CN110718067A - Violation behavior warning method and related device - Google Patents
Violation behavior warning method and related device Download PDFInfo
- Publication number
- CN110718067A CN110718067A CN201910899842.8A CN201910899842A CN110718067A CN 110718067 A CN110718067 A CN 110718067A CN 201910899842 A CN201910899842 A CN 201910899842A CN 110718067 A CN110718067 A CN 110718067A
- Authority
- CN
- China
- Prior art keywords
- violation
- target person
- target
- detection
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
- G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Alarm Systems (AREA)
- Closed-Circuit Television Systems (AREA)
- Image Analysis (AREA)
Abstract
The application discloses a violation behavior warning method and a related device. The violation behavior warning method comprises the following steps: respectively carrying out target detection on multiple frames of images shot by the camera device to obtain a detection area corresponding to at least one target person in each frame of image; judging whether at least one target person has violation behaviors or not based on the detection area; counting the judgment result of the violation behaviors of at least one target person to obtain the number of current violation times of each target person; whether violation warning information about the current target person is output is determined based on the number of violations of each current target person. According to the scheme, the warning of the illegal action can be accurately realized.
Description
Technical Field
The present application relates to the field of information technologies, and in particular, to a violation warning method and a related device.
Background
With the rapid development of information technology, electronic products such as smart phones and tablet computers are increasingly deeply penetrated into daily work and life of people, and meanwhile, the dependence of people on the electronic products is also increasingly serious. Along with the above, violations such as driving, on duty, playing electronic products in class and the like are also frequently happened, and particularly, the playing of electronic products during driving is illegal and traffic accidents are easily caused; when the electronic product is on duty, the rules and regulations of a unit are violated, and the attention of the person on duty is easily dispersed, so that the person on duty is careless, and the consequences such as production accidents and the like are possibly generated; when the electronic products are played in class, the attention of students is dispersed, and the class discipline is disturbed. Therefore, it is necessary to monitor and alarm the violation so as to standardize discipline, serious regulation and reduce the occurrence of accidents. In view of this, how to accurately implement violation warning becomes an urgent problem to be solved.
Disclosure of Invention
The technical problem mainly solved by the application is to provide a violation warning method and a related device, which can accurately realize violation warning.
In order to solve the above problems, a first aspect of the present application provides a method for alarming violation, including performing target detection on multiple frames of images captured by an image capture device, respectively, to obtain a detection area corresponding to at least one target person in each frame of image; judging whether at least one target person has violation behaviors or not based on the detection area; counting the judgment result of the violation behaviors of at least one target person to obtain the number of current violation times of each target person; whether violation warning information about the current target person is output is determined based on the number of violations of each current target person.
In order to solve the above problems, a second aspect of the present application provides an illegal behavior warning device, including a detection module, a judgment module, a statistics module, and a determination module, where the detection module is configured to perform target detection on multiple frames of images captured by a camera device, respectively, to obtain a detection area corresponding to at least one target person in each frame of image; the judgment module is used for judging whether at least one target person has violation behaviors or not based on the detection area; the statistical module is used for counting the judgment result of the violation behaviors of at least one target person to obtain the number of current violation times of each target person; the determining module is used for determining whether to output violation warning information about each current target person based on the number of violations of each current target person.
In order to solve the above problem, a third aspect of the present application provides an illegal behavior warning device, which includes a memory and a processor coupled to each other, where the processor is configured to execute program instructions stored in the memory to implement the method in the first aspect.
In order to solve the above problem, a fourth aspect of the present application provides a storage device storing program instructions executable by a processor, the program instructions being for implementing the method of the first aspect.
According to the scheme, the target detection is carried out on the multi-frame images obtained by the shooting device respectively, the detection areas corresponding to at least one target person in each frame of image are obtained, whether the violation behaviors exist in at least one target person is judged based on the detection areas, the judgment results of the violation behaviors exist in at least one target person are counted, the violation number of each current target person is obtained, whether violation warning information about the current target person is output or not is determined based on the violation number of each current target person, whether the violation behaviors exist in the target person can be tracked and judged, the judgment results of the violation behaviors existing for multiple times are fused, and the accuracy of violation warning can be improved.
Drawings
FIG. 1 is a schematic flowchart of an embodiment of a violation warning method according to the present application;
FIG. 2 is a flowchart illustrating an embodiment of step S12 in FIG. 1;
FIG. 3 is a flowchart illustrating an embodiment of step S123 in FIG. 2;
FIG. 4 is a flowchart illustrating an embodiment of step S31 in FIG. 3;
FIG. 5 is a flowchart illustrating an embodiment of step S32 in FIG. 3;
FIG. 6 is a flowchart illustrating another exemplary embodiment of a violation warning method according to the present application;
FIG. 7 is a flowchart illustrating a violation warning method according to another embodiment of the present application;
FIG. 8 is a flowchart illustrating an embodiment of step S72 in FIG. 7;
FIG. 9 is a block diagram of an embodiment of an illegal behavior warning device according to the present application;
FIG. 10 is a block diagram of another embodiment of an illegal behavior warning device according to the present application;
FIG. 11 is a block diagram of an embodiment of a memory device according to the present application.
Detailed Description
The following describes in detail the embodiments of the present application with reference to the drawings attached hereto.
In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular system structures, interfaces, techniques, etc. in order to provide a thorough understanding of the present application.
The terms "system" and "network" are often used interchangeably herein. The term "and/or" herein is merely an association describing an associated object, meaning that three relationships may exist, e.g., a and/or B, may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" herein generally indicates that the former and latter related objects are in an "or" relationship. Further, the term "plurality" herein means two or more than two.
Referring to fig. 1, fig. 1 is a schematic flowchart illustrating an exemplary violation warning method according to the present application. Specifically, the method may include the steps of:
step S11: and respectively carrying out target detection on multiple frames of images shot by the camera device to obtain a detection area corresponding to at least one target person in each frame of image.
Different types of image pickup devices can be selected according to different application scenes. For example, for a place with a dark environment and poor lighting, the camera equipment can be a night vision camera or an infrared camera; the camera device can be a general digital camera or a network camera for indoor places with bright light, such as classrooms and duty rooms; for outdoor non-blocking scenes such as roads, the camera device may be a waterproof camera, and this embodiment is not particularly limited.
Depending on the specific application scenario, the target person may include, but is not limited to: students, operators on duty, drivers, etc., and the embodiment is not particularly limited herein.
In this embodiment, a video stream captured by the image capturing device may be obtained online in Real Time through a communication Protocol such as an RTSP Protocol (Real Time Streaming Protocol), and the video stream is converted into a multi-frame image for subsequent target detection. In addition, the multi-frame image captured by the imaging device may also be obtained in an offline manner, such as offline copy, and the embodiment is not limited in this respect.
In one implementation scenario, target detection may be performed on multiple frames of images based on an algorithm combining candidate regions and deep learning, for example: R-CNN (Region based Convolutional Neural Network), Fast R-CNN, and the like. In another implementation scenario, target detection may also be performed on multiple frames of images based on a regression algorithm for deep learning, for example: yolo (young Only Look once), ssd (single shotmultitox detector), etc., and the embodiment is not limited in detail herein. The specific algorithms for R-CNN, Fast R-CNN, FasterR-CNN, YOLO, SSD are prior art in the art, and the detailed description thereof is omitted here.
The detection area is an area including the target person, for example, a rectangular area including the target person, or the like. Specifically, the detection area may be an area in which the target person moves, so as to determine whether the target person has an illegal action based on the detection area.
The number of target persons in each frame of image may be one, two, three, etc., and the embodiment is not limited in this respect. For example, for the violation warning of the operator on duty, there may be one target person in each frame of image; aiming at the violation behavior warning of the driver, one target person is in each frame of image; for the violation alarm of the student, there may be a plurality of target persons in each frame of image, and so on, and this embodiment is not illustrated here.
In the embodiment, the target detection is performed on the multiple frames of images, so that the detection area corresponding to each target person in each frame of image is obtained.
Step S12: and judging whether the at least one target person has the violation behavior or not based on the detection area.
In one implementation scenario, the detection area may be detected by a pre-trained detection model, so as to determine whether there is an illegal action. In another implementation scenario, the behavior characteristics of the target person in the detection area may be analyzed to determine whether the target person has an illegal behavior. In another implementation scenario, in order to improve the accuracy of determining whether the target person has an illegal action, the detection area may be analyzed by combining the detection model and the action characteristics, and then, whether the target person has an illegal action may be determined.
Step S13: and counting the judgment result of the violation behaviors of at least one target person to obtain the current violation times of each target person.
And counting the judgment result of whether the detected target personnel have the violation behaviors or not so as to obtain the number of current violation times of each target personnel. In an implementation scenario, in the process of determining whether the target person has an illegal action in the above steps, a corresponding number of violations may be initialized and accumulated for each target person. For example, the acquired multi-frame images are p1, p2, p3, … …, and pn, where p1 is a first frame image in the acquired multi-frame images, p2 is a second frame image in the acquired multi-frame images, and so on, pn is a current frame image in the acquired multi-frame images, and when target persons target1 and target persons target2 are detected in the frame image p1, corresponding violation times sum1 and sum2 are initialized for the target persons, and both are given initial values of 0, and by determining whether there is any violation behavior in target persons target1 and target person target2, it is determined whether the violation times sum1 and sum1 are accumulated, and so on, until the current frame image pn, the violation times of each target person at present can be counted, and the method includes: sum1, sum2, sum3 and sum4, which are not illustrated in this embodiment.
Step S14: whether violation warning information about the current target person is output is determined based on the number of violations of each current target person.
And determining whether to output violation warning information about each current target person based on the number of violations of each current target person. In an implementation scenario, whether violation warning information about the current target person is output may be determined according to the number of violations of each current target person and a violation duration preset by a user, for example, when the total violation duration preset by the user is 10 seconds, and the frame rate of an image pickup device is 25fps (frame per second), that is, when the number of violations reaches 250 times, the total violation duration reaches 10 seconds, and when the number of violations of a certain target person reaches 250 times, the violation warning information about the target person is output; or, the user may further preset the maximum violation duration to be 10 seconds, and the frame rate of the camera device to be 25fps, that is, when the number of violations of a certain target person is detected in consecutive 250 frames of images, that is, when the number of violations of each target person is counted in step S13, if the number of violations of the certain target person is detected in a certain frame, and the next frame does not have any violations, determine whether the current number of violations has been accumulated to 250 times, if not, clear, count again, and if it has reached 250 times, output violation warning information about the target person; or, the total violation duration and the maximum violation duration may be set at the same time, and if the current violation times of a certain target person both satisfy the conditions on the total violation duration and the maximum violation duration, violation warning information on the target person is output.
In another implementation scenario, whether violation warning information about the current target person is output or not may be determined according to the number of violations of each current target person and the number of violations preset by the user. For example, a user may preset the total number of violations, and if the number of violations of a certain target person reaches the total number of violations, output violation warning information about the target person; the user can also preset the maximum continuous violation times, if the violation times of a certain target person in a certain frame reach the maximum continuous violation times, violation warning information about the target person is output, and if the violation times of the target person in the certain frame do not reach the maximum continuous violation times, the violation times corresponding to the target person are reset, and violation behaviors are counted again; the user can also preset the total violation times and the maximum continuous violation times at the same time, and when a certain target person currently meets the conditions about the total violation times and the maximum continuous violation times at the same time, violation warning information about the target person is output.
According to the scheme, the target detection is carried out on the multi-frame images obtained by the shooting device respectively, the detection areas corresponding to at least one target person in each frame of image are obtained, whether the violation behaviors exist in at least one target person is judged based on the detection areas, the judgment results of the violation behaviors exist in at least one target person are counted, the violation number of each current target person is obtained, whether violation warning information about the current target person is output or not is determined based on the violation number of each current target person, whether the violation behaviors exist in the target person can be tracked and judged, the judgment results of the violation behaviors existing for multiple times are fused, and the accuracy of violation warning can be improved.
Referring to fig. 2, fig. 2 is a schematic flowchart illustrating an embodiment of step S12 in fig. 1. Specifically, the method may include the steps of:
step S121: and respectively extracting parts corresponding to the detection areas in each frame of image as detection images.
And respectively extracting parts corresponding to the detection areas in each frame of image as detection images.
Step S122: and detecting the detection image by using a detection model to obtain a candidate image suspected of having the violation.
In one implementation scenario, the detection model is obtained by neural network training using a training image set and a test image set containing the target person and the offending article. Specifically, the training image set and the test image set contain target personnel and illegal articles, and the target personnel have illegal behaviors. For example, the offending item may include, but is not limited to: smart phones, tablet computers, and the like. The illegal action can be that the target person touches a smart phone by hand, the target person touches a tablet computer by hand, and the like, so that the target person is associated with the illegal item, whether the illegal action is suspected to exist is judged through the association between the target person and the illegal item, and the accuracy of illegal action warning is further improved.
And detecting the detection image by using a detection model to obtain a candidate image suspected of having the violation.
Step S123: and judging whether the target person has illegal behaviors or not based on the behavior characteristics of the target person in the candidate image.
And analyzing the behavior characteristics of the target person in the candidate image, and judging whether the target person has an illegal behavior.
According to the scheme, the detection image is detected through the detection model, the candidate image suspected of having the violation is obtained, and whether the violation exists in the target person is judged based on the behavior characteristics of the target person in the candidate image, so that double detection is performed, and the accuracy of violation warning is further guaranteed.
Referring to fig. 3, fig. 3 is a schematic flowchart illustrating an embodiment of step S123 in fig. 2. Specifically, the method may include the steps of:
step S31: and analyzing the behavior characteristics of the target person in the candidate image.
When the target person has an illegal behavior, the sight direction of the target person or the face direction of the target person is basically overlapped with the direction perpendicular to the illegal object, or an included angle in a certain range is formed. By using the characteristics, the behavior characteristics of the target person in the candidate image can be analyzed, and in particular, with reference to fig. 4, the method includes the following steps:
step S311: and analyzing the candidate image to obtain a first vector of the target person.
The first vector is any one of the sight line direction of the target person and the face direction of the target person.
Step S312: the candidate image is analyzed to obtain an offending item in the candidate image and a second vector perpendicular to the offending item is determined.
Offending items include, but are not limited to: smart phones, tablet computers.
In this embodiment, the direction of the second vector is a direction approaching the target person.
Step S313: and acquiring a vector included angle between the first vector and the second vector.
And acquiring a vector included angle between the first vector and the second vector based on the first vector and the second vector obtained by analysis.
Step S32: and judging whether the behavior characteristics accord with preset violation behavior characteristics. If so, go to step S33, otherwise, go to step S34.
By utilizing the characteristics, whether the behavior characteristics of the target personnel meet the preset violation behavior characteristics can be judged. Referring to fig. 5, the method includes the following steps:
step S321: and judging whether the vector included angle is larger than a preset included angle or not. If so, go to step S322, otherwise go to step S323.
The preset included angle may be 100 degrees, 120 degrees, 140 degrees, and the like, which is not illustrated herein. If the vector included angle between the first vector and the second vector is larger than the preset included angle, the behavior feature of the target person can be determined to accord with the preset violation behavior feature, and otherwise, the behavior feature of the target person can be determined to not accord with the preset violation behavior feature.
Step S322: and determining that the behavior characteristic accords with a preset violation behavior characteristic.
When the vector included angle between the first vector and the second vector is larger than the preset included angle, the behavior feature of the target person can be determined to accord with the preset violation behavior feature.
Step S323: determining that the behavior feature does not conform to the preset violation feature.
When the vector included angle between the first vector and the second vector is smaller than or equal to the preset included angle, the behavior feature of the target person can be determined not to accord with the preset violation behavior feature.
Step S33: and determining that the target person has the violation behavior.
When the behavior characteristics of the target person meet the preset violation behavior characteristics, the target person can be determined to have violation behavior.
Step S34: and determining that the target person does not have the violation.
When the behavior feature of the target person does not accord with the preset violation feature, it can be determined that the target person does not have the violation.
According to the scheme, the illegal behavior is detected by combining the detection model and the behavior characteristics of the target personnel, and the accuracy of illegal behavior warning is improved.
Referring to fig. 6, fig. 6 is a schematic flowchart illustrating another embodiment of the violation warning method according to the present application. Specifically, the method may include:
step S61: and respectively carrying out target detection on multiple frames of images shot by the camera device to obtain a detection area corresponding to at least one target person in each frame of image.
See step S11 for details.
Step S62: and judging whether the at least one target person has the violation behavior or not based on the detection area.
See step S12 for details.
Step S63: and counting the judgment result of the violation behaviors of at least one target person to obtain the current violation times of each target person.
See step S13 for details.
Step S64: and judging whether the violation times of the current target personnel are greater than or equal to a preset threshold value, if so, executing the step S65, and otherwise, executing the step S68.
After target detection is carried out on the current frame image and whether violation behaviors exist in target personnel is judged, whether the violation times of each target personnel are larger than a preset threshold value is counted. The preset threshold may be the total number of violations or the maximum number of consecutive violations, which may specifically refer to step S14, and this embodiment is not described herein again. If the value is larger than or equal to the preset threshold value, the corresponding target person can be considered to have the violation, otherwise, the corresponding target person can be considered to have no violation.
Step S65: and outputting violation warning information about the current target personnel.
When the value is greater than or equal to the preset threshold, outputting violation warning information about the current target person, for example, "please notice that there is a violation at XXX" and the like, where the violation warning information may be in the form of text, voice, picture and the like, and the embodiment is not limited specifically herein.
In an implementation scenario, the violation warning information may also be sent to a terminal of a contact person pre-associated with the target person, for example, when the target person is an on-duty person, the violation warning information may be sent to a terminal of an on-duty supervisor and a personnel supervisor; when the target person is a student, violation warning information can be sent to terminals for teaching teachers and parents of the student; when the target person is a driver, violation warning information may be sent to a traffic management department, and the like, which is not illustrated here.
Step S66: and clearing the violation times of the target personnel corresponding to the violation warning information.
In this embodiment, in order to support the warning of repeated violation of the target person, after the violation warning information of the target person is output, the violation times of the target person corresponding to the violation warning information may also be cleared, so as to perform a new round of violation warning on the target person.
Step S67: step S61 and subsequent steps are re-executed.
After the violation times of the corresponding target persons are cleared, step S61 and the subsequent steps are re-executed, so that the detection and the warning of the violation behaviors are continued for each target person.
Step S68: step S61 and subsequent steps are re-executed.
And when the number of violation times of the target personnel is smaller than the preset threshold, re-executing the step S61 and the subsequent steps, so that the detection and the alarm of the violation behaviors can be continuously performed on each target personnel.
Referring to fig. 7, fig. 7 is a schematic flowchart illustrating a violation warning method according to another embodiment of the present application. Specifically, the method may include the steps of:
step S71: and respectively carrying out target detection on the multi-frame images to obtain detection areas corresponding to target personnel.
In this embodiment, the head and shoulder detection may be performed on the multi-frame image to obtain a head and shoulder area corresponding to the at least one target person, and the head and shoulder area may be expanded according to a preset proportional relationship to obtain a detection area corresponding to the at least one target person.
In one implementation scenario, head and shoulder detection is performed on multiple frames of images based on a deep learning head and shoulder detection model, so that a head and shoulder area of a target person is obtained.
In one implementation scenario, in order to enable the detection area after the head and shoulder area is expanded to include the activity area of the target person, the detection area may be obtained by expanding the detection area in a first direction, for example, in the left and right directions, according to a first preset ratio based on the length of the head and shoulder area, and expanding the detection area in a second direction, for example, in the downward direction, according to a second preset ratio based on the width of the head and shoulder area. In this embodiment, the first preset ratio is smaller than the second preset ratio, for example, the first preset ratio may be 50%, and the second preset ratio may be 150%, which is not illustrated in this embodiment.
Step S72: and marking a target identifier for the detection area based on the target personnel in the detection area, wherein the target identifier corresponds to the target personnel one by one.
Referring to fig. 8, in order to mark different target persons with different target identifiers, the following steps can be adopted:
step S721: and judging whether the current frame image is the first frame in the multi-frame image. If so, go to step S722, otherwise go to step S723.
Judging whether the current frame image is the first frame in the multi-frame image, if so, assigning a new target identifier to the detection area in the current frame image, and if not, further determining whether a target person in the detection area in the current frame image newly appears in the current frame image so as to determine whether to assign a new target identifier or an old target identifier to the target person.
Step S722: a new object identifier is marked for the detection area in the current frame image.
If the current frame image is the first frame of the multi-frame image, a new target identifier is marked for the detection area in the current frame image. For example, the current frame image is a multi-frame image { f1,f2,f3,…,fNThe first frame f in (f)1In the frame image f1To obtain a detection region R by performing target detection1Then is the detection region R1A new object identifier, e.g., ID 01, is marked.
Step S723: and judging whether the target person in the detection area in the current frame image exists in the detection area of the image before the current frame image, if so, executing step 724, and otherwise, executing step 725.
If the current frame image is not the first frame in the multi-frame image, whether the target person in the detection area in the current frame image exists in the detection area of the image before the current frame image is continuously judged, if yes, the detection area in the current frame image is marked as a target identifier corresponding to the detection area in which the target person exists, and if not, a new target identifier is marked for the detection area in the current frame image.
Step S724: and marking the detection area in the current frame image as a target identifier corresponding to the detection area where the target person exists.
And if the target person in the detection area in the current frame image already exists in the detection area of the image before the current frame image, marking the detection area in the current frame image as a target identifier corresponding to the detection area in which the target person already exists. For example, the current frame image fn(N is a natural number greater than 1 and smaller than N), it is obvious that the current frame image fnNot a plurality of frame images f1,f2,f3,…,fNIf the current frame image f is the first frame in the sequencenDetection region R ofnHas appeared at f1Detection region R of1In (3), the current frame image fnDetection region R in (1)nDetection region R marked as target person already existing1The corresponding destination identifier ID _ 01.
Step S725: a new object identifier is marked for the detection area in the current frame image.
If the target person in the detection area in the current frame image does not exist in the detection area of the image before the current frame image, a new target identifier is marked for the detection area in the current frame image. For example, the current frame image fn(N is a natural number greater than 1 and smaller than N), it is obvious that currentlyFrame image fnNot a plurality of frame images f1,f2,f3,…,fNIf the current frame image f is the first frame in the sequencenDetection region R ofnDoes not appear in { f1,f2,f3,…,fn-1In any frame of image, it is the current frame of image fnDetection region R ofnA new target identifier ID _ n is marked.
According to the scheme, different target identifiers can be marked for each detection area in the multi-frame image according to different target people.
Step S73: and judging whether the at least one target person has the violation behavior or not based on the detection area.
Please refer to step S12.
Step S74: and counting the number of times of violations detected by the detection area corresponding to each target identifier.
And counting the number of times of violations detected by the detection area corresponding to each target identifier. For example, the number of violations { sum _01, sum _02, sum _03, …, sum _ M } for which the detection region corresponding to each target identifier { ID _01, ID _02, ID _03, …, ID _ M } has been detected is counted.
Step S75: and determining the number of current violations of each target person based on the corresponding relation between each target identifier and the target person.
Since each target identifier is substantially in one-to-one correspondence with the target person, the number of current violations of each target person can be determined based on the correspondence between each target identifier and the target person.
Step S76: whether violation warning information about the current target person is output is determined based on the number of violations of each current target person.
See step S14 for details.
According to the scheme, different target identifiers are marked for each detection area according to different target personnel, so that whether the target personnel have violation behaviors or not is judged on the basis of detection of the detection areas, the violation times of different target personnel are obtained, and whether violation alarm information about each current target personnel is output or not can be determined on the basis of the violation times of each current target personnel.
Referring to fig. 9, fig. 9 is a schematic diagram of a framework of an embodiment of an illegal behavior warning device 90 according to the present application. The violation warning device 90 includes a detection module 91, a determination module 92, a statistics module 93, and a determination module 94, where the detection module 91 is configured to perform target detection on multiple frames of images captured by the image capture device, respectively, to obtain a detection area corresponding to at least one target person in each frame of image, the determination module 92 is configured to determine whether there is a violation on at least one target person based on the detection area, the statistics module 93 is configured to perform statistics on a determination result that there is a violation on at least one target person, to obtain a current number of violations on each target person, and the determination module 94 is configured to determine whether to output violation warning information about the current target person based on the current number of violations on each target person.
According to the scheme, the target detection is carried out on the multi-frame images obtained by the shooting device respectively, the detection areas corresponding to at least one target person in each frame of image are obtained, whether the violation behaviors exist in at least one target person is judged based on the detection areas, the judgment results of the violation behaviors exist in at least one target person are counted, the violation number of each current target person is obtained, whether violation warning information about the current target person is output or not is determined based on the violation number of each current target person, whether the violation behaviors exist in the target person can be tracked and judged, the judgment results of the violation behaviors existing for multiple times are fused, and the accuracy of violation warning can be improved.
In some embodiments, the determining module 92 includes an extracting sub-module configured to extract, as the detection images, portions corresponding to the detection areas in each frame of image, respectively, the determining module 92 further includes an obtaining sub-module configured to detect the detection images by using the detection model, and obtain candidate images suspected of having the violation, and the determining module 92 further includes a determining sub-module configured to determine whether the violation exists in the target person based on behavior characteristics of the target person in the candidate images.
In some embodiments, the judgment sub-module includes an analysis unit configured to analyze the behavior feature of the target person in the candidate image, and the judgment sub-module further includes a first judgment unit configured to judge whether the behavior feature meets a preset violation feature, determine that the target person has a violation if the behavior feature meets the preset violation feature, and determine that the target person does not have a violation if the behavior feature does not meet the preset violation feature.
In some embodiments, the analysis unit is specifically configured to analyze a first vector of the candidate image acquisition target person, analyze the candidate image acquisition target person to acquire an illegal item in the candidate image, determine a second vector perpendicular to the illegal item, and acquire a vector included angle between the first vector and the second vector; the first judging unit is specifically configured to judge whether the vector included angle is greater than a preset included angle, determine that the behavior feature conforms to a preset violation feature if the vector included angle is greater than the preset included angle, and determine that the behavior feature does not conform to the preset violation feature if the vector included angle is not greater than the preset included angle. In one implementation scenario, the first vector is any one of a direction of a line of sight of the target person and a direction of a face of the target person. In one implementation scenario, the offending item includes a smartphone, a tablet computer.
In some embodiments, the violation warning device 90 further includes an output module, configured to output violation warning information about the current target person when the determining module 94 determines that the violation number of the current target person is greater than or equal to a preset threshold, and the determining module 94 is further configured to, in combination with the detecting module 91, the judging module 92, and the counting module 93, re-perform target detection on multiple frames of images captured by the image capturing device when the violation number of the current target person is determined to be less than the preset threshold, to obtain a detection area corresponding to at least one target person in each frame of image, and perform subsequent steps.
In some embodiments, the violation warning device 90 further includes a resetting module, configured to clear the number of violations of the target person corresponding to the violation warning information after the violation warning information about the current target person is output, and perform target detection on multiple frames of images captured by the image pickup device again by using the detection module 91, the judgment module 92, the statistics module 93, and the determination module 94, to obtain a detection area corresponding to at least one target person in each frame of image, and subsequent steps.
In some embodiments, the detection module 91 further includes a detection sub-module configured to perform target detection on the multiple frames of images respectively to obtain a detection area corresponding to the target person, and the detection module 91 further includes a marking sub-module configured to mark a target identifier on the detection area based on the target person in the detection area, where the target identifier corresponds to the target person one to one; the counting module 93 further includes a counting submodule configured to count the number of violations detected in the detection area corresponding to each target identifier, and the counting module 93 further includes a corresponding submodule configured to determine the number of violations of each target person based on a corresponding relationship between each target identifier and the target person.
In some embodiments, the detection sub-module further includes a detection unit configured to perform head and shoulder detection on the multiple frames of images respectively to obtain a head and shoulder area corresponding to the at least one target person, and the detection sub-module further includes an extension unit configured to extend the head and shoulder area according to a preset proportional relationship to obtain a detection area corresponding to the at least one target person.
In some embodiments, the extension unit is specifically configured to extend in a first direction according to a first preset proportion based on the length of the head-shoulder region, and extend in a second direction according to a second preset proportion based on the width of the head-shoulder region, so as to obtain the detection region. In one implementation scenario, the first direction includes a left direction and/or a right direction, and the second direction includes a down direction; in one implementation scenario, the first predetermined ratio is less than the second predetermined ratio.
In some embodiments, the marking sub-module includes a second determination unit configured to determine whether the current frame image is a first frame of the multiple frame images, the marking sub-module further includes a marking unit configured to mark a new target identifier for the detection area in the current frame image when the current frame image is the first frame of the multiple frame images, the second determination unit is further configured to continue to determine whether a target person in the detection area in the current frame image is already present in the detection area of the image before the current frame image when the current frame image is not the first frame of the multiple frame images, the marking unit is further configured to mark the detection area in the current frame image with a new target identifier when the target person in the detection area in the current frame image is not present in the detection area of the image before the current frame image, and the marking unit is further configured to mark the detection area in the current frame image with a new target person already present in the detection area of the image before the current frame image And when the current frame image is detected, marking the detection area in the current frame image as a target identifier corresponding to the detection area where the target person exists.
Referring to fig. 10, fig. 10 is a schematic diagram of a framework of an embodiment of the violation warning device 100 according to the present application. In this embodiment, the violation alerting device 100 includes a memory 110 and a processor 120 coupled to each other, and the processor 120 is configured to execute program instructions stored in the memory 110 to implement steps in any of the above-described violation alerting method embodiments.
In particular, the processor 120 is configured to control itself and the memory 110 to implement the steps in any of the above-described embodiments of the violation alerting method. Processor 120 may also be referred to as a CPU (Central Processing Unit). The processor 120 may be an integrated circuit chip having signal processing capabilities. The Processor 120 may also be a general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete gate or transistor logic, discrete hardware components. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. In addition, processor 120 may be commonly implemented by multiple integrated circuit chips.
In this embodiment, the processor 120 is configured to perform target detection on multiple frames of images captured by the image capture device, respectively, to obtain a detection area corresponding to at least one target person in each frame of image, the processor 120 is further configured to determine whether the at least one target person has an illegal action based on the detection area, the processor 120 is further configured to count a determination result of the at least one target person having the illegal action, to obtain the number of violations of each current target person, and the processor 120 is further configured to determine whether to output violation warning information about the current target person based on the number of violations of each current target person.
According to the scheme, the target detection is carried out on the multi-frame images obtained by the shooting device respectively, the detection areas corresponding to at least one target person in each frame of image are obtained, whether the violation behaviors exist in at least one target person is judged based on the detection areas, the judgment results of the violation behaviors exist in at least one target person are counted, the violation number of each current target person is obtained, whether violation warning information about the current target person is output or not is determined based on the violation number of each current target person, whether the violation behaviors exist in the target person can be tracked and judged, the judgment results of the violation behaviors existing for multiple times are fused, and the accuracy of violation warning can be improved.
In some embodiments, the processor 120 is further configured to extract, as detection images, portions corresponding to the detection areas in each frame of image, respectively, the processor 120 is further configured to detect the detection images by using a detection model, and obtain candidate images suspected of having an illegal action, and the processor 120 is further configured to determine whether the target person has the illegal action based on behavior features of the target person in the candidate images.
In some embodiments, the processor 120 is further configured to analyze the behavior feature of the target person in the candidate image, and the processor 120 is further configured to determine whether the behavior feature meets a preset violation behavior feature, determine that the target person has a violation behavior if the behavior feature meets the preset violation behavior feature, and determine that the target person does not have a violation behavior if the behavior feature does not meet the preset violation behavior feature.
In some embodiments, the processor 120 is further configured to analyze the candidate image to obtain a first vector of the target person, the processor 120 is further configured to analyze the candidate image to obtain an illegal item in the candidate image, and determine a second vector perpendicular to the illegal item, the processor 120 is further configured to obtain a vector included angle between the first vector and the second vector, the processor 120 is further configured to determine whether the vector included angle is greater than a preset included angle, if so, it is determined that the behavior feature conforms to the preset illegal behavior feature, and if not, it is determined that the behavior feature does not conform to the preset illegal behavior feature. In one implementation scenario, the first vector is any one of a direction of a line of sight of the target person and a direction of a face of the target person. In one implementation scenario, the offending item includes a smartphone, a tablet computer.
In some embodiments, the detection model is obtained by neural network training using a training image set and a test image set containing the target person and the offending item.
In some embodiments, the processor 120 is further configured to output violation warning information about the current target person if it is determined that the number of violations of the current target person is greater than or equal to a preset threshold, and the processor 120 is further configured to perform target detection on multiple frames of images captured by the image capturing device again if it is determined that the number of violations of the current target person is less than the preset threshold, so as to obtain a detection area corresponding to at least one target person in each frame of image, and subsequent steps.
In some embodiments, the processor 120 is further configured to, after outputting violation warning information about the current target person, clear the violation times of the target person corresponding to the violation warning information, and perform target detection on multiple frames of images captured by the image capture device again to obtain a detection area corresponding to at least one target person in each frame of image, and perform subsequent steps.
In some embodiments, the processor 120 is further configured to perform target detection on multiple frames of images respectively to obtain detection areas corresponding to target persons, the processor 120 is further configured to mark a target identifier on the detection area based on the target persons in the detection area, where the target identifier corresponds to the target persons one to one, the processor 120 is further configured to count the number of violations that violation behaviors are detected in the detection area corresponding to each target identifier, and the processor 120 is further configured to determine the number of violations of each target person currently based on a corresponding relationship between each target identifier and the target person.
In some embodiments, the processor 120 is further configured to perform head and shoulder detection on the multiple frames of images respectively to obtain a head and shoulder area corresponding to the at least one target person, and the processor 120 is further configured to expand the head and shoulder area according to a preset proportional relationship to obtain a detection area corresponding to the at least one target person.
In some embodiments, the processor 120 is further configured to expand in a first direction according to a first preset ratio based on the length of the head-shoulder region, and the processor 120 is further configured to expand in a second direction according to a second preset ratio based on the width of the head-shoulder region, so as to obtain the detection region. In one implementation scenario, the first direction includes a left direction and/or a right direction, and the second direction includes a down direction. In one implementation scenario, the first predetermined ratio is less than the second predetermined ratio.
In some embodiments, the processor 120 is further configured to determine whether the current frame image is a first frame of a multi-frame image, the processor 120 is further configured to mark a new target identifier for the detection area in the current frame image if the determination result is yes, the processor 120 is further configured to continue to determine whether the target person in the detection area in the current frame image already exists in the detection area of the image before the current frame image if the determination result is no, the processor 120 is further configured to mark a new target identifier for the detection area in the current frame image if the determination result is no, and the processor 120 is further configured to mark the detection area in the current frame image as a target identifier corresponding to the detection area in which the target person already exists if the determination result is yes.
In some embodiments, the violation warning device 100 further includes an image capturing device for capturing a plurality of frames of images.
Referring to fig. 11, fig. 11 is a block diagram illustrating a memory device 1100 according to an embodiment of the present application. The memory device 1100 stores program instructions 1110 capable of being executed by the processor, the program instructions 1110 being for implementing the steps in any of the above-described violation alerting method embodiments.
According to the scheme, the target detection is carried out on the multi-frame images obtained by the shooting device respectively, the detection areas corresponding to at least one target person in each frame of image are obtained, whether the violation behaviors exist in at least one target person is judged based on the detection areas, the judgment results of the violation behaviors exist in at least one target person are counted, the violation number of each current target person is obtained, whether violation warning information about the current target person is output or not is determined based on the violation number of each current target person, whether the violation behaviors exist in the target person can be tracked and judged, the judgment results of the violation behaviors existing for multiple times are fused, and the accuracy of violation warning can be improved.
In the several embodiments provided in the present application, it should be understood that the disclosed method and apparatus may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of a module or a unit is merely a logical division, and an actual implementation may have another division, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some interfaces, and may be in an electrical, mechanical or other form.
Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be substantially implemented or contributed to by the prior art, or all or part of the technical solution may be embodied in a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, a network device, or the like) or a processor (processor) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
Claims (17)
1. A violation warning method is characterized by comprising the following steps:
respectively carrying out target detection on multiple frames of images shot by the camera device to obtain a detection area corresponding to at least one target person in each frame of image;
judging whether the at least one target person has violation behaviors or not based on the detection area;
counting the judgment result of the violation behaviors of the at least one target person to obtain the current violation times of each target person;
and determining whether to output violation warning information about each current target person based on the number of current violations of each current target person.
2. The violation behavior warning method according to claim 1, wherein the determining whether the violation behavior exists for the at least one target person based on the detection area comprises:
respectively extracting parts corresponding to the detection areas in each frame of image as detection images;
detecting the detection image by using a detection model to obtain a candidate image suspected of having violation behaviors;
and judging whether the target person has illegal behaviors or not based on the behavior characteristics of the target person in the candidate image.
3. The violation behavior warning method according to claim 2, wherein determining whether the target person has a violation behavior based on the behavior feature of the target person in the candidate image comprises:
analyzing the behavior characteristics of the target person in the candidate image;
judging whether the behavior characteristics accord with preset violation behavior characteristics or not;
if so, determining that the target personnel has violation behaviors;
if not, determining that the target personnel has no illegal behaviors.
4. The violation behavior warning method according to claim 3, wherein the analyzing the behavior feature of the target person in the candidate image comprises:
analyzing the candidate image to obtain a first vector of the target person;
analyzing the candidate image to obtain an illegal article in the candidate image, and determining a second vector perpendicular to the illegal article;
acquiring a vector included angle between the first vector and the second vector;
the judging whether the behavior characteristic accords with a preset violation behavior characteristic comprises the following steps:
judging whether the vector included angle is larger than a preset included angle or not;
if so, determining that the behavior characteristic accords with a preset violation behavior characteristic;
if not, determining that the behavior characteristic does not accord with the preset violation behavior characteristic.
5. The violation behavior warning method according to claim 4, wherein the first vector is any one of a direction of a line of sight of the target person and a direction of a face of the target person; and/or the presence of a gas in the gas,
the illegal articles comprise a smart phone and a tablet personal computer.
6. The violation behavior warning method according to claim 2, wherein the detection model is obtained by performing neural network training using a training image set and a test image set comprising the target person and the violation item.
7. The violation behavior warning method according to claim 1, wherein the determining whether to output violation warning information about the current target person based on the number of violations for each current target person comprises:
if the number of violation times of the current target personnel is determined to be larger than or equal to a preset threshold value, outputting violation warning information about the current target personnel;
and if the number of violation times of the target person is smaller than the preset threshold value, re-executing the step of respectively carrying out target detection on the multiple frames of images shot by the camera device to obtain a detection area corresponding to at least one target person in each frame of image and the subsequent steps.
8. The violation behavior warning method according to claim 7, wherein after outputting violation warning information about the current target person, the method further comprises:
and clearing the violation times of the target personnel corresponding to the violation warning information, and re-executing the step of performing target detection on the multi-frame images obtained by shooting by the camera respectively to obtain a detection area corresponding to at least one target personnel in each frame of image and the subsequent steps.
9. The violation behavior warning method according to claim 1, wherein the step of performing target detection on the multiple frames of images captured by the image capturing device respectively to obtain a detection area corresponding to at least one target person in each frame of image comprises:
respectively carrying out target detection on the multi-frame images to obtain detection areas corresponding to the target personnel;
marking a target identifier for the detection area based on target persons in the detection area, wherein the target identifier corresponds to the target persons one by one;
the counting of the judgment result of the violation behaviors of the at least one target person to obtain the current violation times of each target person comprises:
counting the number of violations of the violation behaviors detected in the detection area corresponding to each target identifier;
and determining the number of current violations of each target person based on the corresponding relation between each target identifier and the target person.
10. The violation behavior warning method according to claim 9, wherein the performing target detection on the multiple frames of images respectively and acquiring the detection areas corresponding to the target persons comprises:
respectively carrying out head and shoulder detection on the multi-frame images to obtain a head and shoulder area corresponding to the at least one target person;
and expanding the head and shoulder area according to a preset proportional relation to obtain a detection area corresponding to the at least one target person.
11. The violation behavior warning method according to claim 10, wherein the expanding the head-shoulder area according to the preset proportional relationship to obtain the detection area corresponding to the at least one target person comprises:
expanding the head and shoulder area to a first direction according to a first preset proportion based on the length of the head and shoulder area;
and expanding the head and shoulder area to a second direction according to a second preset proportion based on the width of the head and shoulder area to obtain the detection area.
12. The violation warning method according to claim 11, wherein said first direction comprises a left direction and/or a right direction, and said second direction comprises a down direction; and/or the presence of a gas in the gas,
the first preset proportion is smaller than the second preset proportion.
13. The violation behavior warning method according to claim 9, wherein said marking the detection area with a target identifier based on a target person in the detection area comprises:
judging whether the current frame image is the first frame in the multi-frame image;
if yes, marking a new target identifier for the detection area in the current frame image;
if not, judging whether the target person in the detection area in the current frame image already exists in the detection area of the image before the current frame image;
if not, marking a new target identifier for the detection area in the current frame image;
if so, marking the detection area in the current frame image as a target identifier corresponding to the detection area where the target person exists.
14. An illegal action warning device, comprising:
the detection module is used for respectively carrying out target detection on multiple frames of images shot by the camera device to obtain a detection area corresponding to at least one target person in each frame of image;
the judging module is used for judging whether the at least one target person has violation behaviors or not based on the detection area;
the statistical module is used for counting the judgment result of the violation behaviors of the at least one target person to obtain the current violation times of each target person;
and the determining module is used for determining whether violation warning information about the current target personnel is output or not based on the number of current violations of each target personnel.
15. An violation alerting device comprising a memory and a processor coupled to each other, the processor being configured to execute program instructions stored in the memory to implement the violation alerting method of any of claims 1-13.
16. The violation behavior warning device according to claim 15, further comprising an image capturing device for capturing a plurality of frames of images.
17. A storage device storing program instructions executable by a processor to implement a violation warning method according to any one of claims 1-13.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910899842.8A CN110718067A (en) | 2019-09-23 | 2019-09-23 | Violation behavior warning method and related device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910899842.8A CN110718067A (en) | 2019-09-23 | 2019-09-23 | Violation behavior warning method and related device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110718067A true CN110718067A (en) | 2020-01-21 |
Family
ID=69210749
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910899842.8A Pending CN110718067A (en) | 2019-09-23 | 2019-09-23 | Violation behavior warning method and related device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110718067A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111666945A (en) * | 2020-05-11 | 2020-09-15 | 深圳力维智联技术有限公司 | Storefront violation identification method and device based on semantic segmentation and storage medium |
| CN112464818A (en) * | 2020-11-27 | 2021-03-09 | 北京软通智慧城市科技有限公司 | Kitchen supervision method, device, equipment and storage medium |
| CN113255542A (en) * | 2021-06-02 | 2021-08-13 | 北京容联易通信息技术有限公司 | Illegal object identity recognition method based on illegal behavior recognition |
| CN113408464A (en) * | 2021-06-30 | 2021-09-17 | 深圳市商汤科技有限公司 | Behavior detection method and device, electronic equipment and storage medium |
| CN113449581A (en) * | 2020-03-24 | 2021-09-28 | 杭州海康威视数字技术股份有限公司 | Target area detection method and device and electronic equipment |
| CN113628370A (en) * | 2021-08-05 | 2021-11-09 | 国家核安保技术中心 | Intelligent protection channel control system for electronic equipment |
| CN114092889A (en) * | 2022-01-10 | 2022-02-25 | 深圳市明源云科技有限公司 | Violation detection method and device, electronic equipment and readable storage medium |
| CN115484063A (en) * | 2022-08-12 | 2022-12-16 | 国家管网集团北方管道有限责任公司 | Network security prevention and control method and system for industrial control system |
| CN116092280A (en) * | 2023-02-07 | 2023-05-09 | 深圳市冠标科技发展有限公司 | Supervision method and device based on remote communication |
| WO2024027330A1 (en) * | 2022-08-04 | 2024-02-08 | 深圳市震有智联科技有限公司 | Real-time monitoring method based on user behavior, and related device |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090243844A1 (en) * | 2006-05-31 | 2009-10-01 | Nec Corporation | Suspicious activity detection apparatus and method, and program and recording medium |
| CN106133750A (en) * | 2014-02-04 | 2016-11-16 | 弗劳恩霍夫应用研究促进协会 | For determining the 3D rendering analyzer of direction of visual lines |
| CN106778528A (en) * | 2016-11-24 | 2017-05-31 | 四川大学 | A kind of method for detecting fatigue driving based on gaussian pyramid feature |
| CN106875310A (en) * | 2017-04-19 | 2017-06-20 | 哈尔滨理工大学 | Smart classroom teaching multifunction system based on intelligent video |
| CN107301384A (en) * | 2017-06-09 | 2017-10-27 | 湖北天业云商网络科技有限公司 | A kind of driver takes phone behavioral value method and system |
| CN107610393A (en) * | 2017-09-21 | 2018-01-19 | 深圳市鑫汇达机械设计有限公司 | A kind of intelligent office monitoring system |
| CN109284737A (en) * | 2018-10-22 | 2019-01-29 | 广东精标科技股份有限公司 | A kind of students ' behavior analysis and identifying system for wisdom classroom |
| CN109614939A (en) * | 2018-12-13 | 2019-04-12 | 四川长虹电器股份有限公司 | " playing mobile phone " behavioral value recognition methods based on human body attitude estimation |
| CN109685083A (en) * | 2019-01-09 | 2019-04-26 | 安徽睿极智能科技有限公司 | The multi-dimension testing method of driver's driving Misuse mobile phone |
| CN110032947A (en) * | 2019-03-22 | 2019-07-19 | 深兰科技（上海）有限公司 | A kind of method and device that monitor event occurs |
| CN110070003A (en) * | 2019-04-01 | 2019-07-30 | 浙江大华技术股份有限公司 | The method and relevant apparatus that unusual checking and light stream autocorrelation determine |
| CN110135282A (en) * | 2019-04-25 | 2019-08-16 | 沈阳航空航天大学 | A kind of examinee based on depth convolutional neural networks model later plagiarizes cheat detection method |
-
2019
- 2019-09-23 CN CN201910899842.8A patent/CN110718067A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090243844A1 (en) * | 2006-05-31 | 2009-10-01 | Nec Corporation | Suspicious activity detection apparatus and method, and program and recording medium |
| CN106133750A (en) * | 2014-02-04 | 2016-11-16 | 弗劳恩霍夫应用研究促进协会 | For determining the 3D rendering analyzer of direction of visual lines |
| CN106778528A (en) * | 2016-11-24 | 2017-05-31 | 四川大学 | A kind of method for detecting fatigue driving based on gaussian pyramid feature |
| CN106875310A (en) * | 2017-04-19 | 2017-06-20 | 哈尔滨理工大学 | Smart classroom teaching multifunction system based on intelligent video |
| CN107301384A (en) * | 2017-06-09 | 2017-10-27 | 湖北天业云商网络科技有限公司 | A kind of driver takes phone behavioral value method and system |
| CN107610393A (en) * | 2017-09-21 | 2018-01-19 | 深圳市鑫汇达机械设计有限公司 | A kind of intelligent office monitoring system |
| CN109284737A (en) * | 2018-10-22 | 2019-01-29 | 广东精标科技股份有限公司 | A kind of students ' behavior analysis and identifying system for wisdom classroom |
| CN109614939A (en) * | 2018-12-13 | 2019-04-12 | 四川长虹电器股份有限公司 | " playing mobile phone " behavioral value recognition methods based on human body attitude estimation |
| CN109685083A (en) * | 2019-01-09 | 2019-04-26 | 安徽睿极智能科技有限公司 | The multi-dimension testing method of driver's driving Misuse mobile phone |
| CN110032947A (en) * | 2019-03-22 | 2019-07-19 | 深兰科技（上海）有限公司 | A kind of method and device that monitor event occurs |
| CN110070003A (en) * | 2019-04-01 | 2019-07-30 | 浙江大华技术股份有限公司 | The method and relevant apparatus that unusual checking and light stream autocorrelation determine |
| CN110135282A (en) * | 2019-04-25 | 2019-08-16 | 沈阳航空航天大学 | A kind of examinee based on depth convolutional neural networks model later plagiarizes cheat detection method |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113449581A (en) * | 2020-03-24 | 2021-09-28 | 杭州海康威视数字技术股份有限公司 | Target area detection method and device and electronic equipment |
| CN111666945A (en) * | 2020-05-11 | 2020-09-15 | 深圳力维智联技术有限公司 | Storefront violation identification method and device based on semantic segmentation and storage medium |
| CN112464818A (en) * | 2020-11-27 | 2021-03-09 | 北京软通智慧城市科技有限公司 | Kitchen supervision method, device, equipment and storage medium |
| CN112464818B (en) * | 2020-11-27 | 2024-04-16 | 北京软通智慧科技有限公司 | Kitchen supervision method, device, equipment and storage medium |
| CN113255542A (en) * | 2021-06-02 | 2021-08-13 | 北京容联易通信息技术有限公司 | Illegal object identity recognition method based on illegal behavior recognition |
| CN113408464A (en) * | 2021-06-30 | 2021-09-17 | 深圳市商汤科技有限公司 | Behavior detection method and device, electronic equipment and storage medium |
| CN113628370B (en) * | 2021-08-05 | 2022-06-24 | 国家核安保技术中心 | Intelligent protection channel control system for electronic equipment |
| CN113628370A (en) * | 2021-08-05 | 2021-11-09 | 国家核安保技术中心 | Intelligent protection channel control system for electronic equipment |
| CN114092889A (en) * | 2022-01-10 | 2022-02-25 | 深圳市明源云科技有限公司 | Violation detection method and device, electronic equipment and readable storage medium |
| CN114092889B (en) * | 2022-01-10 | 2022-04-15 | 深圳市明源云科技有限公司 | Violation detection method and device, electronic equipment and readable storage medium |
| WO2024027330A1 (en) * | 2022-08-04 | 2024-02-08 | 深圳市震有智联科技有限公司 | Real-time monitoring method based on user behavior, and related device |
| CN115484063A (en) * | 2022-08-12 | 2022-12-16 | 国家管网集团北方管道有限责任公司 | Network security prevention and control method and system for industrial control system |
| CN115484063B (en) * | 2022-08-12 | 2023-05-30 | 国家管网集团北方管道有限责任公司 | Network security prevention and control method and system for industrial control system |
| CN116092280A (en) * | 2023-02-07 | 2023-05-09 | 深圳市冠标科技发展有限公司 | Supervision method and device based on remote communication |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110718067A (en) | Violation behavior warning method and related device | |
| CN111507283B (en) | Student behavior identification method and system based on classroom scene | |
| KR102260120B1 (en) | Apparatus for Performing Recognition of Activity Based on Deep Learning and Driving Method Thereof | |
| Leyva et al. | The LV dataset: A realistic surveillance video dataset for abnormal event detection | |
| CN105405150B (en) | Anomaly detection method and device based on fusion feature | |
| WO2020000912A1 (en) | Behavior detection method and apparatus, and electronic device and storage medium | |
| Yin et al. | The infrared moving object detection and security detection related algorithms based on W4 and frame difference | |
| CN107229894A (en) | Intelligent video monitoring method and system based on computer vision analysis technology | |
| CN106165391A (en) | The image capturing strengthened | |
| CN107920223B (en) | Object behavior detection method and device | |
| CN113191699A (en) | Power distribution construction site safety supervision method | |
| CN110717358B (en) | Visitor number counting method and device, electronic equipment and storage medium | |
| CN111241913A (en) | Method, device and system for detecting falling of personnel | |
| CN111325954B (en) | Personnel loss early warning method, device, system and server | |
| Han et al. | Improved visual background extractor using an adaptive distance threshold | |
| CN103049748A (en) | Behavior-monitoring method and behavior-monitoring system | |
| CN113920585A (en) | Behavior recognition method and device, equipment and storage medium | |
| Vashistha et al. | An architecture to identify violence in video surveillance system using ViF and LBP | |
| CN111753587A (en) | Method and device for detecting falling to ground | |
| CN110533889B (en) | Sensitive area electronic equipment monitoring and positioning device and method | |
| CN111144260A (en) | Detection method, device and system of crossing gate | |
| KR101407394B1 (en) | System for abandoned and stolen object detection | |
| CN113392776B (en) | Seat leaving behavior detection method and storage device combining seat information and machine vision | |
| CN113554682B (en) | Target tracking-based safety helmet detection method | |
| CN114550049A (en) | Behavior recognition method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200121 |
|
| RJ01 | Rejection of invention patent application after publication |