Method, system, apparatus and medium for analyzing dangerous behavior of personnel

Abstract

The present application relates to the field of security technology, and specifically provides a personnel dangerous behavior analysis method, system, device and medium, aiming at the problem that the existing management and control mode cannot timely and effectively warn the dangerous behavior and dangerous information without specific dangerous action. For this purpose, the personnel dangerous behavior analysis method can obtain images containing faces and environments of personnel, respectively perform face recognition and text recognition on the images, analyze whether the personnel have dangerous behavior according to the face recognition result and the text recognition result, can identify the specific position, special identifier and dangerous information of the personnel corresponding to the face recognition result according to the text recognition result, further analyze whether the personnel corresponding to the face recognition result have dangerous behavior, form complete and accurate clues, and facilitate the management and control mechanism to effectively warn the dangerous behavior and dangerous information without dangerous action.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of security and protection technology, and specifically provides a personnel dangerous behavior analysis method, system, device and medium. BACKGROUND

[0002] With the continuous development of security and protection technology and artificial intelligence technology, the construction of a safe city has become a hot topic at present. And it is an important trend for the development of the security and protection industry to effectively utilize the massive video resources obtained by security and protection equipment. At present, the video resources obtained by the security and protection equipment are applied by the control institutions, and the control of dangerous behaviors and dangerous information is mainly through the post-research and judgment mode. Although there is also a control mode of applying the video resources obtained by the security and protection equipment to pre-warning management for some action behaviors, but since this control mode only considers the action behaviors recorded in the video, the pre-warning can only be carried out after the action behavior is analyzed to be a dangerous action such as a gun action, and those without specific dangerous actions but still exist dangerous situations cannot be timely and accurately warned. For example, after a specific person appears at a specific place, although no specific dangerous action is shown, it is extremely likely to produce dangerous behavior, at which time pre-warning is also needed, but the above control mode cannot timely and accurately pre-warn.

[0003] Correspondingly, there is a need in the art for a new personnel dangerous behavior analysis scheme to solve the above problems. SUMMARY

[0004] In order to overcome the above defects, the present application is proposed to solve or partially solve the problem that the existing control mode cannot timely and effectively pre-warn the dangerous behaviors and dangerous information without specific dangerous actions.

[0005] In a first aspect, the present application provides a personnel dangerous behavior analysis method, comprising:

[0006] obtaining an image containing a face of a person and an environment where the person is located;

[0007] respectively performing face recognition and text recognition on the image to obtain a face recognition result and a text recognition result;

[0008] analyzing whether the person has a dangerous behavior according to the face recognition result and the text recognition result.

[0009] In one technical solution of the above personnel dangerous behavior analysis method, the step of "analyzing whether the person has a dangerous behavior according to the face recognition result and the text recognition result" specifically comprises:

[0010] judging whether the person is a preset specific person according to the face recognition result;

[0011] When the person is the specific person, according to the character recognition result, it is judged whether there is dangerous information related to dangerous behavior in the environment where the person is located;

[0012] When the dangerous information exists, it is judged according to the dangerous information whether the specific person has dangerous behavior.

[0013] In one of the technical solutions of the personnel dangerous behavior analysis method, the dangerous information includes a preset place name of a specific place and / or a preset specific information, and the step of "judging whether the specific person has dangerous behavior according to the dangerous information" specifically includes:

[0014] Obtaining position information of an image acquisition device that acquires the image and time information of acquiring the image;

[0015] Judging whether the specific person has dangerous behavior according to the position information, the time information and the dangerous information.

[0016] In one of the technical solutions of the personnel dangerous behavior analysis method, the step of "judging whether the specific person has dangerous behavior according to the position information, the time information and the dangerous information" includes:

[0017] Setting a warning rule of the dangerous behavior;

[0018] According to the warning rule, analyzing the position information, the time information and the dangerous information existing in the environment where the specific person is located;

[0019] Judging whether the specific person has dangerous behavior according to the analysis result;

[0020] The warning rule includes:

[0021] If the specific person appears at a preset specific place, there is dangerous behavior; and / or, if the specific person appears at the specific place with a frequency exceeding a preset frequency threshold, there is dangerous behavior; and / or, if the dangerous information contains a preset specific information, there is dangerous behavior;

[0022] And / or, the image acquisition device includes a fixedly set image acquisition device and an image acquisition device set on a movable object.

[0023] In one of the technical solutions of the personnel dangerous behavior analysis method, before the step of "analyzing whether the person has dangerous behavior according to the face recognition result and the character recognition result", the method further includes:

[0024] determine whether the dangerous information related to dangerous behavior exists in the environment where the person is located according to the character recognition result;

[0025] when the dangerous information exists in the environment, it is directly determined that the person has dangerous behavior;

[0026] The step of "respectively performing face recognition and character recognition on the image to obtain a face recognition result and a character recognition result" comprises:

[0027] extracting face feature information of the image, and obtaining the face recognition result according to the face feature information;

[0028] performing character recognition on the image to extract character information in the image;

[0029] applying semantic recognition technology to perform semantic recognition on the character information, obtaining the character recognition result according to the semantic recognition result, or applying a word segmentation tool to perform word segmentation on the character information, and obtaining the character recognition result according to the word segmentation result.

[0030] In a second aspect, the present application provides a personnel dangerous behavior analysis system, which comprises:

[0031] an image acquisition module configured to acquire an image containing a face of a person and an environment where the person is located;

[0032] a face and character recognition module configured to respectively perform face recognition and character recognition on the image to obtain a face recognition result and a character recognition result;

[0033] a dangerous behavior analysis module configured to analyze whether the person has dangerous behavior according to the face recognition result and the character recognition result.

[0034] In one technical solution of the above personnel dangerous behavior analysis system, the dangerous behavior analysis module comprises:

[0035] a specific person determination sub-module configured to determine whether the person is a specific person according to the face recognition result;

[0036] a dangerous information determination sub-module configured to determine whether the dangerous information related to dangerous behavior exists in the environment where the person is located according to the character recognition result when the person is the specific person;

[0037] a dangerous behavior determination sub-module configured to determine whether the specific person has dangerous behavior according to the dangerous information when the dangerous information exists.

[0038] In one of the above technical solutions of the personnel dangerous behavior analysis system, the dangerous information includes a preset location name of a specific location and / or a preset specific information, and the dangerous behavior judgment submodule includes:

[0039] a position and time information acquisition unit configured to acquire position information of an image acquisition device that acquires the image and time information of acquiring the image;

[0040] a dangerous behavior judgment unit configured to judge whether the specific personnel has a dangerous behavior according to the position information, the time information, and the dangerous information.

[0041] In one of the above technical solutions of the personnel dangerous behavior analysis system, the dangerous behavior judgment unit includes:

[0042] a pre-warning rule setting subunit configured to set a pre-warning rule of the dangerous behavior;

[0043] a dangerous information analysis subunit configured to analyze the position information, the time information, and dangerous information existing in an environment where the specific personnel is located according to the pre-warning rule;

[0044] a dangerous behavior judgment subunit configured to judge whether the specific personnel has a dangerous behavior according to an analysis result;

[0045] wherein the pre-warning rule includes:

[0046] if the specific personnel appears at a preset specific location, there is a dangerous behavior; and / or, if a frequency at which the specific personnel appears at the specific location exceeds a preset frequency threshold, there is a dangerous behavior; and / or, if the dangerous information contains a preset specific information, there is a dangerous behavior;

[0047] and / or, the image acquisition device includes a fixedly set image acquisition device and an image acquisition device set on a movable object.

[0048] In one of the above technical solutions of the personnel dangerous behavior analysis system, the system further includes a dangerous behavior related information judgment module configured to perform the following steps:

[0049] judging whether there is dangerous information related to a dangerous behavior in an environment where the personnel is located according to the text recognition result;

[0050] when the dangerous information exists in the environment, it is directly determined that the personnel has a dangerous behavior;

[0051] and / or, the face and text recognition module includes:

[0052] a face recognition result obtaining sub-module configured to extract face feature information of the image, and obtain the face recognition result according to the face feature information;

[0053] a text information extracting sub-module configured to perform text recognition on the image, and extract text information in the image;

[0054] a text recognition result obtaining sub-module configured to perform semantic recognition on the text information by using a semantic recognition technology, and obtain the text recognition result according to a result of the semantic recognition, or perform word segmentation on the text information by using a word segmentation tool, and obtain the text recognition result according to a result of the word segmentation.

[0055] In a third aspect, a control device is provided, which includes a processor and a storage device, the storage device being adapted to store a plurality of program codes, the program codes being adapted to be loaded and run by the processor to execute the personnel dangerous behavior analysis method according to any one of the technical solutions of the personnel dangerous behavior analysis method.

[0056] In a fourth aspect, a computer readable storage medium is provided, which has a plurality of program codes stored therein, the program codes being adapted to be loaded and run by a processor to execute the personnel dangerous behavior analysis method according to any one of the technical solutions of the personnel dangerous behavior analysis method.

[0057] The above one or more technical solutions of the present application have at least one or more of the following beneficial effects:

[0058] In the implementation of the technical solutions of the present application, the image containing the face of the personnel and the environment where the personnel is located can be obtained, and the face recognition and the text recognition are performed on the image respectively, and then the dangerous behavior of the personnel is analyzed according to the face recognition result and the text recognition result. Through the above configuration, the text recognition result of the environment where the personnel is located can be obtained through the text recognition, and according to the text recognition result, whether the personnel corresponding to the face recognition result appears at a specific location, whether there is specific information in the environment where the personnel is located, whether there is dangerous information related to the dangerous behavior in the environment where the personnel is located, and the like can be identified, and according to the above content, whether the personnel corresponding to the face recognition result has the dangerous behavior can be analyzed, so that complete and accurate clue information is formed, so that the management and control mechanism can timely and effectively perform early warning processing on various dangerous behaviors and dangerous information without specific dangerous actions. BRIEF DESCRIPTION OF DRAWINGS

[0059] The disclosure of the present application will become more apparent from the following description in conjunction with the accompanying drawings. It is readily understood by those skilled in the art that the drawings are merely intended to illustrate the present application and are not intended to limit the scope of protection of the present application. In the drawings:

[0060] Figure 1 is a main step flowchart diagram of a personnel dangerous behavior analysis method according to an embodiment of the present application;

[0061] Figure 2 is a main structure block diagram of a personnel dangerous behavior analysis system according to an embodiment of the present application;

[0062] Figure 3 is a main step flowchart diagram of a personnel dangerous behavior analysis method according to an embodiment of the present application. DETAILED DESCRIPTION

[0063] Some embodiments of the present application will be described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present application, and are not intended to limit the protection scope of the present application.

[0064] In the description of the present application, "module" and "processor" can include hardware, software or a combination of both. A module can include hardware circuit, various suitable sensors, communication port, memory, and can also include software part such as program code, and can be a combination of software and hardware. The processor can be a central processing unit, microprocessor, image processor, digital signal processor or any other suitable processor. The processor has data and / or signal processing functions. The processor can be implemented in software, hardware or a combination of both. The non-transitory computer readable storage medium includes any suitable medium that can store program code, such as magnetic disk, hard disk, optical disk, flash memory, read-only memory, random access memory, etc. The term "A and / or B" means all possible combinations of A and B, such as only A, only B or both A and B. The term "at least one of A or B" or "at least one of A and B" has similar meaning as "A and / or B", and can include only A, only B or both A and B. The singular form of the term "one", "this" can also include plural forms.

[0065] Referring to the drawings Figure 1 , Figure 1 is a main step flowchart diagram of a personnel dangerous behavior analysis method according to an embodiment of the present application. As shown in Figure 1 , the personnel dangerous behavior analysis method in the embodiment of the present application mainly includes the following steps S101-S103.

[0066] Step S101: Obtain an image containing both the face of a person and the environment where the person is located.

[0067] In this embodiment, an image containing both the face of a person and the environment where the person is located can be obtained.

[0068] In an embodiment, the image containing the face of the person and the environment where the person is located can be an image obtained by an image acquisition device, which can contain face snapshot data of the person and image data of the environment where the person is located.

[0069] In an embodiment, the image acquisition device can include a fixedly arranged image acquisition device and an image acquisition device arranged on a movable object. For example, the image acquisition device can be a security monitoring camera arranged in a city, a fixedly arranged face socket monitoring device, a socket snapshot device arranged on a vehicle, a snapshot device arranged on a taxi, etc.

[0070] In an embodiment, the image can be an image obtained by an image acquisition device, an image obtained from an offline video file, or an image obtained from a network picture.

[0071] Step S102: respectively performing face recognition and text recognition on the image to obtain a face recognition result and a text recognition result.

[0072] In this embodiment, face recognition can be performed on the image obtained in step S101 to obtain a face recognition result, and text recognition can be performed on the image obtained in step S101 to obtain a text recognition result. It should be noted that in this embodiment, a conventional face recognition method and a conventional text recognition method in the field of image processing can be used to respectively perform face recognition and text recognition on the image to obtain a face recognition result and a text recognition result.

[0073] Step S103: analyzing whether the person has a dangerous behavior according to the face recognition result and the text recognition result.

[0074] In this embodiment, the face recognition result and the text recognition result obtained in step S102 can be analyzed to determine whether the person corresponding to the face recognition result has a dangerous behavior.

[0075] Based on the steps S101-S103, the application can obtain the image containing the face of the person and the environment where the person is located, and perform face recognition and text recognition on the image respectively, and further analyze whether the person has dangerous behavior according to the face recognition result and the text recognition result. Through the above configuration, the text recognition result of the environment where the person is located can be obtained through text recognition, and whether the person corresponding to the face recognition result appears in a specific place, whether there is specific information in the environment where the person is located, whether there is dangerous information related to dangerous behavior in the environment where the person is located, and the like can be identified according to the text recognition result. According to the above content, whether the person corresponding to the face recognition result has dangerous behavior can be further analyzed, so as to form complete and accurate clue information, so that the management and control mechanism can timely and effectively perform early warning processing on various dangerous behaviors and dangerous information without specific dangerous actions.

[0076] The steps S102 and S103 are further described below.

[0077] In an embodiment of the application, the step S102 can further include the following steps:

[0078] Step S1021: Extract the face feature information of the image, and obtain the face recognition result according to the face feature information.

[0079] In the embodiment, the face feature information in the image obtained in the step S101 can be extracted, and the face recognition result can be obtained according to the face feature information.

[0080] Step S1022: Perform text recognition on the image, and extract the text information in the image.

[0081] In the embodiment, the image obtained in the step S101 can be subjected to text recognition, and the text information in the image can be extracted.

[0082] In an embodiment, the OCR (Optical Character Recognition) recognition method can be used to extract the text information in the image.

[0083] Step S1023: Apply semantic recognition technology to perform semantic recognition on the text information, and obtain the text recognition result according to the semantic recognition result.

[0084] In the embodiment, the semantic recognition technology can be applied to perform named entity recognition on the text information obtained in step S1022, to obtain entities in the text information and further obtain entity relationships between the entities, and generate structured data corresponding to the text information according to the entities and the entity relationships, and take the structured data as the text recognition result. The semantic recognition technology refers to a technology for enabling effective communication between a person and a computer by using a natural language. The named entity recognition is a basic task in the semantic recognition technology, that is, to recognize entities with strong specific meaning or reference in a text. After the named entity recognition, the relationship extraction between the entities can be further performed to obtain the entity relationships between the entities.

[0085] In one embodiment, the entity relationship can include attribute classification of the entity, for example, if the entity obtained according to the semantic recognition on the text information includes a bank-related entity, it can be determined that the person is at the bank according to the entity relationship of the entity.

[0086] In one embodiment of the embodiment of the present application, the text recognition result can also be obtained according to the following steps:

[0087] Step S1024: applying a word segmentation tool to perform word segmentation on the text information, and obtaining the text recognition result according to the word segmentation result.

[0088] In the embodiment, the word segmentation tool can be applied to perform word segmentation on the text information obtained in step S1022, and the text recognition result can be obtained according to the word segmentation result.

[0089] In one embodiment, the lucene word segmenter or other tools can be applied to perform intelligent word group recognition on the text information, to perform data processing on the text information, so that the text information can be searched, and the text information that can be searched is taken as the text recognition result. The lucene word segmenter is an open source full-text search engine tool, which can provide text query and indexing functions.

[0090] In one embodiment of the embodiment of the present application, step S103 can further include:

[0091] Step S1031: determining whether the person is a preset specific person according to the face recognition result.

[0092] In the embodiment, the face recognition result obtained in step S102 can be used to determine whether the person corresponding to the face recognition result is a preset specific person. The specific person refers to a person who needs to be controlled by a control agency.

[0093] In an embodiment, images obtained by different image acquisition devices can be obtained, and a face clustering technique can be applied to cluster and archive the face recognition results. The face clustering technique refers to an algorithm for grouping faces in a set according to identity (e.g., face feature information).

[0094] In an embodiment, the clustered face feature information can be compared with face feature information in a specific personnel library of a management and control mechanism, to determine whether the personnel corresponding to the obtained face feature information is a specific personnel of the specific personnel library of the management and control mechanism.

[0095] Step S1032: When the personnel is a specific personnel, according to the text recognition result, it is determined whether there is dangerous information related to dangerous behavior in the environment in which the personnel is located.

[0096] In the embodiment, when the personnel corresponding to the face recognition result is a specific personnel, according to the text recognition result obtained in step S102, it can be determined whether there is dangerous information related to dangerous behavior in the environment in which the personnel is located.

[0097] Step S1033: When there is dangerous information, it is determined whether the specific personnel has dangerous behavior according to the dangerous information.

[0098] In the embodiment, when it is determined in step S1032 that there is dangerous information related to dangerous behavior, it can be determined whether the specific personnel has dangerous behavior according to the dangerous information.

[0099] In an embodiment, the dangerous information can include a preset specific location name and a preset specific information, and it can be determined whether the text recognition result includes the specific location name and / or the specific information. If the text recognition result includes the specific location name and / or the specific information, it can be considered that there is dangerous information related to dangerous behavior in the environment in which the personnel is located. A person skilled in the art can set the preset specific location name and the preset specific information according to the needs of actual applications. For example, the preset specific location can be a bank, and the specific location name can be the name of the bank. The preset specific information can be a preset information.

[0100] In an embodiment, step S1033 can further include:

[0101] Step S10331: Obtain position information of an image acquisition device that acquires an image and time information of acquiring the image.

[0102] Step S10332: Determine whether the specific personnel has dangerous behavior according to the position information, the time information, and the dangerous information.

[0103] In the embodiment, the position information of the image acquisition device acquiring the image and the time information of acquiring the image can be acquired, and it is determined whether the specific person has dangerous behavior according to the position information, the time information, and the dangerous information determined in step S1032.

[0104] In one embodiment, step S10332 can further include:

[0105] Step S103321: setting a warning rule of dangerous behavior;

[0106] Step S103322: analyzing the position information, the time information, and the dangerous information existing in the environment where the specific person is located according to the warning rule;

[0107] Step S103323: determining whether the specific person has dangerous behavior according to the analysis result.

[0108] The warning rule includes: if the specific person appears at a preset specific place, the specific person has dangerous behavior; if the specific person appears at the specific place with a frequency exceeding a preset frequency threshold, the specific person has dangerous behavior; and if the dangerous information contains preset specific information, the specific person has dangerous behavior.

[0109] In the embodiment, the warning rule of dangerous behavior can be set according to the control requirements of the control mechanism. The warning rule can be that if the specific person appears at a preset specific place, it is determined that the specific person has dangerous behavior; or if the specific person appears at the specific place with a frequency exceeding a preset frequency threshold, it is determined that the specific person has dangerous behavior; or if the dangerous information existing in the environment where the specific person is located contains preset specific information, it is determined that the specific person has dangerous behavior. A person skilled in the art can set the specific place, the frequency threshold, and the specific information according to the actual application requirements. The position information of the image acquisition device, the time information, and the dangerous information existing in the environment where the specific person is located can be analyzed according to the warning rule, and it is determined whether the specific person has dangerous behavior according to the analysis result.

[0110] Specifically, the dangerous information existing in the environment where the specific person is located can be analyzed according to the position information of the image acquisition device and the time information of the acquired image. For example, the spatial position where the specific person is located can be determined according to the position information of the image acquisition device, and the time when the dangerous information exists can be determined according to the time information of the acquired image. By comprehensively analyzing the dangerous information existing in the environment where the specific person is located according to the position information of the image acquisition device and the time information of the acquired image, it can be determined whether the specific person appears at a specific place controlled by the control mechanism, whether the specific person frequently appears at the specific place (the frequency of appearing at the specific place exceeds a preset frequency threshold), and whether the environment where the specific person appears contains specific information controlled by the control mechanism. In this way, the time and space information of the specific person can be comprehensively mastered, and timely warning can be given when dangerous behavior occurs.

[0111] In one embodiment of the embodiment of the present application, in addition to steps S101-S103, the present application can further include step S104:

[0112] Step S104: determining whether dangerous information related to dangerous behavior exists in the environment where the person is located according to the text recognition result; when dangerous information exists in the environment, it is directly determined that the person has dangerous behavior.

[0113] In the present embodiment, it can be determined whether dangerous information related to dangerous behavior exists in the environment where the person is located according to the text recognition result. If dangerous information related to dangerous behavior exists in the environment where the person is located, it can be directly determined that the person in the environment has dangerous behavior. Specifically, the control mechanism can set dangerous information related to dangerous behavior, such as keywords related to dangerous behavior. If such keywords exist in the environment where the person is located, it can be directly determined that the person has dangerous behavior. That is, whether the person is a preset specific person or not, if dangerous information exists in the environment, it can be determined that the person has dangerous behavior.

[0114] In one embodiment, the face recognition result can be clustered and archived to determine whether the face recognition result is a person whose identity information is not recorded in the control mechanism. If it is a person whose identity information is not recorded in the control mechanism, a warning rule can be set, and dangerous behavior analysis can be performed on such person to timely warn the dangerous behavior.

[0115] In one embodiment, the face recognition result, the text recognition result, the position information of the image acquisition device, and the time information of the acquired image of the person can be combined to generate a database of the person, and the behavior of the person can be analyzed according to the database of the person.

[0116] Referring to the drawings Figure 3 , Figure 3is a main step flow schematic diagram of a personnel dangerous behavior analysis method according to an embodiment of the present application. As shown in Figure 3 In the embodiment, the image acquisition device can be used to acquire an image containing a face of a person and an environment where the person is located. The image containing the face can be used as a small image, and the image of the environment where the person is located can be used as a panoramic image. After face recognition of the small image, a personnel profile cluster is generated by face clustering technology. After OCR text recognition of the panoramic image, text information in the panoramic image is acquired, and semantic recognition technology is applied to recognize the text information to enrich the information dimension of the personnel profile cluster. Further, a specific personnel management model and an abnormal behavior management model can be applied to analyze the personnel profile cluster, and a specific information early warning model can be applied to analyze the semantic recognition result, and a personnel dangerous behavior early warning and judgment can be performed according to the analysis results of the specific personnel management model, the abnormal behavior management model and the specific information early warning model. The specific functions of the specific personnel management model can be referred to the steps S1031 to S1033 of the foregoing method embodiment, the specific functions of the abnormal behavior management model can be referred to the steps S103321 to S103323 of the foregoing method embodiment, and the specific functions of the specific information early warning model can be referred to the step S104 of the foregoing method embodiment.

[0117] It should be noted that although the steps in the above embodiments are described in a specific order, those skilled in the art can understand that, in order to achieve the effect of the present application, the steps do not have to be executed in this order, and they can be executed simultaneously (in parallel) or in other orders, and these changes are within the protection scope of the present application.

[0118] Further, the present application also provides a personnel dangerous behavior analysis system.

[0119] Referring to the accompanying Figure 2 , Figure 2 is a main structure block diagram of a personnel dangerous behavior analysis system according to an embodiment of the present application. As shown in Figure 2 The personnel dangerous behavior analysis system in the embodiment of the present application can include an image acquisition module, a face and text recognition module and a dangerous behavior analysis module. In the embodiment, the image acquisition module can be configured to acquire an image containing a face of a person and an environment where the person is located. The face and text recognition module can be configured to perform face recognition and text recognition on the image respectively to obtain face recognition results and text recognition results. The dangerous behavior analysis module can be configured to analyze whether the person has dangerous behavior according to the face recognition results and the text recognition results.

[0120] In an embodiment, the dangerous behavior analysis module can include a specific person judgment sub-module, a dangerous information judgment sub-module, and a dangerous behavior judgment sub-module. In this embodiment, the specific person judgment sub-module can be configured to judge whether a person is a preset specific person according to a face recognition result. The dangerous information judgment sub-module can be configured to judge whether there is dangerous information related to a dangerous behavior in an environment where the person is located according to a text recognition result when the person is the specific person. The dangerous behavior judgment sub-module can be configured to judge whether the specific person has a dangerous behavior according to the dangerous information when there is the dangerous information.

[0121] In an embodiment, the dangerous information can include a location name of a preset specific location and / or a preset specific information, and the dangerous behavior judgment sub-module can include a location and time information acquisition unit and a dangerous behavior judgment unit. In this embodiment, the location and time information acquisition unit can be configured to acquire location information of an image acquisition device that acquires an image and time information of acquiring the image. The dangerous behavior judgment unit can be configured to judge whether the specific person has a dangerous behavior according to the location information, the time information, and the dangerous information.

[0122] In an embodiment, the dangerous behavior judgment unit can include a pre-warning rule setting sub-unit, a dangerous information analysis sub-unit, and a dangerous behavior judgment sub-unit. In this embodiment, the pre-warning rule setting sub-unit can be configured to set a pre-warning rule of a dangerous behavior. The dangerous information analysis sub-unit can be configured to analyze the location information, the time information, and dangerous information existing in an environment where the specific person is located according to the pre-warning rule. The dangerous behavior judgment sub-unit can be configured to judge whether the specific person has a dangerous behavior according to an analysis result. The pre-warning rule includes: if the specific person appears at a preset specific location, the specific person has a dangerous behavior; if a frequency at which the specific person appears at the specific location exceeds a preset frequency threshold, the specific person has a dangerous behavior; and if the dangerous information contains a preset specific information, the specific person has a dangerous behavior.

[0123] In an embodiment, the personnel dangerous behavior analysis system can further include a dangerous behavior related information judgment module. In this embodiment, the dangerous behavior related information judgment module can be configured to perform the following steps: judging whether there is dangerous information related to a dangerous behavior in an environment where a person is located according to a text recognition result; and directly judging that the person has a dangerous behavior when there is the dangerous information in the environment.

[0124] In one embodiment, the face and text recognition module can include a face recognition result obtaining sub-module, a text information extracting sub-module, and a text recognition result obtaining sub-module. In this embodiment, the face recognition result obtaining sub-module can be configured to extract face feature information of the image, and obtain a face recognition result according to the face feature information. The text information extracting sub-module can be configured to perform text recognition on the image, and extract text information in the image. The text recognition result obtaining sub-module can be configured to perform semantic recognition on the text information by applying a semantic recognition technology, and obtain the text recognition result according to a result of the semantic recognition, or perform word segmentation on the text information by applying a word segmentation tool, and obtain the text recognition result according to a result of the word segmentation.

[0125] In one embodiment, the image acquisition device can include a fixedly arranged image acquisition device and an image acquisition device arranged on a movable object.

[0126] The personnel dangerous behavior analysis system described above is used to execute Figure 1 and Figure 3 The technical principles, technical problems solved, and technical effects of the two are similar. For the convenience and brevity of description, the personnel dangerous behavior analysis system can refer to the content described in the embodiment of the personnel dangerous behavior analysis method, and details will not be repeated here.

[0127] Those skilled in the art can understand that all or part of the processes in the method of the above-mentioned embodiment can also be completed by a computer program instructing related hardware. The computer program can be stored in a computer readable storage medium. When the computer program is executed by a processor, the steps of each method embodiment described above can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or some intermediate forms. The computer readable storage medium can include any entity or device, medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal, and software distribution medium. It should be noted that the content included in the computer readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, according to legislation and patent practice in some jurisdictions, the computer readable storage medium does not include electrical carrier signals and telecommunication signals.

[0128] Further, the present application also provides a control device. In an embodiment of the control device according to the present application, the control device comprises a processor and a storage device, the storage device can be configured to store a program of the personnel dangerous behavior analysis method according to the above-mentioned embodiments of the method, and the processor can be configured to execute the program in the storage device, which includes but is not limited to the program of the personnel dangerous behavior analysis method according to the above-mentioned embodiments of the method. For the convenience of illustration, only the parts related to the embodiments of the present application are shown, and the specific technical details not disclosed are referred to the method embodiments of the present application. The control device can be a control device device formed by various electronic devices.

[0129] Further, the present application also provides a computer readable storage medium. In an embodiment of the computer readable storage medium according to the present application, the computer readable storage medium can be configured to store a program of the personnel dangerous behavior analysis method according to the above-mentioned embodiments of the method, which can be loaded and run by a processor to implement the personnel dangerous behavior analysis method according to the above-mentioned embodiments of the method. For the convenience of illustration, only the parts related to the embodiments of the present application are shown, and the specific technical details not disclosed are referred to the method embodiments of the present application. The computer readable storage medium can be a storage device device formed by various electronic devices, and optionally, the computer readable storage medium in the embodiments of the present application is a non-transitory computer readable storage medium.

[0130] Further, it should be understood that, since the setting of each module is only for illustrating the functional units of the device of the present application, the corresponding physical device of the module can be the processor itself, or a part of software, hardware or the combination of software and hardware in the processor. Therefore, the number of each module in the figure is only illustrative.

[0131] Those skilled in the art can understand that each module in the device can be adaptively split or combined. Such splitting or combining of specific modules does not cause the technical solution to deviate from the principles of the present application, and therefore, the technical solution after splitting or combining will fall within the protection scope of the present application.

[0132] So far, the technical solution of the present application has been described in combination with the preferred embodiments shown in the drawings, but those skilled in the art can easily understand that the protection scope of the present application is obviously not limited to these specific embodiments. Those skilled in the art can make equivalent changes or replacements to the related technical features without deviating from the principles of the present application, and the technical solution after such changes or replacements will fall within the protection scope of the present application.

Claims

1. A method of analyzing dangerous behavior of a person, characterized by, The method comprises: acquiring an image containing a face of a person and an environment in which the person is located; performing face recognition and text recognition on the image respectively to obtain a face recognition result and a text recognition result; analyzing whether the person has a dangerous behavior according to the face recognition result and the text recognition result; the step of analyzing whether the person has a dangerous behavior according to the face recognition result and the text recognition result specifically comprises: judging whether the person is a specific person according to the face recognition result; when the person is the specific person, judging whether there is dangerous information related to a dangerous behavior in the environment in which the person is located according to the text recognition result; the dangerous information comprises a place name of a specific place and / or a specific information; when the dangerous information exists, acquiring position information of an image acquisition device that acquires the image and time information of acquiring the image; setting a warning rule of the dangerous behavior; analyzing the position information, the time information and the dangerous information existing in the environment in which the specific person is located according to the warning rule; and judging whether the specific person has a dangerous behavior according to an analysis result; wherein the warning rule comprises: if the specific person appears at a specific place, there is a dangerous behavior; and / or, if the specific person appears at the specific place with a frequency exceeding a preset frequency threshold, there is a dangerous behavior; and / or, if the dangerous information contains a specific information, there is a dangerous behavior.

2. The method according to claim 1, wherein the image acquisition device comprises a fixed image acquisition device and an image acquisition device arranged on a movable object. Before the step of analyzing whether the person has a dangerous behavior according to the face recognition result and the text recognition result, the method further comprises:

3. The method of claim 1, wherein, judging whether there is dangerous information related to a dangerous behavior in the environment in which the person is located according to the text recognition result; when the dangerous information exists in the environment, it is directly determined that the person has a dangerous behavior; and / or, the step of performing face recognition and text recognition on the image respectively to obtain a face recognition result and a text recognition result comprises: extracting face feature information of the image, and obtaining the face recognition result according to the face feature information; performing text recognition on the image to extract text information in the image; applying semantic recognition technology to perform semantic recognition on the text information, and obtaining the text recognition result according to a result of the semantic recognition, or applying a word segmentation tool to perform word segmentation on the text information, and obtaining the text recognition result according to a result of the word segmentation. The system comprises:

4. A system for analyzing dangerous behavior of a person, characterized in that an image acquisition module configured to acquire an image containing a face of a person and an environment in which the person is located; a face and text recognition module configured to perform face recognition and text recognition on the image respectively to obtain a face recognition result and a text recognition result; and an analysis module configured to analyze whether the person has a dangerous behavior according to the face recognition result and the text recognition result. a dangerous behavior analysis module configured to analyze whether the personnel has dangerous behavior according to the face recognition result and the text recognition result; the dangerous behavior analysis module comprises: a specific personnel judgment sub-module configured to judge whether the personnel is a preset specific personnel according to the face recognition result; a dangerous information judgment sub-module configured to judge whether there is dangerous information related to dangerous behavior in an environment where the personnel is located according to the text recognition result when the personnel is the specific personnel; a dangerous behavior judgment sub-module configured to judge whether the specific personnel has dangerous behavior according to the dangerous information when the dangerous information exists; the dangerous information comprises a preset specific location name and / or a preset specific information, and the dangerous behavior judgment sub-module comprises: a position and time information acquisition unit configured to acquire position information of an image acquisition device that acquires the image and time information of acquiring the image; a dangerous behavior judgment unit configured to judge whether the specific personnel has dangerous behavior according to the position information, the time information and the dangerous information; the dangerous behavior judgment unit comprises: a pre-warning rule setting sub-unit configured to set a pre-warning rule of the dangerous behavior; a dangerous information analysis sub-unit configured to analyze the position information, the time information and dangerous information existing in an environment where the specific personnel is located according to the pre-warning rule; a dangerous behavior judgment sub-unit configured to judge whether the specific personnel has dangerous behavior according to an analysis result; wherein the pre-warning rule comprises: if the specific personnel appears at a preset specific location, there is dangerous behavior; and / or, if a frequency of the specific personnel appearing at the specific location exceeds a preset frequency threshold, there is dangerous behavior; and / or, if the dangerous information contains a preset specific information, there is dangerous behavior.

5. The personnel dangerous behavior analysis system according to claim 4, wherein the image acquisition device comprises a fixedly arranged image acquisition device and an image acquisition device arranged on a movable object.

6. The system for analyzing dangerous behavior of people according to claim 4, wherein The system further comprises a dangerous behavior related information judgment module configured to perform the following steps: judging whether there is dangerous information related to dangerous behavior in an environment where the personnel is located according to the text recognition result; when the dangerous information exists in the environment, it is directly determined that the personnel has dangerous behavior; and / or, the face and text recognition module comprises: a face recognition result acquisition sub-module configured to extract face feature information of the image, and obtain the face recognition result according to the face feature information; a text information extraction sub-module configured to perform text recognition on the image, and extract text information in the image; The character recognition result obtaining submodule is configured to apply a semantic recognition technology to perform semantic recognition on the character information, to obtain the character recognition result according to a result of the semantic recognition, or to apply a word segmentation tool to perform word segmentation on the character information, to obtain the character recognition result according to a result of the word segmentation.

7. A control device comprising a processor and a storage device, the storage device being adapted to store a plurality of program codes, characterized in that, The program code is adapted to be loaded and run by the processor to perform the personnel dangerous behavior analysis method of any one of claims 1 to 3.

8. A computer readable storage medium having stored therein a plurality of program codes, characterized in that, The program code is adapted to be loaded and run by the processor to perform the personnel dangerous behavior analysis method of any one of claims 1 to 3.

Images