Public security monitoring method and device based on intelligent street lamp
Through the sound wave acquisition of intelligent street lights and multiple algorithms, precise monitoring of urban security is achieved, and the problem that existing technology cannot achieve real-time and precise public security management is solved, and the level of urban security is improved.
Patent Information
- Application Number
- CN202211106571.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2022-09-12
- Publication Date
- 2024-11-19
AI Technical Summary
The existing public security monitoring technology cannot achieve real-time and precise urban security management, and mainly relies on cameras and manpower, and cannot effectively improve the safety level of the city.
The sound wave acquisition technology based on intelligent street lights is adopted to obtain sound wave information through multiple intelligent street lights, and combined with voice analysis, probability prediction algorithms and three-dimensional modeling, we judge whether there is a public security incident in the target area and perform corresponding public security maintenance operations.
It realizes more accurate and intelligent security monitoring, improves the safety level of the city, and can quickly respond to and handle public security incidents.
Smart Images

Figure CN118982894A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data processing, and in particular to a public security monitoring method and device based on intelligent street lamps. Background Art
[0002] With the acceleration of urbanization and the increase in urban population density, urban public security issues have become more and more important. Many companies or government agencies have begun to seek the help of algorithm technology to improve the level of urban public security management. However, the existing public security monitoring technology is generally limited to video acquisition through commonly installed cameras and relies on the manpower of the masses or the public security system. It is impossible to achieve more real-time and accurate public security monitoring. It can be seen that the existing technology has defects that need to be solved urgently. Summary of the invention
[0003] The technical problem to be solved by the present invention is to provide a public security monitoring method and device based on intelligent street lamps, which can achieve more accurate and intelligent public security monitoring based on multiple intelligent algorithms and the sound wave collection function of intelligent street lamps, thereby improving the safety level of the city.
[0004] In order to solve the above technical problems, the first aspect of the present invention discloses a public security monitoring method based on smart street lamps, the method comprising:
[0005] Acquire multiple sound wave collection information from multiple smart street lamps; the smart street lamps are arranged on streets in a target area; the smart street lamps are provided with sound wave collection equipment and playback equipment;
[0006] Determining regional sound information of the target area based on the plurality of sound wave collection information;
[0007] Determine the voice text and voice type corresponding to the regional voice information based on the voice analysis algorithm model;
[0008] Based on the probability prediction algorithm model, the probability of a public security incident occurring in the target area is determined according to the sound text and the sound type;
[0009] Determine whether the probability of a public security incident occurring in the target area is greater than a preset probability threshold. If the judgment result is yes, perform a public security maintenance operation; the public security maintenance operation includes sending a public security maintenance notification to the corresponding public security terminal and sending a public security alarm information to the playback device of the smart street lamp for playback.
[0010] As an optional implementation, in the first aspect of the present invention, determining the regional sound information of the target area based on the multiple sound wave collection information includes:
[0011] Processing the plurality of sound wave collection information by a noise suppression algorithm to obtain a plurality of noise-removed sound wave information;
[0012] After aligning the sound wave data of the plurality of noise-removed sound wave information according to the same time dimension, calculating the sound wave similarity of any two sound wave data of the noise-removed sound wave information in any time period;
[0013] For any synthesis time period, mixing and synthesizing the sound wave data of at least two noise-removed sound wave information whose sound wave similarity in the synthesis time period is greater than a preset first similarity threshold, to obtain synthesized sound wave information corresponding to the synthesis time period;
[0014] The synthesized sound wave information corresponding to all the synthesized time periods is combined to obtain the regional sound information of the target area.
[0015] As an optional implementation, in the first aspect of the present invention, determining the sound text and sound type corresponding to the regional sound information based on the speech analysis algorithm model includes:
[0016] Determining, based on a sound classification algorithm model, a portion containing dialogue speech and a portion containing non-dialogue speech in the regional sound information;
[0017] Determine the audio text corresponding to the dialogue speech part based on the speech recognition algorithm model;
[0018] Based on a sound template matching algorithm, the sound type corresponding to the non-dialogue speech part is determined.
[0019] As an optional implementation, in the first aspect of the present invention, the determining the dialogue speech part and the non-dialogue speech part in the regional sound information based on the sound classification algorithm model includes:
[0020] The regional sound information is input into a trained sound classification neural network model to determine the dialogue speech part and the non-dialogue speech part in the regional sound information; the sound classification neural network model is trained by a training data set including a plurality of training sound information and corresponding sound type labels; the sound type labels include dialogue speech labels and non-dialogue speech labels.
[0021] As an optional implementation, in the first aspect of the present invention, determining the sound type corresponding to the non-dialogue speech part based on the sound template matching algorithm includes:
[0022] Calculate the data similarity between the non-dialogue speech part and multiple public security sound templates, and determine the type of the public security sound template with the highest data similarity as the sound type of the non-dialogue speech part; the type of the public security sound template or the sound type includes at least one of the sound of a vehicle collision, the sound of hitting glass, gunshots, human shouting and human groaning.
[0023] As an optional implementation, in the first aspect of the present invention, the probability of a public security incident occurring in the target area is determined based on the probability prediction algorithm model according to the sound text and the sound type, including:
[0024] The sound text and sound type are input into a trained public security prediction neural network model to determine the probability of a public security incident occurring in the target area; the public security prediction neural network model is trained using a training data set including multiple training sound texts, training sound types and corresponding public security incident labels.
[0025] As an optional implementation, in the first aspect of the present invention, the smart street lamp is further provided with an image acquisition device; and the method further comprises:
[0026] If the judgment result is yes, an image acquisition instruction is sent to the image acquisition device of the multiple smart street lamps in the target area that meet the three-dimensional position condition; the three-dimensional position condition is that at least two of the connecting lines between the positions of the multiple smart street lamps are not parallel to each other;
[0027] Receiving a plurality of image acquisition information acquired by the plurality of intelligent street lamps in response to the image acquisition instruction;
[0028] Based on the three-dimensional modeling algorithm, a regional three-dimensional model of the target area is established according to the plurality of image acquisition information;
[0029] Calculating the geometric similarity between the regional three-dimensional model and a plurality of public security three-dimensional templates;
[0030] When the geometric similarity between the regional three-dimensional model and at least one of the public security three-dimensional templates is greater than a preset second similarity threshold, it is determined whether a public security incident occurs in the target area.
[0031] As an optional embodiment, in the first aspect of the present invention, the method further comprises:
[0032] Obtaining public security incident processing information fed back by the public security terminal;
[0033] Determine the information type of the public security incident processing information;
[0034] When the information type is that no public security incident has occurred, increasing the probability threshold;
[0035] When the information type is a public security incident, the probability threshold is lowered.
[0036] The second aspect of the present invention discloses a public security monitoring device based on an intelligent street lamp, which comprises:
[0037] A sound wave acquisition module, used to acquire a plurality of sound wave collection information from a plurality of smart street lamps; the smart street lamps are arranged on streets in a target area; the smart street lamps are provided with a sound wave collection device and a playback device;
[0038] A first determination module, configured to determine regional sound information of the target area based on the plurality of sound wave collection information;
[0039] A second determination module is used to determine the sound text and sound type corresponding to the regional sound information based on the speech analysis algorithm model;
[0040] A third determination module is used to determine the probability of a public security incident occurring in the target area based on the probability prediction algorithm model and according to the sound text and the sound type;
[0041] The public security operation module is used to determine whether the probability of a public security incident occurring in the target area is greater than a preset probability threshold. If so, a public security maintenance operation is performed; the public security maintenance operation includes sending a public security maintenance notification to the corresponding public security terminal and sending a public security alarm information to the playback device of the smart street lamp for playback.
[0042] As an optional implementation, in the second aspect of the present invention, the specific manner in which the first determination module determines the regional sound information of the target area based on the multiple sound wave collection information includes:
[0043] Processing the plurality of sound wave collection information by a noise suppression algorithm to obtain a plurality of noise-removed sound wave information;
[0044] After aligning the sound wave data of the plurality of noise-removed sound wave information according to the same time dimension, calculating the sound wave similarity of any two sound wave data of the noise-removed sound wave information in any time period;
[0045] For any synthesis time period, mixing and synthesizing the sound wave data of at least two noise-removed sound wave information whose sound wave similarity in the synthesis time period is greater than a preset first similarity threshold, to obtain synthesized sound wave information corresponding to the synthesis time period;
[0046] The synthesized sound wave information corresponding to all the synthesized time periods is combined to obtain the regional sound information of the target area.
[0047] As an optional implementation, in the second aspect of the present invention, the second determination module determines the specific manner of the sound text and sound type corresponding to the regional sound information based on the speech analysis algorithm model, including:
[0048] Determining, based on a sound classification algorithm model, a portion containing dialogue speech and a portion containing non-dialogue speech in the regional sound information;
[0049] Determine the audio text corresponding to the dialogue speech part based on the speech recognition algorithm model;
[0050] Based on a sound template matching algorithm, the sound type corresponding to the non-dialogue speech part is determined.
[0051] As an optional implementation, in the second aspect of the present invention, the second determination module determines the specific manner of the dialogue voice part and the non-dialogue voice part in the regional sound information based on the sound classification algorithm model, including:
[0052] The regional sound information is input into a trained sound classification neural network model to determine the dialogue speech part and the non-dialogue speech part in the regional sound information; the sound classification neural network model is trained by a training data set including a plurality of training sound information and corresponding sound type labels; the sound type labels include dialogue speech labels and non-dialogue speech labels.
[0053] As an optional implementation, in the second aspect of the present invention, the specific manner in which the second determination module determines the sound type corresponding to the non-dialogue speech portion based on a sound template matching algorithm includes:
[0054] Calculate the data similarity between the non-dialogue speech part and multiple public security sound templates, and determine the type of the public security sound template with the highest data similarity as the sound type of the non-dialogue speech part; the type of the public security sound template or the sound type includes at least one of the sound of a vehicle collision, the sound of hitting glass, gunshots, human shouting and human groaning.
[0055] As an optional implementation, in the second aspect of the present invention, the third determination module determines the specific manner of the probability of a public security incident occurring in the target area based on the probability prediction algorithm model and according to the sound text and the sound type, including:
[0056] The sound text and sound type are input into a trained public security prediction neural network model to determine the probability of a public security incident occurring in the target area; the public security prediction neural network model is trained using a training data set including multiple training sound texts, training sound types and corresponding public security incident labels.
[0057] As an optional implementation, in the second aspect of the present invention, the intelligent street lamp is further provided with an image acquisition device; the device further comprises a three-dimensional judgment module for performing the following steps:
[0058] If the judgment result is yes, an image acquisition instruction is sent to the image acquisition device of the multiple smart street lamps in the target area that meet the three-dimensional position condition; the three-dimensional position condition is that at least two of the connecting lines between the positions of the multiple smart street lamps are not parallel to each other;
[0059] Receiving a plurality of image acquisition information acquired by the plurality of intelligent street lamps in response to the image acquisition instruction;
[0060] Based on the three-dimensional modeling algorithm, a regional three-dimensional model of the target area is established according to the plurality of image acquisition information;
[0061] Calculating the geometric similarity between the regional three-dimensional model and a plurality of public security three-dimensional templates;
[0062] When the geometric similarity between the regional three-dimensional model and at least one of the public security three-dimensional templates is greater than a preset second similarity threshold, it is determined whether a public security incident occurs in the target area.
[0063] As an optional implementation, in the second aspect of the present invention, the device further includes a threshold adjustment module, configured to perform the following steps:
[0064] Obtaining public security incident processing information fed back by the public security terminal;
[0065] Determine the information type of the public security incident processing information;
[0066] When the information type is that no public security incident has occurred, increasing the probability threshold;
[0067] When the information type is a public security incident, the probability threshold is lowered.
[0068] The third aspect of the present invention discloses another public security monitoring device based on an intelligent street lamp, the device comprising:
[0069] A memory storing executable program code;
[0070] a processor coupled to the memory;
[0071] The processor calls the executable program code stored in the memory to execute part or all of the steps in the public security monitoring method based on smart street lamps disclosed in the first aspect of the embodiment of the present invention.
[0072] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:
[0073] In an embodiment of the present invention, a plurality of sound wave collection information from a plurality of intelligent street lamps is obtained; the intelligent street lamps are arranged on streets in a target area; the intelligent street lamps are provided with sound wave collection equipment and playback equipment; based on the plurality of sound wave collection information, the regional sound information of the target area is determined; based on a speech analysis algorithm model, the sound text and sound type corresponding to the regional sound information are determined; based on a probability prediction algorithm model, the probability of a public security incident occurring in the target area is determined according to the sound text and sound type; it is determined whether the probability of a public security incident occurring in the target area is greater than a preset probability threshold, and if so, a public security maintenance operation is performed; the public security maintenance operation includes sending a public security maintenance notification to a corresponding public security terminal and sending a public security alarm information to the playback equipment of the intelligent street lamp. It can be seen that the present invention can realize more accurate and more intelligent public security monitoring based on a variety of intelligent algorithms and based on the sound wave collection function of intelligent street lamps, thereby improving the safety level of the city. BRIEF DESCRIPTION OF THE DRAWINGS
[0074] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0075] Figure 1 It is a flow chart of a public security monitoring method based on intelligent street lamps disclosed in an embodiment of the present invention;
[0076] Figure 2 It is a structural schematic diagram of a public security monitoring device based on an intelligent street lamp disclosed in an embodiment of the present invention;
[0077] Figure 3 It is a structural schematic diagram of another public security monitoring device based on intelligent street lamps disclosed in an embodiment of the present invention. DETAILED DESCRIPTION
[0078] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only 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.
[0079] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or end including a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units inherent to these processes, methods, products or ends.
[0080] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.
[0081] The present invention discloses a public security monitoring method and device based on intelligent street lamps, which can realize more accurate and intelligent public security monitoring based on multiple intelligent algorithms and the sound wave collection function of intelligent street lamps, thereby improving the safety level of the city. The following are detailed descriptions.
[0082] Embodiment 1
[0083] See also Figure 1 , Figure 1 FIG. 1 is a flow chart of a public security monitoring method based on intelligent street lamps disclosed in an embodiment of the present invention. Figure 1 As shown, the public security monitoring method based on smart street lamps may include the following operations:
[0084] 101. Acquire multiple sound wave collection information from multiple smart street lamps.
[0085] Optionally, the smart street lamp is set on the street of the target area. Optionally, the target area can be an area in the city, such as a living area, a business circle or a scenic area. Optionally, the smart street lamp is provided with a sound wave collection device and a playback device, wherein the sound wave collection device can adopt a sound collection device such as a microphone, and the playback device can adopt a device such as a speaker. Optionally, the smart street lamp can be an existing ordinary street lamp that has been modified, and the modification can be completed by connecting the sound wave collection device and the playback device to the controller of the street lamp or directly installing them on the ordinary street lamp. Optionally, the smart street lamp should have a wireless communication function so that it can upload the collected data to the server or receive commands from the server.
[0086] 102. Determine regional sound information of the target area based on the information collected from the multiple sound waves.
[0087] 103. Based on the speech analysis algorithm model, determine the sound text and sound type corresponding to the regional sound information.
[0088] 104. Based on the probability prediction algorithm model, determine the probability of public security incidents occurring in the target area according to the sound text and sound type;.
[0089] 105. Determine whether the probability of a public security incident occurring in the target area is greater than a preset probability threshold. If the determination result is yes, perform a public security maintenance operation.
[0090] Optionally, the public security maintenance operation includes sending a public security maintenance notification to a corresponding public security terminal and sending a public security alarm message to a playback device of the smart street lamp for playback.
[0091] It can be seen that the embodiments of the present invention can achieve more accurate and intelligent public security monitoring based on a variety of intelligent algorithms and the sound wave collection function of intelligent street lamps, thereby improving the safety level of the city.
[0092] As an optional implementation, in the above step 102, determining the regional sound information of the target area based on the multiple sound wave collection information includes:
[0093] The multiple sound wave collection information is processed by a noise suppression algorithm to obtain multiple noise-removed wave information;
[0094] After aligning the sound wave data of multiple noise-removed sound wave information according to the same time dimension, calculate the sound wave similarity of any two sound wave data of noise-removed sound wave information in any time period;
[0095] For any synthesis time period, mixing and synthesizing the sound wave data of at least two noise-removed sound wave information whose sound wave similarity in the synthesis time period is greater than a preset first similarity threshold, to obtain synthesized sound wave information corresponding to the synthesis time period;
[0096] The synthesized sound wave information corresponding to all the synthesized time periods is combined to obtain the regional sound information of the target area.
[0097] Through the above settings, the noise in the sound wave information can be removed first, and then the common sound of the intersection of multiple sound wave information can be screened and synthesized, so as to obtain a more accurate and realistic regional sound of the target area, so as to facilitate a more accurate judgment on whether a public security incident occurs in the area in the future.
[0098] As an optional implementation, in the above step 103, determining the sound text and sound type corresponding to the regional sound information based on the speech analysis algorithm model includes:
[0099] Based on the sound classification algorithm model, determine the conversational speech part and the non-conversational speech part in the regional sound information;
[0100] Based on the speech recognition algorithm model, determine the sound text corresponding to the dialogue speech part;
[0101] Based on the sound template matching algorithm, the sound type corresponding to the non-dialogue speech part is determined.
[0102] Optionally, in the above steps, determining the dialogue voice part and the non-dialogue voice part in the regional sound information based on the sound classification algorithm model includes:
[0103] The regional sound information is input into the trained sound classification neural network model to determine the dialogue speech part and the non-dialogue speech part in the regional sound information.
[0104] Specifically, the sound classification neural network model is trained by a training data set including a plurality of training sound information and corresponding sound type labels, wherein the sound type labels include dialogue speech labels and non-dialogue speech labels.
[0105] Optionally, the sound classification neural network model may include a sound feature extraction network, a sound feature reconstruction network, and a sound classification network. Optionally, a training model including a sound classification neural network model, a loss function calculation module, and a parameter optimization module is determined, and the training data of the training data set is continuously input into the training model. The loss function calculation module calculates the cross entropy loss function value between the prediction result output by the sound classification neural network model and the training data, and the parameter optimization module continuously optimizes the model parameters of the training model based on the gradient descent method, so that the cross entropy loss function value converges until the training is completed, so that a trained sound classification neural network model can be obtained.
[0106] Optionally, in the above step, determining the sound type corresponding to the non-dialogue speech part based on the sound template matching algorithm includes:
[0107] The data similarity between the non-dialogue speech part and a plurality of security-targeted sound templates is calculated, and the type of the security-targeted sound template with the highest data similarity is determined as the sound type of the non-dialogue speech part.
[0108] Optionally, the type or sound type of the public security sound template includes at least one of the sounds of vehicle collision, hitting glass, gunshot, human shouting and human groaning. Optionally, the method for determining any one of the public security sound templates can be to obtain multiple sound samples of the type, calculate the sound features corresponding to the multiple sound samples, such as the feature intersection between the Mel frequency cepstral coefficients, and determine the feature intersection as the public security sound template corresponding to the type. Similarly, when calculating the data similarity between the non-dialogue speech part and multiple public security sound templates, it is necessary to first determine the sound features of the non-dialogue speech part, and then calculate the feature similarity between the sound features and the sound template to obtain the data similarity.
[0109] As an optional implementation, in the above step 104, based on the probability prediction algorithm model, according to the sound text and the sound type, determining the probability of a public security incident occurring in the target area includes:
[0110] The sound text and sound type are input into the trained public security prediction neural network model to determine the probability of public security incidents occurring in the target area.
[0111] Specifically, the public security prediction neural network model is trained by a training data set including multiple training sound texts, training sound types and corresponding public security event labels. Optionally, the public security prediction neural network model may include a feature extraction network, a feature reconstruction network and a classification network. Optionally, a training model including a public security prediction neural network model, a loss function calculation module and a parameter optimization module is determined, and the training data of the training data set is continuously input into the training model. The cross entropy loss function value between the prediction result output by the public security prediction neural network model and the training data is calculated by the loss function calculation module, and the model parameters of the training model are continuously optimized by the parameter optimization module based on the gradient descent method, so that the cross entropy loss function value converges until the training is completed, so that a trained public security prediction neural network model can be obtained.
[0112] As an optional implementation, the smart street lamp is further provided with an image acquisition device, which may be a camera device. Accordingly, the method further includes:
[0113] If the judgment result is yes, an image acquisition instruction is sent to image acquisition devices of multiple smart street lamps in the target area that meet the three-dimensional position conditions;
[0114] Receiving a plurality of image acquisition information acquired by a plurality of intelligent street lamps in response to an image acquisition instruction;
[0115] Based on the 3D modeling algorithm, a regional 3D model of the target area is established according to the information collected from multiple images;
[0116] Calculate the geometric similarity between the regional 3D model and multiple security 3D templates;
[0117] When the geometric similarity between the regional three-dimensional model and at least one public security three-dimensional template is greater than a preset second similarity threshold, it is determined whether a public security incident occurs in the target area.
[0118] Optionally, the three-dimensional position condition is that at least two of the lines connecting the positions of the multiple smart street lights are not parallel to each other. Through such a setting, the multiple smart street lights have formed a relatively three-dimensional positional relationship in terms of position, so that the images acquired by each of them can capture the target area from multiple angles, thereby greatly improving the success rate of subsequently establishing a regional three-dimensional model of the target area based on multiple image acquisition information.
[0119] As an optional implementation, the method further includes:
[0120] Obtaining public security incident processing information fed back by public security terminals;
[0121] Determine the information type of public security incident handling information;
[0122] When the information type is that no public security incident has occurred, the probability threshold is increased;
[0123] When the information type is a public security incident, lower the probability threshold.
[0124] Through the above settings, the success or failure of the judgment of this public security monitoring method can be determined according to the actual processing results, and the probability threshold can be lowered when it is successful to improve the sensitivity, so that the next public security monitoring method can more easily judge the public security incident. In the event of failure, that is, when the prediction fails, the probability threshold is increased to avoid the system being too sensitive due to a too low probability threshold, thereby playing a role in adjusting the system sensitivity.
[0125] Embodiment 2
[0126] See also Figure 2 , Figure 2 : is a schematic diagram of a public security monitoring device based on an intelligent street lamp disclosed in an embodiment of the present invention. Figure 2 The described device can be applied to a corresponding financial data accounting terminal, a financial data accounting device or a financial data accounting server, and the server can be a local server or a cloud server, which is not limited in the embodiment of the present invention. Figure 2 As shown, the device may include:
[0127] The sound wave acquisition module 201 is used to acquire a plurality of sound wave collection information from a plurality of smart street lamps.
[0128] Optionally, the smart street lamp is set on the street of the target area. Optionally, the target area can be an area in the city, such as a living area, a business circle or a scenic area. Optionally, the smart street lamp is provided with a sound wave collection device and a playback device, wherein the sound wave collection device can adopt a sound collection device such as a microphone, and the playback device can adopt a device such as a speaker. Optionally, the smart street lamp can be an existing ordinary street lamp that has been modified, and the modification can be completed by connecting the sound wave collection device and the playback device to the controller of the street lamp or directly installing them on the ordinary street lamp. Optionally, the smart street lamp should have a wireless communication function so that it can upload the collected data to the server or receive commands from the server.
[0129] The first determination module 202 is used to determine the regional sound information of the target area based on the multiple sound wave collection information.
[0130] The second determination module 203 is used to determine the sound text and sound type corresponding to the regional sound information based on the speech analysis algorithm model.
[0131] The third determination module 204 is used to determine the probability of a public security incident occurring in the target area based on the probability prediction algorithm model according to the sound text and the sound type.
[0132] The public security operation module 205 is used to determine whether the probability of a public security incident occurring in the target area is greater than a preset probability threshold. If the determination result is yes, a public security maintenance operation is performed.
[0133] Optionally, the public security maintenance operation includes sending a public security maintenance notification to a corresponding public security terminal and sending a public security alarm message to a playback device of the smart street lamp for playback.
[0134] It can be seen that the embodiments of the present invention can achieve more accurate and intelligent public security monitoring based on a variety of intelligent algorithms and the sound wave collection function of intelligent street lamps, thereby improving the safety level of the city.
[0135] As an optional implementation manner, the first determination module 202 determines the specific manner of the regional sound information of the target area based on the multiple sound wave collection information, including:
[0136] The multiple sound wave collection information is processed by a noise suppression algorithm to obtain multiple noise-removed wave information;
[0137] After aligning the sound wave data of multiple noise-removed sound wave information according to the same time dimension, calculate the sound wave similarity of any two sound wave data of noise-removed sound wave information in any time period;
[0138] For any synthesis time period, mixing and synthesizing the sound wave data of at least two noise-removed sound wave information whose sound wave similarity in the synthesis time period is greater than a preset first similarity threshold, to obtain synthesized sound wave information corresponding to the synthesis time period;
[0139] The synthesized sound wave information corresponding to all the synthesized time periods is combined to obtain the regional sound information of the target area.
[0140] Through the above settings, the noise in the sound wave information can be removed first, and then the common sound of the intersection of multiple sound wave information can be screened and synthesized, so as to obtain a more accurate and realistic regional sound of the target area, so as to facilitate a more accurate judgment on whether a public security incident occurs in the area in the future.
[0141] As an optional implementation, the second determination module 203 determines the specific manner of the sound text and sound type corresponding to the regional sound information based on the speech analysis algorithm model, including:
[0142] Based on the sound classification algorithm model, determine the conversational speech part and the non-conversational speech part in the regional sound information;
[0143] Based on the speech recognition algorithm model, determine the sound text corresponding to the dialogue speech part;
[0144] Based on the sound template matching algorithm, the sound type corresponding to the non-dialogue speech part is determined.
[0145] As an optional implementation, the second determination module 203 determines the specific manner of including the dialogue voice part and the non-dialogue voice part in the regional sound information based on the sound classification algorithm model, including:
[0146] The regional sound information is input into the trained sound classification neural network model to determine the dialogue speech part and the non-dialogue speech part in the regional sound information.
[0147] Specifically, the sound classification neural network model is trained by a training data set including a plurality of training sound information and corresponding sound type labels, wherein the sound type labels include dialogue speech labels and non-dialogue speech labels.
[0148] Optionally, the sound classification neural network model may include a sound feature extraction network, a sound feature reconstruction network, and a sound classification network. Optionally, a training model including a sound classification neural network model, a loss function calculation module, and a parameter optimization module is determined, and the training data of the training data set is continuously input into the training model. The loss function calculation module calculates the cross entropy loss function value between the prediction result output by the sound classification neural network model and the training data, and the parameter optimization module continuously optimizes the model parameters of the training model based on the gradient descent method, so that the cross entropy loss function value converges until the training is completed, so that a trained sound classification neural network model can be obtained.
[0149] As an optional implementation, the second determination module 203 determines the specific manner of the sound type corresponding to the non-dialogue speech part based on the sound template matching algorithm, including:
[0150] The data similarity between the non-dialogue speech part and a plurality of security-targeted sound templates is calculated, and the type of the security-targeted sound template with the highest data similarity is determined as the sound type of the non-dialogue speech part.
[0151] Optionally, the type or sound type of the public security sound template includes at least one of the sounds of vehicle collision, hitting glass, gunshot, human shouting and human groaning. Optionally, the method for determining any one of the public security sound templates can be to obtain multiple sound samples of the type, calculate the sound features corresponding to the multiple sound samples, such as the feature intersection between the Mel frequency cepstral coefficients, and determine the feature intersection as the public security sound template corresponding to the type. Similarly, when calculating the data similarity between the non-dialogue speech part and multiple public security sound templates, it is necessary to first determine the sound features of the non-dialogue speech part, and then calculate the feature similarity between the sound features and the sound template to obtain the data similarity.
[0152] As an optional implementation, the third determination module 204 determines the specific manner of determining the probability of a public security incident occurring in the target area based on the probability prediction algorithm model according to the sound text and the sound type, including:
[0153] The sound text and sound type are input into the trained public security prediction neural network model to determine the probability of public security incidents occurring in the target area.
[0154] Specifically, the public security prediction neural network model is trained by a training data set including multiple training sound texts, training sound types and corresponding public security event labels. Optionally, the public security prediction neural network model may include a feature extraction network, a feature reconstruction network and a classification network. Optionally, a training model including a public security prediction neural network model, a loss function calculation module and a parameter optimization module is determined, and the training data of the training data set is continuously input into the training model. The cross entropy loss function value between the prediction result output by the public security prediction neural network model and the training data is calculated by the loss function calculation module, and the model parameters of the training model are continuously optimized by the parameter optimization module based on the gradient descent method, so that the cross entropy loss function value converges until the training is completed, so that a trained public security prediction neural network model can be obtained.
[0155] As an optional implementation, the smart street lamp is further provided with an image acquisition device, which may be a camera device. Accordingly, the device further includes a three-dimensional judgment module for performing the following steps:
[0156] If the judgment result is yes, an image acquisition instruction is sent to image acquisition devices of multiple smart street lamps in the target area that meet the three-dimensional position conditions;
[0157] Receiving a plurality of image acquisition information acquired by a plurality of intelligent street lamps in response to an image acquisition instruction;
[0158] Based on the 3D modeling algorithm, a regional 3D model of the target area is established according to the information collected from multiple images;
[0159] Calculate the geometric similarity between the regional 3D model and multiple security 3D templates;
[0160] When the geometric similarity between the regional three-dimensional model and at least one public security three-dimensional template is greater than a preset second similarity threshold, it is determined whether a public security incident occurs in the target area.
[0161] Optionally, the three-dimensional position condition is that at least two of the lines connecting the positions of the multiple smart street lights are not parallel to each other. Through such a setting, the multiple smart street lights have formed a relatively three-dimensional positional relationship in terms of position, so that the images acquired by each of them can capture the target area from multiple angles, thereby greatly improving the success rate of subsequently establishing a regional three-dimensional model of the target area based on multiple image acquisition information.
[0162] As an optional implementation, the device further includes a threshold adjustment module, configured to perform the following steps:
[0163] Obtaining public security incident processing information fed back by public security terminals;
[0164] Determine the information type of public security incident handling information;
[0165] When the information type is that no public security incident has occurred, the probability threshold is increased;
[0166] When the information type is a public security incident, lower the probability threshold.
[0167] Through the above settings, the success or failure of the judgment of this public security monitoring method can be determined according to the actual processing results, and the probability threshold can be lowered when it is successful to improve the sensitivity, so that the next public security monitoring method can more easily judge the public security incident. In the event of failure, that is, when the prediction fails, the probability threshold is increased to avoid the system being too sensitive due to a too low probability threshold, thereby playing a role in adjusting the system sensitivity.
[0168] Embodiment 3
[0169] See also Figure 3 , Figure 3 FIG. 1 is a schematic diagram of the structure of another public security monitoring device based on an intelligent street lamp disclosed in an embodiment of the present invention. Figure 3 As shown, the device may include:
[0170] A memory 301 storing executable program codes;
[0171] a processor 302 coupled to the memory 301;
[0172] The processor 302 calls the executable program code stored in the memory 301 to execute part or all of the steps in the public security monitoring method based on smart street lamps disclosed in the first embodiment of the present invention.
[0173] Embodiment 4
[0174] An embodiment of the present invention discloses a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute some or all steps of the public security monitoring method based on smart street lamps disclosed in Embodiment 1 of the present invention.
[0175] The above describes specific embodiments of the present specification, and other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily have to be performed in the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0176] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer-readable storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.
[0177] The apparatus, device, non-volatile computer-readable storage medium and method provided in the embodiments of this specification correspond to each other, and therefore, the apparatus, device, and non-volatile computer storage medium also have similar beneficial technical effects as the corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the corresponding apparatus, device, and non-volatile computer storage medium will not be repeated here.
[0178] In the 1990s, improvements to a technology could be clearly distinguished as hardware improvements (for example, improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the method flow). However, with the development of technology, many improvements to the method flow today can be regarded as direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement in a method flow cannot be implemented using a hardware entity module. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit whose logical function is determined by the user's programming of the device. Designers can "integrate" a digital system on a PLD by programming it themselves, without having to ask a chip manufacturer to design and produce a dedicated integrated circuit chip. Moreover, nowadays, instead of manually making integrated circuit chips, this kind of programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when developing and writing programs, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL). There is not only one HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones are VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also know that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages and program it into the integrated circuit, and then it is easy to obtain the hardware circuit that implements the logic method flow.
[0179] The controller can be implemented in any appropriate manner, for example, the controller can take the form of a microprocessor or processor and a computer-readable medium storing a computer-readable program code (such as software or firmware) that can be executed by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320. The memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that in addition to implementing the controller in a purely computer-readable program code manner, the controller can be implemented in the form of a logic gate, a switch, an application-specific integrated circuit, a programmable logic controller, and an embedded microcontroller by logically programming the method steps. Therefore, this controller can be considered as a hardware component, and the devices included therein for implementing various functions can also be regarded as structures within the hardware component. Or even, the devices for implementing various functions can be regarded as both software modules for implementing the method and structures within the hardware component.
[0180] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
[0181] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.
[0182] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification may be in the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the embodiments of this specification may be in the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.
[0183] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0184] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0185] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.
[0186] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0187] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.
[0188] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.
[0189] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0190] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.
[0191] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.
[0192] Finally, it should be noted that the public security monitoring method and device based on smart street lamps disclosed in the embodiments of the present invention only disclose the preferred embodiments of the present invention, which are only used to illustrate the technical solution 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 the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features can be replaced by equivalents. 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. A public security monitoring method based on intelligent street lamps, characterized in that: The method comprises: Acquire multiple sound wave collection information from multiple smart street lamps; the smart street lamps are arranged on streets in a target area; the smart street lamps are provided with sound wave collection equipment and playback equipment; Determining regional sound information of the target area based on the plurality of sound wave collection information; Determine the voice text and voice type corresponding to the regional voice information based on the voice analysis algorithm model; Based on the probability prediction algorithm model, the probability of a public security incident occurring in the target area is determined according to the sound text and the sound type; Determine whether the probability of a public security incident occurring in the target area is greater than a preset probability threshold. If the judgment result is yes, perform a public security maintenance operation; the public security maintenance operation includes sending a public security maintenance notification to the corresponding public security terminal and sending a public security alarm information to the playback device of the smart street lamp for playback.
2. The public security monitoring method based on intelligent street lamps according to claim 1 is characterized in that: The determining the regional sound information of the target area based on the plurality of sound wave collection information includes: Processing the plurality of sound wave collection information by a noise suppression algorithm to obtain a plurality of noise-removed sound wave information; After aligning the sound wave data of the plurality of noise-removed sound wave information according to the same time dimension, calculating the sound wave similarity of any two sound wave data of the noise-removed sound wave information in any time period; For any synthesis time period, mixing and synthesizing the sound wave data of at least two noise-removed sound wave information whose sound wave similarity in the synthesis time period is greater than a preset first similarity threshold, to obtain synthesized sound wave information corresponding to the synthesis time period; The synthesized sound wave information corresponding to all the synthesized time periods is combined to obtain the regional sound information of the target area.
3. The public security monitoring method based on intelligent street lamps according to claim 1 is characterized in that: The determining of the sound text and sound type corresponding to the regional sound information based on the speech analysis algorithm model includes: Determining, based on a sound classification algorithm model, a portion containing dialogue speech and a portion containing non-dialogue speech in the regional sound information; Determine the audio text corresponding to the dialogue speech part based on the speech recognition algorithm model; Based on a sound template matching algorithm, the sound type corresponding to the non-dialogue speech part is determined.
4. The public security monitoring method based on intelligent street lamps according to claim 3 is characterized in that: The step of determining the dialogue voice part and the non-dialogue voice part in the regional sound information based on the sound classification algorithm model includes: The regional sound information is input into a trained sound classification neural network model to determine the dialogue speech part and the non-dialogue speech part in the regional sound information; the sound classification neural network model is trained by a training data set including a plurality of training sound information and corresponding sound type labels; the sound type labels include dialogue speech labels and non-dialogue speech labels.
5. The public security monitoring method based on intelligent street lamps according to claim 3 is characterized in that: The determining the sound type corresponding to the non-dialogue speech part based on the sound template matching algorithm includes: Calculate the data similarity between the non-dialogue speech part and multiple public security sound templates, and determine the type of the public security sound template with the highest data similarity as the sound type of the non-dialogue speech part; the type of the public security sound template or the sound type includes at least one of the sound of a vehicle collision, the sound of hitting glass, gunshots, human shouting and human groaning.
6. The public security monitoring method based on intelligent street lamps according to claim 1 is characterized in that: The method of determining the probability of a public security incident occurring in the target area based on the probability prediction algorithm model and the sound text and the sound type includes: The sound text and sound type are input into a trained public security prediction neural network model to determine the probability of a public security incident occurring in the target area; the public security prediction neural network model is trained using a training data set including multiple training sound texts, training sound types and corresponding public security incident labels.
7. The public security monitoring method based on intelligent street lamps according to claim 1 is characterized in that: The smart street lamp is also provided with an image acquisition device; the method further comprises: If the judgment result is yes, an image acquisition instruction is sent to the image acquisition device of the multiple smart street lamps in the target area that meet the three-dimensional position condition; the three-dimensional position condition is that at least two of the connecting lines between the positions of the multiple smart street lamps are not parallel to each other; Receiving a plurality of image acquisition information acquired by the plurality of intelligent street lamps in response to the image acquisition instruction; Based on the three-dimensional modeling algorithm, a regional three-dimensional model of the target area is established according to the plurality of image acquisition information; Calculating the geometric similarity between the regional three-dimensional model and a plurality of public security three-dimensional templates; When the geometric similarity between the regional three-dimensional model and at least one of the public security three-dimensional templates is greater than a preset second similarity threshold, it is determined whether a public security incident occurs in the target area.
8. The public security monitoring method based on intelligent street lamps according to claim 1 is characterized in that: The method further comprises: Obtaining public security incident processing information fed back by the public security terminal; Determine the information type of the public security incident processing information; When the information type is that no public security incident has occurred, increasing the probability threshold; When the information type is a public security incident, the probability threshold is lowered.
9. A public security monitoring device based on intelligent street lamps, characterized in that: The device comprises: A sound wave acquisition module, used to acquire a plurality of sound wave collection information from a plurality of smart street lamps; the smart street lamps are arranged on streets in a target area; the smart street lamps are provided with a sound wave collection device and a playback device; A first determination module, configured to determine regional sound information of the target area based on the plurality of sound wave collection information; A second determination module is used to determine the sound text and sound type corresponding to the regional sound information based on the speech analysis algorithm model; A third determination module is used to determine the probability of a public security incident occurring in the target area based on the probability prediction algorithm model and according to the sound text and the sound type; The public security operation module is used to determine whether the probability of a public security incident occurring in the target area is greater than a preset probability threshold. If so, a public security maintenance operation is performed; the public security maintenance operation includes sending a public security maintenance notification to the corresponding public security terminal and sending a public security alarm information to the playback device of the smart street lamp for playback.
10. A public security monitoring device based on intelligent street lamps, characterized in that: The device comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the public security monitoring method based on the smart street lamp as described in any one of claims 1-8.
