combination with a preset weight allocation mechanism; and obtaining weighted fusion features according to the Bluetooth features, the video features, the allocation weights of the corresponding Bluetooth features, and the allocation weights of the corresponding video features.

[0009] According to an alarm method based on indoor scene video fusion Bluetooth beacon provided by the present invention, classification is performed using a previously trained decision tree according to weighted fusion features to obtain classification results, including: inputting the weighted fusion features into the decision tree to obtain the classification results output by the decision tree, the classification results including alarm behavior classification results and face recognition results; wherein the decision tree is trained based on training features and result labels corresponding to the training features; the decision tree performs behavior classification and face recognition based on the input weighted fusion features to obtain alarm behavior classification results and face recognition results.

[0010] According to an alarm method for indoor scene-based video fusion Bluetooth beacon provided by the present invention, a decision tree performs face recognition based on input weighted fusion features, including: the decision tree searches and matches the face tags previously stored in the leaf nodes based on the input weighted fusion features to obtain the corresponding face recognition results; or, the decision tree obtains the corresponding face recognition results based on the input weighted fusion features in combination with the face tags previously bound to the Bluetooth beacon device.

[0011] The present invention also provides an alarm device based on indoor scene video fusion Bluetooth beacon, including: an information acquisition module, which acquires corresponding Bluetooth beacon information and video information based on the triggering alarm behavior and the association between the previously configured Bluetooth beacon device and the shooting device; a weighted fusion module, which obtains weighted fusion features based on the Bluetooth beacon information and video information in combination with a preset weight distribution mechanism; a classification module, which classifies according to the weighted fusion features using a previously trained decision tree to obtain a classification result, and determines whether to alarm based on the classification result.

[0012] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for alarming a video fusion Bluetooth marker based on an indoor scene as described in any one of the above is implemented.

[0013] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the alarm method of the video fusion Bluetooth beacon based on the indoor scene as described in any one of the above is implemented.

[0014] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-mentioned indoor scene-based video fusion Bluetooth beacon alarm methods.

[0015] The alarm method and device of the video fusion Bluetooth beacon based on indoor scenes provided by the present invention obtain Bluetooth beacon information and video information based on the association between the Bluetooth beacon device and the shooting device when the alarm is triggered, so as to facilitate multi-source fusion of the obtained Bluetooth beacon information and video information and perform behavior classification, thereby reducing false alarms and delays, improving the efficiency and practicality of alarms, and further ensuring whether abnormal behavior occurs and whether an alarm is needed, while ensuring the accuracy of the alarm, avoiding the situation where the accuracy and reliability of single-factor alarms are poor. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 It is one of the flow charts of the alarm method based on indoor scene video fusion Bluetooth beacon provided by the present invention; Figure 2 This is the second flow chart of the alarm method based on indoor scene video fusion Bluetooth beacon provided by the present invention; Figure 3 It is a structural schematic diagram of an alarm device based on indoor scene video fusion Bluetooth beacon provided by the present invention; Figure 4 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] Figure 1 A schematic flow chart of an alarm method based on indoor scene video fusion Bluetooth beacon provided by the present invention is shown, and the method comprises: S11, based on the triggering alarm behavior and the association between the previously configured Bluetooth beacon device and the shooting device, obtaining corresponding Bluetooth beacon information and video information; S12, obtaining weighted fusion features according to the Bluetooth beacon information and the video information in combination with a preset weight allocation mechanism; S13 performs classification based on the weighted fusion features using a previously trained decision tree to obtain a classification result, and determines whether to issue an alarm based on the classification result.

[0020] It should be noted that the step numbers "S1N" in this specification do not represent the order of the alarm method based on the video fusion Bluetooth location marker of the indoor scene. Figure 2 The present invention describes the alarm method of the video fusion Bluetooth beacon based on indoor scenes.

[0021] Step S11, based on the alarm triggering behavior and the association between the previously configured Bluetooth beacon device and the shooting device, obtain the corresponding Bluetooth beacon information and video information.

[0022] In this embodiment, based on the triggering alarm behavior and the association between the previously configured Bluetooth beacon device and the shooting device, the corresponding Bluetooth beacon information and video information are obtained, including: based on the triggering alarm behavior, if the alarm behavior is Bluetooth triggered, the corresponding Bluetooth beacon information is obtained, and based on the shooting device previously associated with the Bluetooth beacon device, the corresponding video information is obtained; if the alarm behavior is video triggered, the corresponding video information is obtained, and based on the Bluetooth beacon associated with the shooting device corresponding to the video, the corresponding Bluetooth information is obtained to obtain the Bluetooth beacon information.

[0023] It should be added that Bluetooth triggered alarms include active triggering and passive triggering. Active triggering mainly uses special devices to actively trigger alarms, while passive triggering mainly uses signal strength or distance to judge the proximity between devices to issue alarms. Video triggered alarms mainly use video combined with AI image recognition, and issue alarms based on the set similarity threshold.

[0024] It is worth noting that the alarm is triggered by Bluetooth or video to obtain the Bluetooth beacon information and video information in the corresponding room, so as to facilitate the subsequent multi-source fusion based on the obtained Bluetooth beacon information and video information, so as to classify the behavior and determine whether to issue an alarm based on abnormal behavior, thereby improving the accuracy of the alarm and avoiding the poor accuracy and reliability of single-factor alarms.

[0025] In addition, abnormal behavior can be determined based on the actual indoor scenes involved. For example, in elderly care scenarios, it can be determined whether the elderly have abnormal behaviors such as falling. In another example, in public indoor places, whether there are behaviors that cause financial or safety risks to others. No further limitations are made here.

[0026] In an optional embodiment, after acquiring the Bluetooth beacon information, the method includes: calibrating and optimizing the Bluetooth beacon information to ensure the stability and accuracy of the Bluetooth beacon information.

[0027] In an optional embodiment, after acquiring the video information, the method includes: saving the video information for subsequent archiving and tracing.

[0028] In an optional embodiment, before obtaining corresponding Bluetooth beacon information and video information based on the triggered alarm behavior and the association between the previously configured Bluetooth beacon device and the shooting device, it includes: obtaining the shooting device location information and the Bluetooth beacon device location information; determining the distance between any shooting device and any Bluetooth beacon device based on the shooting device location information and the Bluetooth beacon device location information, the distance including the horizontal distance and the vertical distance; determining whether the vertical distance is less than a preset height threshold; based on the vertical distance being less than the preset height threshold, filtering the shooting devices and Bluetooth beacon devices with the corresponding vertical distance, and eliminating the shooting devices and Bluetooth beacon devices corresponding to the vertical distance greater than or equal to the preset height threshold; determining whether the horizontal distance of the screened shooting devices and Bluetooth beacon devices is less than or equal to the preset horizontal distance threshold; based on the horizontal distance of the screened shooting devices and Bluetooth beacon devices being less than or equal to the preset horizontal distance threshold, associating the shooting devices and Bluetooth beacon devices with the corresponding vertical distance.

[0029] It should be supplemented that determining whether the vertical distance is less than the preset height threshold also includes: if the vertical distance is greater than or equal to the preset height threshold, then the corresponding shooting device and Bluetooth beacon device with the vertical distance are not associated. Correspondingly, determining whether the horizontal distance between the screened shooting device and the Bluetooth beacon device is less than or equal to the preset horizontal distance threshold also includes: if the horizontal distance between the screened shooting device and the Bluetooth beacon device is greater than the preset horizontal distance threshold, then the corresponding shooting device and Bluetooth beacon device with the horizontal distance are not associated.

[0030] In addition, the preset height threshold and the preset horizontal distance threshold can be set according to the actual scene in which the shooting device and the Bluetooth beacon device are located. For example, in an indoor scene, the preset height threshold can be set according to the actual floor height, and the preset horizontal distance threshold can be set based on the associated range of the Bluetooth beacon device and the video device on the same floor. No further restrictions are made here.

[0031] In an optional embodiment, before associating the shooting device and the Bluetooth beacon device, it also includes: scanning the application scenario, establishing a virtual three-dimensional space, and configuring multiple Bluetooth beacon devices in the application scenario; obtaining shooting device parameters, and configuring a virtual shooting device in the virtual three-dimensional space according to the shooting device parameters; based on the above-mentioned association method, associating the shooting device and the Bluetooth beacon device in the virtual three-dimensional space, thereby realizing intelligent association between the Bluetooth device and the video content, facilitating subsequent joint analysis of Bluetooth and video, and improving the accuracy and reliability of the alarm.

[0032] Step S12, obtaining weighted fusion features according to the Bluetooth beacon information and the video information in combination with a preset weight allocation mechanism.

[0033] In this embodiment, based on the Bluetooth beacon information and the video information, combined with a preset weight allocation mechanism, a weighted fusion feature is obtained, including: extracting features from the Bluetooth beacon information to obtain Bluetooth features, and determining the Bluetooth alarm classification result based on the Bluetooth features; extracting features from the video information to obtain video features, and determining the video alarm classification result based on the video features; determining the allocation weights of the corresponding Bluetooth features and the corresponding video features based on the Bluetooth alarm classification results and the video alarm classification results, combined with a preset weight allocation mechanism; obtaining weighted fusion features based on the Bluetooth features, the video features, the allocation weights of the corresponding Bluetooth features, and the allocation weights of the corresponding video features.

[0034] It should be added that Bluetooth features may include signal strength, distance, etc., and video features include behavioral features and facial features. By using a preset weight allocation mechanism, Bluetooth features and video features are weighted and fused, so that the classification tree can be used to conduct a comprehensive analysis and judgment of Bluetooth features and video features, reduce false alarms and delays, and improve the efficiency and practicality of alarms. In addition, the preset weight allocation mechanism includes: configuring fixed weights for Bluetooth features and video features based on Bluetooth and algorithms for extracting video features; or, dynamically configuring the weights of Bluetooth features and video features based on the algorithm type. Specifically, the weights can be dynamically configured based on the risk and maturity of the algorithm. For example, an algorithm with low risk and high maturity can increase the allocation weight of video features between Bluetooth features and video features. No further limitation is made here.

[0035] In an optional embodiment, before obtaining the weighted fusion features based on the Bluetooth beacon information and the video information in combination with a preset weight distribution mechanism, it includes: performing preprocessing operations such as normalization and standardization on the Bluetooth features and the video features to improve the accuracy and generalization ability of the model.

[0036] Step S13 classifies the information using the previously trained decision tree based on the weighted fusion features to obtain a classification result, and determines whether to issue an alarm based on the classification result.

[0037] Specifically, according to the weighted fusion features, a previously trained decision tree is used for classification to obtain a classification result, including: inputting the weighted fusion features into the decision tree to obtain a classification result output by the decision tree, the classification result including an alarm behavior classification result and a face recognition result; wherein the decision tree is trained based on the training features and the result labels corresponding to the training features; the decision tree performs behavior classification and face recognition based on the input weighted fusion features to obtain an alarm behavior classification result and a face recognition result.

[0038] In this embodiment, before training the decision tree, a decision tree model is constructed using training features, and a suitable decision tree algorithm is selected for training and optimization, such as adjusting the depth of the tree, pruning, etc., to avoid overfitting, which can be specifically set according to the actual design. Each node of the decision tree represents a feature, and during the application process, it is divided into different branches according to the value of the weighted fusion feature. In addition, in the branches of the decision tree, further divisions can be made according to the weighted fusion features to achieve behavior classification.

[0039] In an optional embodiment, the decision tree performs face recognition based on the input weighted fusion features, including: the decision tree searches and matches the face labels previously stored in the leaf nodes based on the input weighted fusion features to obtain the corresponding face recognition results.

[0040] It should be noted that when the face labels are stored in the leaf nodes of the decision tree, the corresponding face identifiers can be found by searching and matching in the decision tree, thereby achieving face recognition.

[0041] In another optional embodiment, the decision tree performs face recognition based on the input weighted fusion features, including: the decision tree obtains the corresponding face recognition result based on the input weighted fusion features combined with the face tag previously bound to the Bluetooth beacon device.

[0042] It should be noted that if the face identifier is not stored in the leaf node of the decision tree, and the Bluetooth feature is bound to the face tag, then when using the decision tree for behavior classification, if a node related to the Bluetooth feature is encountered, the Bluetooth feature can be used to assist in face recognition.

[0043] Furthermore, based on the input weighted fusion features and combined with the face tags previously bound to the Bluetooth beacon device, corresponding face recognition results are obtained, including: for the nodes related to the Bluetooth features in the decision tree, based on the face tags bound to the Bluetooth beacon device, determining the face candidate range; combining the face features in the weighted fusion features, searching and matching in the face candidate range to obtain the corresponding face recognition results.

[0044] To summarize, the embodiments of the present invention obtain Bluetooth beacon information and video information based on the association between the Bluetooth beacon device and the shooting device when an alarm is triggered, so as to facilitate multi-source fusion and behavior classification of the obtained Bluetooth beacon information and video information, reduce false alarms and delays, improve the efficiency and practicality of alarms, and further ensure whether abnormal behavior occurs and whether an alarm is needed, while ensuring the accuracy of the alarm, avoiding the poor accuracy and reliability of single-factor alarms.

[0045] The following is a description of the alarm device for the video fusion Bluetooth marker based on the indoor scene provided by the present invention. The alarm device for the video fusion Bluetooth marker based on the indoor scene described below and the alarm method for the video fusion Bluetooth marker based on the indoor scene described above can be referenced to each other.

[0046] Figure 3 The schematic diagram of the structure of an alarm device based on indoor scene video fusion Bluetooth beacon is shown, the device includes: The information acquisition module 31 acquires the corresponding Bluetooth beacon information and video information based on the triggering alarm behavior and the association between the previously configured Bluetooth beacon device and the shooting device; The weighted fusion module 32 obtains weighted fusion features according to the Bluetooth beacon information and the video information in combination with a preset weight distribution mechanism; The classification module 33 performs classification based on the weighted fusion features and the previously trained decision tree to obtain a classification result, and determines whether to issue an alarm based on the classification result.

[0047] In this embodiment, the information acquisition module 31 includes: an information acquisition unit, based on the triggered alarm behavior, if the alarm behavior is Bluetooth triggered, the corresponding Bluetooth beacon information is obtained, and based on the shooting device previously associated with the Bluetooth beacon device, the corresponding video information is obtained; if the alarm behavior is video triggered, the corresponding video information is obtained, and based on the Bluetooth beacon associated with the corresponding shooting device of the video, the corresponding Bluetooth information is obtained to obtain the Bluetooth beacon information.

[0048] In an optional embodiment, the device further includes: an information processing module, which calibrates and optimizes the Bluetooth beacon information after acquiring the Bluetooth beacon information to ensure the stability and accuracy of the Bluetooth beacon information.

[0049] In an optional embodiment, the device further includes: an information reporting module, which saves the video information after acquiring the video information for subsequent archiving and tracing.

[0050] In an optional embodiment, the device further includes: a position acquisition module, which acquires the position information of the shooting device and the position information of the Bluetooth beacon device before acquiring the corresponding Bluetooth beacon information and video information based on the triggering alarm behavior and the association between the previously configured Bluetooth beacon device and the shooting device; a distance determination module, which determines the distance between any shooting device and any Bluetooth beacon device according to the position information of the shooting device and the position information of the Bluetooth beacon device, and the distance includes a horizontal distance and a vertical distance; a vertical distance judgment module, which determines whether the vertical distance is less than a preset height threshold; a screening module, which screens the shooting devices and Bluetooth beacon devices that obtain the corresponding vertical distance based on the vertical distance being less than the preset height threshold, and eliminates the shooting devices and Bluetooth beacon devices corresponding to the vertical distance greater than or equal to the preset height threshold; a horizontal distance judgment module, which determines whether the horizontal distance of the screened shooting devices and Bluetooth beacon devices is less than or equal to the preset horizontal distance threshold; an association module, which associates the shooting devices and Bluetooth beacon devices with the corresponding vertical distance based on the horizontal distance of the screened shooting devices and Bluetooth beacon devices being less than or equal to the preset horizontal distance threshold.

[0051] Furthermore, the association module is also used to: if the vertical distance is greater than or equal to a preset height threshold, then the corresponding shooting device and Bluetooth beacon device with the vertical distance are not associated. Correspondingly, the association module is also used to: if the horizontal distance between the screened shooting device and the Bluetooth beacon device is greater than a preset horizontal distance threshold, then the shooting device and the Bluetooth beacon device with the horizontal distance are not associated.

[0052] In an optional embodiment, the device also includes: a three-dimensional space construction module, which scans the application scene and establishes a virtual three-dimensional space before associating the shooting device and the Bluetooth beacon device, and multiple Bluetooth beacon devices are configured in the application scene; a virtual camera configuration module, which obtains the shooting device parameters and configures the virtual shooting device in the virtual three-dimensional space according to the shooting device parameters; an association module, which associates the shooting device and the Bluetooth beacon device in the virtual three-dimensional space based on the above-mentioned association method, thereby realizing intelligent association between the Bluetooth device and the video content, facilitating the subsequent joint analysis of Bluetooth and video, and improving the accuracy and reliability of the alarm.

[0053] The weighted fusion module 32 includes: a Bluetooth feature processing unit, which extracts features from Bluetooth beacon information to obtain Bluetooth features, and determines Bluetooth alarm classification results based on the Bluetooth features; a video feature processing unit, which extracts features from video information to obtain video features, and determines video alarm classification results based on the video features; a weight allocation unit, which determines the allocation weights of corresponding Bluetooth features and corresponding video features based on the Bluetooth alarm classification results and the video alarm classification results in combination with a preset weight allocation mechanism; a weighted fusion unit, which obtains weighted fusion features based on Bluetooth features, video features, the allocation weights of corresponding Bluetooth features, and the allocation weights of corresponding video features.

[0054] In an optional embodiment, the device further includes: a feature processing module, which performs preprocessing operations such as normalization and standardization on the Bluetooth features and video features before obtaining weighted fusion features based on the Bluetooth beacon information and the video information in combination with a preset weight distribution mechanism, so as to improve the accuracy and generalization ability of the model.

[0055] The classification module 33 includes: a classification unit, which inputs the weighted fusion features into a decision tree to obtain a classification result output by the decision tree, and the classification result includes an alarm behavior classification result and a face recognition result; wherein the decision tree is trained based on training features and result labels corresponding to the training features; the decision tree performs behavior classification and face recognition based on the input weighted fusion features to obtain an alarm behavior classification result and a face recognition result.

[0056] Specifically, before training the decision tree, the decision tree model is constructed using the training features, and an appropriate decision tree algorithm is selected for training and optimization, such as adjusting the depth of the tree, pruning, etc., to avoid overfitting, which can be set according to the actual design. Each node of the decision tree represents a feature. During the application process, it is divided into different branches according to the value of the weighted fusion feature. In addition, in the branches of the decision tree, further divisions can be made according to the weighted fusion features to achieve behavior classification.

[0057] In an optional embodiment, the classification module 33 is further used for: the decision tree searches and matches the face labels previously stored in the leaf nodes based on the input weighted fusion features to obtain the corresponding face recognition results.

[0058] In another optional embodiment, the classification module 33 is further used to: the decision tree is based on the weighted fusion features of the input, combined with the face tag previously bound to the Bluetooth beacon device, to obtain the corresponding face recognition result.

[0059] Furthermore, the classification module 33 includes: a range screening unit, which determines the face candidate range for the nodes related to the Bluetooth features in the decision tree based on the face tags bound to the Bluetooth beacon device; a face recognition unit, which searches and matches in the face candidate range in combination with the face features in the weighted fusion features to obtain the corresponding face recognition results.

[0060] To summarize, in the embodiment of the present invention, when an alarm is triggered, the information acquisition module obtains Bluetooth beacon information and video information based on the association between the Bluetooth beacon device and the shooting device, so that the weighted fusion module can perform multi-source fusion on the obtained Bluetooth beacon information and video information, and use the classification module to perform behavior classification, thereby reducing false alarms and delays, and improving the efficiency and practicality of alarms, thereby ensuring whether abnormal behavior occurs and whether an alarm is needed, while ensuring the accuracy of the alarm, avoiding the situation where single-factor alarms have poor accuracy and reliability.

[0061] Figure 4 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 4 As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430 and a communication bus 440, wherein the processor 410, the communication interface 420 and the memory 430 communicate with each other through the communication bus 440. The processor 410 may call the logic instructions in the memory 430 to execute the alarm method based on the video fusion Bluetooth beacon of the indoor scene, the method comprising: based on the triggering alarm behavior and the association between the previously configured Bluetooth beacon device and the shooting device, obtaining the corresponding Bluetooth beacon information and video information; according to the Bluetooth beacon information and the video information, combined with the preset weight distribution mechanism, obtaining the weighted fusion feature; according to the weighted fusion feature, using the previously trained decision tree for classification, obtaining the classification result, and determining whether to alarm according to the classification result.

[0062] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0063] On the other hand, the present invention also provides a computer program product, which includes a computer program, and the computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the alarm method of the video fusion Bluetooth beacon based on the indoor scene provided by the above methods, and the method includes: based on the triggering alarm behavior and the association between the previously configured Bluetooth beacon device and the shooting device, obtaining the corresponding Bluetooth beacon information and video information; according to the Bluetooth beacon information and video information, combined with a preset weight distribution mechanism, obtaining weighted fusion features; according to the weighted fusion features, using a previously trained decision tree for classification to obtain a classification result, and determining whether to alarm based on the classification result.

[0064] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it is implemented to execute the alarm method of the video-fused Bluetooth beacon based on the indoor scene provided by the above-mentioned methods. The method includes: based on the triggering alarm behavior and the association between the previously configured Bluetooth beacon device and the shooting device, obtaining the corresponding Bluetooth beacon information and video information; according to the Bluetooth beacon information and the video information, combined with a preset weight distribution mechanism, obtaining weighted fusion features; according to the weighted fusion features, using a previously trained decision tree for classification to obtain a classification result, and determining whether to alarm based on the classification result.

[0065] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0066] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0067] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An alarm method based on indoor scene video fusion Bluetooth beacon, characterized in that: include: Based on the triggering alarm behavior and the association between the previously configured Bluetooth beacon device and the shooting device, the corresponding Bluetooth beacon information and video information are obtained; According to the Bluetooth beacon information and the video information, combined with a preset weight distribution mechanism, a weighted fusion feature is obtained; According to the weighted fusion features, a previously trained decision tree is used for classification to obtain a classification result, and according to the classification result, it is determined whether to issue an alarm.

2. The alarm method based on indoor scene video fusion Bluetooth beacon according to claim 1 is characterized in that: Based on the triggering alarm behavior and the association between the previously configured Bluetooth beacon device and the shooting device, the corresponding Bluetooth beacon information and video information are obtained, including: Based on the triggering alarm behavior, if the alarm behavior is triggered by Bluetooth, obtaining corresponding Bluetooth beacon information, and obtaining corresponding video information based on a shooting device previously associated with the Bluetooth beacon device; If the alarm behavior is triggered by a video, the corresponding video information is obtained, and based on the Bluetooth beacon associated with the corresponding shooting device of the video, the corresponding Bluetooth information is obtained to obtain the Bluetooth beacon information.

3. The alarm method based on indoor scene video fusion Bluetooth beacon according to claim 1 is characterized in that: Before obtaining corresponding Bluetooth beacon information and video information based on the triggering alarm behavior and the association between the previously configured Bluetooth beacon device and the shooting device, the method includes: Get the location information of the camera device and the Bluetooth beacon device; Determine the distance between any shooting device and any Bluetooth beacon device according to the shooting device location information and the Bluetooth beacon device location information, wherein the distance includes a horizontal distance and a vertical distance; Determining whether the vertical distance is less than a preset height threshold; Based on the fact that the vertical distance is less than a preset height threshold, the shooting devices and Bluetooth beacon devices that obtain the corresponding vertical distances are screened, and the shooting devices and Bluetooth beacon devices corresponding to the vertical distances greater than or equal to the preset height threshold are eliminated; Determine whether the horizontal distance between the screened camera device and the Bluetooth beacon device is less than or equal to a preset horizontal distance threshold; Based on the fact that the horizontal distance between the filtered shooting devices and the Bluetooth beacon devices is less than or equal to a preset horizontal distance threshold, the shooting devices and the Bluetooth beacon devices of the corresponding vertical distances are associated.

4. The alarm method based on indoor scene video fusion Bluetooth beacon according to claim 1 is characterized in that: According to the Bluetooth beacon information and the video information, combined with a preset weight distribution mechanism, a weighted fusion feature is obtained, including: Extracting features from the Bluetooth beacon information to obtain Bluetooth features, and determining a Bluetooth alarm classification result based on the Bluetooth features; Extracting features from the video information to obtain video features, and determining a video alarm classification result based on the video features; According to the Bluetooth alarm classification result and the video alarm classification result, combined with a preset weight allocation mechanism, determine the allocation weight of the corresponding Bluetooth feature and the allocation weight of the corresponding video feature; A weighted fusion feature is obtained according to the Bluetooth feature, the video feature, the allocation weight corresponding to the Bluetooth feature, and the allocation weight corresponding to the video feature.

5. The alarm method based on indoor scene video fusion Bluetooth beacon according to claim 4 is characterized in that: According to the weighted fusion features, the previously trained decision tree is used for classification to obtain classification results, including: Input the weighted fusion features into a decision tree to obtain a classification result output by the decision tree, wherein the classification result includes an alarm behavior classification result and a face recognition result; wherein the decision tree is trained based on training features and result labels corresponding to the training features; The decision tree performs behavior classification and face recognition based on the input weighted fusion features to obtain alarm behavior classification results and face recognition results.

6. The alarm method based on indoor scene video fusion Bluetooth beacon according to claim 5 is characterized in that: The decision tree performs face recognition based on the input weighted fusion features, including: The decision tree searches and matches the face labels previously stored in the leaf nodes based on the input weighted fusion features to obtain the corresponding face recognition results; or, The decision tree is based on the input weighted fusion features and combined with the face tag previously bound to the Bluetooth beacon device to obtain the corresponding face recognition result.

7. An alarm device based on indoor scene video fusion Bluetooth beacon, characterized in that: include: An information acquisition module, based on the triggering alarm behavior and the association between the previously configured Bluetooth beacon device and the shooting device, acquires the corresponding Bluetooth beacon information and video information; A weighted fusion module, which obtains weighted fusion features according to the Bluetooth beacon information and the video information in combination with a preset weight distribution mechanism; The classification module performs classification based on the weighted fusion features and a previously trained decision tree to obtain a classification result, and determines whether to issue an alarm based on the classification result.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the alarm method of the video fusion Bluetooth beacon based on the indoor scene is implemented as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the alarm method of the video fusion Bluetooth beacon based on indoor scenes as described in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the alarm method of the video fusion Bluetooth beacon based on indoor scenes as described in any one of claims 1 to 6 is implemented.

Images