Physical violence detection method, system and electronic equipment

Abstract

This application applies to the field of computer analysis technology and provides a method, system, and electronic device for detecting physical violence. The method comprises: determining the kinetic energy of multiple body parts of at least one observed individual in a dynamic image; screening suspected violent individuals from the at least one observed individual based on the kinetic energy; and determining the suspected violent individual as a target individual engaging in physical violence if the motion posture of the multiple body parts of the suspected violent individual matches a predetermined attack posture. This solution can improve the accuracy of detecting physical violence.

Description

Technical Field

[0001] The present application belongs to the field of computer analysis technology, and in particular relates to a method, system, and electronic device for detecting physical violence. Background Art

[0002] Violent behavior primarily manifests itself in two main categories: verbal violence and physical violence. Currently, verbal violence detection relies on voiceprint feature extraction and natural language processing technologies, using emotion recognition and insulting word matching. With the advancement of these technologies, the accuracy of verbal violence detection has increased. However, existing methods for detecting physical violence have a relatively limited range of detection criteria, resulting in relatively low accuracy. Summary of the Invention

[0003] The embodiments of the present application provide a method, system, and electronic device for detecting physical violence to address the problem of low accuracy in detecting physical violence in the prior art.

[0004] A first aspect of an embodiment of the present application provides a method for detecting physical violence, comprising:

[0005] determining kinetic energy of a plurality of body parts of each of the observed individuals based on at least one observed individual in the dynamic image;

[0006] Based on the kinetic energy, screening out a suspected violent individual from at least one of the observed individuals;

[0007] In a case where the movement postures of the multiple body parts of the suspected violent individual conform to the set attack postures, the suspected violent individual is determined to be a target individual with physical violent behavior.

[0008] A second aspect of the embodiments of the present application provides a physical violence behavior detection system, comprising:

[0009] A first determining module is configured to determine kinetic energy of multiple body parts of each observed individual based on at least one observed individual in the dynamic image;

[0010] a screening module, configured to screen out a suspected violent individual from at least one of the observed individuals based on the kinetic energy;

[0011] The second determination module is configured to determine that the suspected violent individual is a target individual with physical violence behavior if the movement postures of the multiple body parts of the suspected violent individual conform to a set attack posture.

[0012] A third aspect of an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method described in the first aspect when executing the computer program.

[0013] A fourth aspect of an embodiment of the present application provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the steps of the method described in the first aspect.

[0014] A fifth aspect of an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the method described in the first aspect are implemented.

[0015] As can be seen above, this application first determines the kinetic energy of multiple body parts of each observed individual in the dynamic image, quantifying the energy characteristics of the observed individual's movements. Then, based on the kinetic energy, suspected violent individuals are screened from at least one observed individual. If the movement postures of multiple body parts of the suspected violent individual match the set attack posture, the suspected violent individual is determined to be a target individual who has committed physical violence. This dual-modal detection and identification mechanism, which achieves multi-dimensional detection and identification through initial kinetic energy screening and posture verification, effectively improves the accuracy of detecting physical violence. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0017] Figure 1 This is a flow chart of a method for detecting physical violence provided in an embodiment of the present application;

[0018] Figure 2 This is a schematic diagram of a human body part movement posture that conforms to a set attack posture provided by an embodiment of the present application. Figure 1 ;

[0019] Figure 3 This is a schematic diagram of a human body part movement posture that conforms to a set attack posture provided by an embodiment of the present application. Figure 2 ;

[0020] Figure 4 This is a structural diagram of a physical violence detection system provided in an embodiment of the present application;

[0021] Figure 5 This is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0022] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0023] It will be understood that when used in this specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0024] It should also be understood that the terms used in this specification are for the purpose of describing specific embodiments only and are not intended to limit the present application. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0025] It should be further understood that the term "and / or" used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0026] As used in this specification and the appended claims, the term "if" can be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0027] In specific implementations, the terminals described in the embodiments of the present application include, but are not limited to, other portable devices such as mobile phones, laptop computers, or tablet computers with touch-sensitive surfaces (e.g., touch screen displays and / or touch pads). It should also be understood that in some embodiments, the device is not a portable communication device, but a desktop computer with a touch-sensitive surface (e.g., touch screen displays and / or touch pads).

[0028] In the following discussion, a terminal including a display and a touch-sensitive surface is described. However, it should be understood that the terminal may include one or more other physical user interface devices such as a physical keyboard, mouse, and / or joystick.

[0029] The terminal supports various applications, such as one or more of the following: a drawing application, a presentation application, a word processing application, a website creation application, a disk burning application, a spreadsheet application, a game application, a phone application, a video conferencing application, an email application, an instant messaging application, a workout support application, a photo management application, a digital camera application, a digital video camera application, a web browsing application, a digital music player application, and / or a digital video player application.

[0030] Various applications that can be executed on the terminal can use at least one common physical user interface device, such as a touch-sensitive surface. One or more functions of the touch-sensitive surface and corresponding information displayed on the terminal can be adjusted and / or changed between applications and / or within a corresponding application. In this way, the common physical architecture of the terminal (e.g., the touch-sensitive surface) can support a variety of applications with user interfaces that are intuitive and transparent to the user.

[0031] It should be understood that the size of the serial numbers of each step in this embodiment does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of this application.

[0032] In order to illustrate the technical solution described in this application, specific embodiments are provided below.

[0033] See also Figure 1 , Figure 1 This is a flow chart of a method for detecting physical violence provided in an embodiment of the present application. Figure 1 As shown, a method for detecting physical violence includes the following steps:

[0034] Step 101 : determining kinetic energy of multiple body parts of each observed individual based on at least one observed individual in a dynamic image.

[0035] Image acquisition equipment is installed in observation areas such as school corridors, hospital lobbies, subway platforms, etc. to capture dynamic images.

[0036] In some embodiments, the image acquisition device is a high-frame-rate wide-angle camera, a thermal imaging device, a depth sensor, etc. The depth sensor may be a red-green-blue depth (RGBD) camera to more accurately identify the observed individual and determine the kinetic energy of the observed individual's body parts in space.

[0037] In some embodiments, dynamic images of the observation area can also be flexibly captured by the image capture equipment carried by the drone.

[0038] Dynamic images are image data captured by image acquisition devices and contain multiple frames of images that are sequential in time, such as video streams and time-series image sequences. Each frame typically includes a corresponding timestamp to support image analysis and processing of time-related information such as speed.

[0039] Observed individuals refer to people who appear in the observation area and are captured by the image acquisition equipment.

[0040] In some embodiments, dynamic images refer to video streams and / or time-series image sequences captured by image acquisition equipment that contain valid observation data of the observed individual, that is, through pre-screening, the content containing valid observation data is screened out to reduce the amount of data that needs to be processed subsequently.

[0041] In some embodiments, the dynamic image is image data that has undergone image preprocessing. Image preprocessing includes Gaussian filtering, histogram equalization, contrast enhancement, and other techniques. Image preprocessing improves the image quality of the dynamic image and enhances the accuracy of detecting physical violence.

[0042] In some embodiments, the images included in the dynamic image are all images of a set image format, such as images with a resolution of 1280×720 (pixels) or 1920×1080 (pixels).

[0043] Body parts refer to those that could potentially strike others, such as the hands, feet, head, elbows, knees, and shoulders. It's important to note that when calculating kinetic energy, unilateral body parts, such as the left hand, right hand, left foot, right foot, left knee, and right knee, are treated as independent parts for the calculation of kinetic energy.

[0044] The core essence of physical violence is that parts of the human body generate high energy output in a short period of time, and kinetic energy is the characteristic quantity that quantifies the energy output of parts of the human body.

[0045] Through dynamic images, the kinetic energy of multiple body parts of the observed individual is determined to detect whether the observed individual has high energy output in a short period of time, thereby realizing the detection of physical violence.

[0046] In some embodiments, the determining of the kinetic energy of multiple body parts of each observed individual based on at least one observed individual in the dynamic image includes: performing personnel detection on the dynamic image to determine at least one observed individual in the dynamic image; performing personnel matching on each observed individual to obtain an image frame sequence of each observed individual in the dynamic image; determining the body part mass of multiple body parts of the observed individual and position data of multiple body key points in the image frame sequence based on the image frame sequence of each observed individual; calculating the moving speed of multiple body parts of the observed individual based on the position change data of the position data of multiple body key points in the image frame sequence; and calculating the kinetic energy of multiple body parts of each observed individual based on the moving speed and the body part mass of each body part.

[0047] By identifying, matching, and analyzing key points of observed individuals in dynamic images, the movement speed and kinetic energy of multiple body parts of each observed individual are accurately calculated, and limb movements are converted into quantifiable physical energy characteristics. Abnormal limb behaviors with high energy output are effectively identified, realizing cross-modal detection of limb violence from visual data to physical quantity judgment.

[0048] In some embodiments, target detection algorithms such as YOLOv8, Faster Region-based Convolutional Neural Network (Faster R-CNN) algorithm, Mask Region-based Convolutional Neural Network (Mask R-CNN) algorithm, etc. are used to perform person detection on dynamic images, thereby accurately identifying at least one observed individual contained in the dynamic images.

[0049] In some embodiments, an object detection algorithm is used to identify an observed individual in multiple frames of images included in a dynamic image.

[0050] After identifying the observed individual through the target detection algorithm, the individual's physical features and motion trajectory are used for person matching, achieving matching mapping of the same observed individual in multiple frames of image, and then obtaining an image frame sequence for each observed individual in the dynamic image. The image frame sequence is image data containing at least one frame of image and corresponding timestamp information.

[0051] In some embodiments, each observed individual's image frame sequence undergoes local enhancement or background blur processing to maintain visual prominence of the observed individual in each frame, facilitating better determination of the mass, movement speed, and kinetic energy of the observed individual's body parts. The images contained in each observed individual's image frame sequence also conform to a predetermined image format.

[0052] After determining the image frame sequence for each observed individual, the body mass of multiple body parts and the position data of multiple body key points in the image frame sequence are further determined based on the image frame sequence. The body key points refer to the skeletal joints of the body parts, which serve as reference points for determining the movement speed of the body parts.

[0053] In some embodiments, multiple frames of a dynamic image captured by a fixed-position image capture device share a common reference coordinate system. For example, the lower left corner of the image is used as the coordinate origin, the image length edge at the coordinate origin is used as the x-axis, and the image width edge at the coordinate origin is used as the y-axis. Accordingly, the image frame sequence of an observed individual in the dynamic image captured by the fixed-position image capture device also shares this reference coordinate system.

[0054] In some embodiments, if the image acquisition device is an RGBD camera, the reference coordinate system is a three-dimensional space coordinate system.

[0055] By introducing a reference coordinate system, we can quickly obtain the position data, i.e., the coordinate value, of each key point of the human body in the same reference coordinate system, thereby making the subsequent calculated movement speed more accurate and more useful.

[0056] In some embodiments, determining the body part masses of the multiple body parts of the observed individual and the position data of the multiple body key points in the image frame sequence based on the image frame sequence of each of the observed individual includes: calculating the body part masses of the multiple body parts of the observed individual based on the image area and mass calculation coefficient of each of the body parts of the observed individual in the image frame sequence; identifying the body key points of the observed individual in each image frame in the image frame sequence, and determining the positions of the identified body key points in the image frame sequence as the position data of the body key points.

[0057] Based on the image frame sequence, the mass of human body parts is calculated, and the position data of key points of the human body are determined to achieve the mapping of image data to biometric data, providing a data basis for kinetic energy calculation.

[0058] For ease of understanding, the image area in the above embodiment is referred to as a first image area. The first image area corresponding to each body part of the observed individual is determined based on the image frame sequence of the observed individual.

[0059] In some embodiments, each observed individual's image frame sequence includes at least one frame of image, and the second image area of ​​each human body part in each frame of image is determined. For each human body part, the average value of the second image area of ​​the human body part in each frame of image can be used as the first image area. The units of the first image area and the second image area can be mm. 2 or pixel 2 , pixel means pixel.

[0060] In some embodiments, the mass calculation coefficient is a conversion parameter that converts the image area of ​​a human body part into the mass of the human body part. Introducing the mass calculation coefficient to predict the mass of a human body part can adapt to different body shapes and improve the accuracy of mass prediction. The unit of the mass calculation coefficient can be kg / mm 2 or kg / pixel 2 For example, the mass calculation factor for the arm is 0.05kg / pixel².

[0061] In some embodiments, the sample mass of each body part of the observed individual in multiple reference images and the sample image area in a predetermined image format are pre-calculated. Based on the sample mass and sample image area, a conversion parameter between the sample mass and sample image area is calculated, i.e., a mass calculation coefficient is calculated. Different mass calculation coefficients correspond to different body parts.

[0062] The first image area of ​​each human body part is multiplied by the mass calculation coefficient, that is, human body part mass = first image area × mass calculation coefficient, and then the human body part mass of the human body part is calculated to realize the conversion from image area to human body part mass.

[0063] The OpenPose algorithm, DeepCut algorithm, or AlphaPose algorithm is used to extract the human skeleton from each frame in the image frame sequence. The human key points of the observed individual in each frame are then obtained based on the human skeleton, and the coordinates of the human key points in the reference coordinate system are determined to obtain position data.

[0064] The position change data of the key points of the human body in the image frame sequence represents the distance s, that is, the coordinate change value of the key points of the human body in different images is the distance s, and the time difference of the corresponding images is the time t. The formula calculates the moving speed of multiple key points of the observed individual's body as the moving speed of the corresponding body parts.

[0065] In some embodiments, the coordinates of the left hand of the observed individual in the first frame image are , the coordinates in the second frame image are , the first frame image is taken earlier than the second frame image, and the time difference between the two is , then the moving speed of the left hand of the observed individual is calculated based on these two frames of images, and the moving speed of the left hand of the observed individual is .

[0066] In some embodiments, for any part of the human body, the average value of the movement speed between multiple frames is taken as the movement speed required for calculating the kinetic energy.

[0067] Given the moving speed v and mass m of a human body part, the kinetic energy is calculated according to the formula , calculate the kinetic energy E of each body part of each observed individual.

[0068] In some embodiments, there is a situation where a body part of the observed individual is obscured. To deal with this situation, based on the image frame sequence, body part detection is performed on each of the observed individuals. When it is detected that the observed individual has an obscured body part, the unobstructed body part of the observed individual is determined; based on the unobstructed body part, the posture of the observed individual is estimated to obtain the part contour of the obscured body part and the body key points, and the position data of the body key points is determined; based on the image area corresponding to the part contour and the mass calculation coefficient, the body part mass of the obscured body part is calculated.

[0069] Specifically, human body part detection is performed on the observed individual in each image in the image frame sequence. Each frame is checked for any obscured human body parts, while unobstructed human body parts are simultaneously determined. If an obscured human body part is detected, the posture of the observed individual is estimated based on the determined unobstructed human body part. The part contour and key points of the obscured human body part are then generated, and the coordinates, i.e., positional data, of the estimated key points are determined accordingly. The body mass of the obscured human body part is then calculated using the image area and mass calculation coefficient corresponding to the part contour of the obscured human body part. Furthermore, the coordinates of the key points corresponding to the obscured human body part are used to calculate the movement speed of the obscured human body part.

[0070] When determining the contours and key points of occluded body parts, the appropriate positions of the key points of the unobstructed body parts are inferred based on the key points of the unobstructed body parts, combined with kinematic constraints and movement trajectories, to generate the corresponding key points. Leveraging technologies such as generative adversarial networks, the contours of the occluded body parts are constructed based on the contours of the unobstructed body parts. This generates the contours of the occluded body parts, achieving highly robust pose recovery in occluded environments, enabling pose reconstruction and part reconstruction in occluded environments, and effectively supporting kinetic energy calculations.

[0071] Step 102: Based on the kinetic energy, a suspected violent individual is screened from at least one of the observed individuals.

[0072] The kinetic energy of multiple body parts of each observed individual is compared with a preset kinetic energy, such as 50 joules.

[0073] If the kinetic energy of any body part of the observed individual is greater than the preset kinetic energy, it is determined that the corresponding body part of the observed individual may have committed physical violence, and the observed individual is a suspected violent individual.

[0074] If the kinetic energy of all parts of the observed individual's body is less than or equal to the preset kinetic energy, it is determined that the observed individual does not have any physical violence behavior and the observed individual is an ordinary observed individual.

[0075] Kinetic energy is a physical quantity that quantifies the intensity of an action. It can be used to initially screen for physically violent behaviors and identify suspected violent individuals. By making judgments based on this objective energy, potential dangers can be quickly identified, high-risk individuals can be targeted, and detection efficiency and accuracy can be improved.

[0076] Step 103 : If the movement postures of the multiple body parts of the suspected violent individual conform to the set attack posture, the suspected violent individual is determined to be a target individual with physical violence behavior.

[0077] Suspected violent individuals are screened based on kinetic energy, but solely considering kinetic energy can lead to false positives. Therefore, the system checks whether the body parts of suspected violent individuals conform to known attack postures, or set attack postures. For example, when punching or kicking, one foot generates kinetic energy greater than the preset kinetic energy, while the other foot remains stationary. This can effectively distinguish false positives in scenarios such as running, enabling further screening and identification, reducing false positives, and identifying individuals who are engaging in physical violence as dangerous individuals, thereby improving detection accuracy.

[0078] Among them, setting the attack posture limits the action posture of some or all parts of the human body. When comparing, it only needs to compare with the action posture of the corresponding part.

[0079] In some embodiments, when the part movement postures of the multiple body parts of the suspected violent individual conform to the set attack posture, before determining that the suspected violent individual is a target individual with physically violent behavior, it also includes: generating the part movement postures of the multiple body parts of the suspected violent individual based on the position data of the multiple body key points of the suspected violent individual.

[0080] Determine the movement posture of each body part of a suspected violent individual to more accurately identify aggressive actions and reduce false detections.

[0081] This application can not only identify dangerous individuals who are committing physical violence, but also identify potentially dangerous individuals who have physical violence and may harm other observing individuals. It upgrades from seeing physical violence to foreseeing physical violence, and realizes the upgrade from passive identification of physical violence to active identification and defense.

[0082] like Figure 2 As shown, Figure 2 This is a schematic diagram of a human body part movement posture that conforms to a set attack posture provided by an embodiment of the present application. Figure 1 . Figure 2 There are two observed individuals in the scenario: observed individual A and observed individual B. Based on the dynamic image, the kinetic energy of observed individual A's body part A1 is calculated to be greater than the preset kinetic energy. That is, the kinetic energy of the punch of body part A1 (left hand) is greater than the preset kinetic energy. Therefore, observed individual A is determined to be a suspected violent individual. At the same time, the movement trajectory of observed individual A's body part A1 points toward observed individual B. That is, the body part movement posture of observed individual A's body part A1 is moving toward observed individual B, which conforms to the preset attack posture of a punching attack. This further determines observed individual A, the suspected violent individual, to be the target individual who has committed physical violence. The body part that is the target of physical violence is observed individual A's body part A1. Accordingly, the target of observed individual A's physical violence is observed individual B.

[0083] like Figure 3 As shown, Figure 3 This is a schematic diagram of a human body part movement posture that conforms to a set attack posture provided by an embodiment of the present application. Figure 2 . Figure 3 There are two observed individuals in the scenario: Observer C and Observer D. Based on the dynamic imagery, the kinetic energy of Observer C's body part C1 is calculated to be greater than the preset kinetic energy. This means that the kinetic energy of the kicking kick from body part C1 (left foot) is greater than the preset kinetic energy. Therefore, Observer C is identified as a suspected violent individual. Furthermore, the trajectory of Observer C's body part C1 points toward Observer D, and body part C2 (right foot) remains stationary on the ground. The standing, kicking posture formed by Observer C's body parts C1 and C2, directed toward Observer D, conforms to the preset attack posture for a kicking attack. This leads to the identification of Observer C, the suspected violent individual, as the target of the physical violence. The body part involved in the physical violence is Observer C's body part C1. Accordingly, the target of Observer C's physical violence is Observer D.

[0084] In some embodiments, after determining that the suspected violent individual is a target individual with physically violent behavior when the movement postures of multiple body parts of the suspected violent individual conform to the set attack posture, it also includes: determining the level of violent behavior of the target individual based on the kinetic energy interval corresponding to the maximum kinetic energy among the kinetic energy of the multiple body parts of the target individual; and / or determining the level of violent behavior of the target individual based on the contact status and body size difference between the target individual and other observed individuals.

[0085] There are different levels of physical violence, and different levels of violence correspond to different degrees of danger and response measures.

[0086] In some embodiments, the kinetic energy of a human body part, the contact status (ie, distance) between the target individual and other observed individuals, and the size difference between the target individual and other observed individuals may all affect the determination of the level of violent behavior.

[0087] In some embodiments, the level of violent behavior can be determined based on the kinetic energy range corresponding to the maximum kinetic energy of a human body part, the degree of contact between the target individual and other observed individuals, the difference in body shape between the target individual and other observed individuals, etc., to clarify the degree of danger of the target individual, so as to determine response measures based on the level of violent behavior, reduce the risk of personal injury, and minimize personal harm.

[0088] In some embodiments, the violent behavior level of the target individual is determined based on the contact status and body size difference between the target individual and the other observed individuals, including: if the distance between the target individual and the other observed individuals is less than or equal to a set distance, then the violent behavior level is increased; if the height ratio of the target individual and the other observed individuals is greater than a set height ratio and / or the mass ratio is greater than a set mass ratio, then the violent behavior level is increased.

[0089] Among them, the other observed individuals used to determine the height ratio and mass ratio are observed individuals within the attack range of the target individual.

[0090] In some embodiments, if the distance between the target individual and the other observed individuals is greater than a set distance, the height ratio between the target individual and the other observed individuals is less than or equal to the set height ratio, and the mass ratio is less than or equal to the set mass ratio, then the violent behavior level is determined to be Level 1. For example, the set distance may be 0.5 meters, the set height ratio may be 1.3, and the set mass ratio may be 1.5. These values ​​are for example purposes only and may be adjusted based on actual circumstances.

[0091] In some embodiments, if the distance between the target individual and other observed individuals is less than or equal to the set distance, the height ratio between the target individual and other observed individuals is greater than the set height ratio, and the mass ratio is greater than the set mass ratio, if any of these three conditions is met, the level of violent behavior is determined to be level two.

[0092] In some embodiments, if the distance between the target individual and other observed individuals is less than or equal to the set distance, the height ratio between the target individual and other observed individuals is greater than the set height ratio, and the mass ratio is greater than the set mass ratio, if two of these three conditions are met, the level of violent behavior is determined to be level three.

[0093] In some embodiments, if the distance between the target individual and other observed individuals is less than or equal to the set distance, the height ratio of the target individual and the other observed individuals is greater than the set height ratio, and the mass ratio is greater than the set mass ratio, if all three conditions are met, the level of violent behavior is determined to be level four.

[0094] In some embodiments, the scores corresponding to each judgment item can be determined based on the kinetic energy range corresponding to the maximum kinetic energy of a human body part, the degree of contact between the target individual and other observed individuals, the difference in body shape between the target individual and other observed individuals, etc., and the scores are summed up to obtain a total score, and the total score is used to determine the level of violent behavior of the target individual.

[0095] In some embodiments, there are multiple kinetic energy ranges, each with a different score. Kinetic energy is measured in joules (J). For example, [50J, 100J) is scored 1 point, [100J, 150J) is scored 3 points, and [150J, +∞) is scored 5 points.

[0096] In some embodiments, the attack risk of the target individual can also be determined based on the movement speed, and the attack risk can be used as a criterion for assessing the level of violent behavior. For example, the maximum movement speed and initial movement speed of the body part of the target individual that is physically violent are determined, and the speed change rate is calculated based on the maximum movement speed and the initial movement speed to obtain the movement acceleration. If the maximum movement speed is greater than or equal to the first set attack speed and the movement acceleration is greater than or equal to the set attack acceleration, the attack risk is determined to be high risk. If the maximum movement speed is greater than or equal to the second set attack speed, the attack risk is determined to be medium risk, where the second set attack speed = 0.7 × the first set attack speed. This is an example. If the above two conditions are not met, the attack risk is determined to be low risk. Different attack risks correspond to different scores, which are used to assess the level of violent behavior.

[0097] In some embodiments, different distance intervals may be set, and different distance intervals may correspond to different scores. Similarly, different height ratio intervals and different mass ratio intervals may be set, and different height ratio intervals may correspond to different scores, and different mass ratio intervals may correspond to different scores.

[0098] Determine the score for each item and then calculate the total score for the level of violence. Different levels of violence correspond to different total scores, and the corresponding level of violence is determined based on the total score.

[0099] Through the level of violent behavior, the degree of danger of physical violence can be clarified, graded warnings can be issued, and timely response measures can be taken to reduce the physical violence injuries suffered by observers.

[0100] In this embodiment, the kinetic energy of multiple body parts of each observed individual in the dynamic image is first determined, quantifying the energy characteristics of the observed individual's movements. Then, based on the kinetic energy, a suspected violent individual is screened from at least one observed individual. If the motion postures of multiple body parts of the suspected violent individual match the predefined attack posture, the suspected violent individual is identified as a target individual engaging in physical violence. This dual-modal detection and identification mechanism, combining initial kinetic energy screening with posture verification, achieves multi-dimensional detection and identification, effectively improving the accuracy of detecting physical violence.

[0101] See also Figure 4 , Figure 4 : is a structural diagram of a physical violence behavior detection system provided in an embodiment of the present application. For ease of explanation, only the parts related to the embodiment of the present application are shown.

[0102] The physical violence behavior detection system 400 includes: a first determination module 401 , a screening module 402 , and a second determination module 403 .

[0103] The first determining module 401 is configured to determine kinetic energy of multiple body parts of each observed individual based on at least one observed individual in a dynamic image.

[0104] The screening module 402 is configured to screen out a suspected violent individual from at least one of the observed individuals based on the kinetic energy.

[0105] The second determining module 403 is configured to determine that the suspected violent individual is a target individual with physical violence behavior if the movement postures of the multiple body parts of the suspected violent individual conform to a set attack posture.

[0106] In some embodiments, the first determining module is specifically configured to:

[0107] Performing person detection on the dynamic image to determine at least one observed individual in the dynamic image;

[0108] Performing person matching on each of the observed individuals to obtain an image frame sequence of each of the observed individuals in the dynamic image;

[0109] determining, based on the image frame sequence of each observed individual, body part masses of the plurality of body parts of the observed individual and position data of a plurality of body key points in the image frame sequence;

[0110] Calculating the moving speeds of the plurality of human body parts of the observed individual based on the position change data of the position data of the plurality of human body key points in the image frame sequence;

[0111] The kinetic energy of the plurality of body parts of each of the observed individuals is calculated based on the moving speed and the mass of the body part.

[0112] In some embodiments, the first determining module is further configured to:

[0113] Calculating the body part mass of the plurality of body parts of the observed individual according to the image area and mass calculation coefficient of each body part of the observed individual in the image frame sequence;

[0114] The human body key points of the observed individual in each image frame in the image frame sequence are identified, and positions of the identified human body key points in the image frame sequence are determined as the position data of the human body key points.

[0115] In some embodiments, the first determining module is further configured to:

[0116] performing body part detection on each of the observed individuals based on the image frame sequence, and determining unobstructed body parts of the observed individuals when an obstructed body part of the observed individuals is detected;

[0117] Based on the unobstructed human body part, performing posture estimation on the observed individual, obtaining the part contour of the obstructed human body part and the human body key points, and determining the position data of the human body key points;

[0118] The human body part mass of the obscured human body part is calculated according to the image area corresponding to the part contour and the mass calculation coefficient.

[0119] In some embodiments, the system further comprises a posture determination module for:

[0120] Based on the position data of the multiple key points of the human body of the suspected violent individual, the part action postures of the multiple parts of the human body of the suspected violent individual are generated.

[0121] In some embodiments, the system further comprises a violence level determination module configured to:

[0122] determining the level of violent behavior of the target individual according to a kinetic energy interval corresponding to a maximum kinetic energy among the kinetic energies of the plurality of body parts of the target individual; and / or,

[0123] The violent behavior level of the target individual is determined based on the contact status and body size difference between the target individual and other observed individuals.

[0124] In some embodiments, the violence level determination module is further configured to:

[0125] If the distance between the target individual and the other observed individuals is less than or equal to the set distance, the level of the violent behavior is increased;

[0126] If the height ratio of the target individual to the other observed individuals is greater than a set height ratio and / or the mass ratio is greater than a set mass ratio, the level of the violent behavior is increased.

[0127] The physical violence behavior detection system provided in the embodiments of the present application can implement each process of the embodiments of the above-mentioned physical violence behavior detection method and can achieve the same technical effects. To avoid repetition, they are not described here.

[0128] Figure 5 : is a structural diagram of an electronic device provided in an embodiment of the present application. As shown in the figure, the electronic device 5 of this embodiment includes: at least one processor 50 ( Figure 5 Only one is shown in the figure), a memory 51 and a computer program 52 stored in the memory 51 and executable on the at least one processor 50, wherein the processor 50 implements the steps of any of the above-mentioned method embodiments when executing the computer program 52.

[0129] The electronic device 5 can be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The electronic device 5 can include, but is not limited to, a processor 50 and a memory 51. It can be understood by those skilled in the art that Figure 5 It is only an example of the electronic device 5 and does not constitute a limitation of the electronic device 5. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0130] The processor 50 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0131] The memory 51 can be an internal storage unit of the electronic device 5, such as a hard drive or memory of the electronic device 5. The memory 51 can also be an external storage device of the electronic device 5, such as a plug-in hard drive, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. Furthermore, the memory 51 can include both an internal storage unit of the electronic device 5 and an external storage device. The memory 51 is used to store the computer program and other programs and data required by the electronic device. The memory 51 can also be used to temporarily store data that has been output or is about to be output.

[0132] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0133] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0134] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0135] In the embodiments provided in this application, it should be understood that the disclosed systems / electronic devices and methods can be implemented in other ways. For example, the system / electronic device embodiments described above are merely schematic. For example, the division of the modules or units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of the system or unit, which can be electrical, mechanical or other forms.

[0136] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of the solution of this embodiment according to actual needs.

[0137] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0138] If the integrated module / unit is implemented as a software functional unit and sold or used as a standalone product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application can implement all or part of the process steps in the above-mentioned method embodiments by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium. It should be noted that the content of the computer-readable medium can be appropriately increased or decreased based on the requirements of legislation and patent practice in a jurisdiction. For example, in some jurisdictions, based on legislation and patent practice, computer-readable media does not include electric carrier signals and telecommunication signals.

[0139] The present application implements all or part of the processes in the above-mentioned embodiment methods, and may also be implemented through a computer program product. When the computer program product runs on an electronic device, the electronic device can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0140] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. 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 various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A method for detecting physical violence, characterized in that: include: determining kinetic energy of a plurality of body parts of each of the observed individuals based on at least one observed individual in the dynamic image; The kinetic energy of each of the human body parts of each observed individual is calculated according to the kinetic energy calculation formula Calculated, where is the moving speed of the human body part, is the body part mass of the human body part, the body part mass is calculated based on the image area of ​​the human body part and a mass calculation coefficient, the mass calculation coefficient is a conversion parameter for converting the image area of ​​the human body part into the body part mass, different types of human body parts have different corresponding mass calculation coefficients, and the mass calculation coefficient is calculated based on the sample mass of each human body part of the observed individual in multiple frames of reference images and the sample image area in a set image format; Based on the kinetic energy, screening out a suspected violent individual from at least one of the observed individuals; generating, based on the position data of multiple key points of the body of the suspected violent individual, body part motion postures of the multiple body parts of the suspected violent individual; the body part motion postures including movement trajectories for identifying whether the suspected violent individual has engaged in aggressive actions; In a case where the movement postures of the multiple body parts of the suspected violent individual conform to the set attack posture, the suspected violent individual is determined to be a target individual with physical violence behavior.

2. The method according to claim 1, characterized in that The determining of the kinetic energy of multiple body parts of each observed individual based on at least one observed individual in the dynamic image includes: Performing person detection on the dynamic image to determine at least one observed individual in the dynamic image; Performing person matching on each of the observed individuals to obtain an image frame sequence of each of the observed individuals in the dynamic image; determining, based on the image frame sequence of each observed individual, the body part masses of the plurality of body parts of the observed individual and the position data of the plurality of body key points in the image frame sequence; Calculating the moving speeds of the plurality of body parts of the observed individual based on position change data of the position data of the plurality of human body key points in the image frame sequence; The kinetic energy of the plurality of body parts of each of the observed individuals is calculated based on the moving speed and the mass of the body part.

3. The method according to claim 2, characterized in that Determining the body part masses of the plurality of body parts of the observed individual and the position data of the plurality of body key points in the image frame sequence according to the image frame sequence of each observed individual includes: Calculating the body part mass of the plurality of body parts of the observed individual according to the image area and the mass calculation coefficient of each body part of the observed individual in the image frame sequence; The human body key points of the observed individual in each image frame in the image frame sequence are identified, and positions of the identified human body key points in the image frame sequence are determined as the position data of the human body key points.

4. The method according to claim 2, characterized in that The method further comprises: performing body part detection on each of the observed individuals based on the image frame sequence, and determining unobstructed body parts of the observed individuals when an obstructed body part of the observed individuals is detected; Based on the unobstructed human body part, performing posture estimation on the observed individual, obtaining the part contour of the obstructed human body part and the human body key points, and determining the position data of the human body key points; The human body part mass of the obscured human body part is calculated according to the image area corresponding to the part contour and the mass calculation coefficient.

5. The method according to claim 1, wherein After determining that the suspected violent individual is a target individual who has committed physical violence, if the movement postures of the multiple body parts of the suspected violent individual conform to the set attack posture, the method further includes: determining the level of violent behavior of the target individual according to a kinetic energy interval corresponding to a maximum kinetic energy among the kinetic energies of the plurality of body parts of the target individual; and / or, The violent behavior level of the target individual is determined based on the contact status and body size difference between the target individual and other observed individuals.

6. The method according to claim 5, characterized in that Determining the level of violent behavior of the target individual based on the contact status and body size difference between the target individual and the other observed individuals includes: If the distance between the target individual and the other observed individuals is less than or equal to the set distance, the level of the violent behavior is increased; If the height ratio of the target individual to the other observed individuals is greater than a set height ratio and / or the mass ratio is greater than a set mass ratio, the level of the violent behavior is increased.

7. A physical violence detection system, characterized in that: include: A first determining module is configured to determine kinetic energy of multiple body parts of each observed individual based on at least one observed individual in the dynamic image; The kinetic energy of each of the human body parts of each observed individual is calculated according to the kinetic energy calculation formula Calculated, where is the moving speed of the human body part, is the body part mass of the human body part, the body part mass is calculated based on the image area of ​​the human body part and a mass calculation coefficient, the mass calculation coefficient is a conversion parameter for converting the image area of ​​the human body part into the body part mass, different types of human body parts have different corresponding mass calculation coefficients, and the mass calculation coefficient is calculated based on the sample mass of each human body part of the observed individual in multiple frames of reference images and the sample image area in a set image format; a screening module, configured to screen out a suspected violent individual from at least one of the observed individuals based on the kinetic energy; a posture determination module, configured to generate, based on the position data of multiple key points of the body of the suspected violent individual, body part movement postures of the suspected violent individual; the body part movement postures including movement trajectories, for identifying whether the suspected violent individual has engaged in aggressive actions; The second determination module is configured to determine that the suspected violent individual is a target individual with physical violence behavior if the movement postures of the multiple body parts of the suspected violent individual conform to a set attack posture.

8. An electronic device, characterized in that: The electronic device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the electronic device implements the method according to any one of claims 1 to 6.

9. A computer program product, characterized in that The invention comprises a computer program which, when executed, causes the method according to any one of claims 1 to 6 to be performed.

Images