CN110572618B - Illegal photographing behavior monitoring method, device and system - Google Patents
Illegal photographing behavior monitoring method, device and system Download PDFInfo
- Publication number
- CN110572618B CN110572618B CN201910907087.3A CN201910907087A CN110572618B CN 110572618 B CN110572618 B CN 110572618B CN 201910907087 A CN201910907087 A CN 201910907087A CN 110572618 B CN110572618 B CN 110572618B
- Authority
- CN
- China
- Prior art keywords
- photographing behavior
- mobile phone
- behavior
- photographing
- position information
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
Abstract
The invention discloses a method, a device and a system for monitoring illegal photographing behaviors, and belongs to the technical field of video monitoring. The illegal photographing behavior monitoring method is characterized in that the photographing behavior is judged by adopting the characteristic value of a monitoring target, and an alarm is given to the illegal photographing behavior. An illegal photographing behavior monitoring device identifies a photographing behavior and sends an alarm in time for the illegal photographing behavior. The invention mainly utilizes the characteristic value of the monitored target to carry out identification and judgment, thereby having little influence by the environment, improving the practicability of the technology, reducing the development cost and difficulty and reducing the misjudgment rate.
Description
Technical Field
The invention relates to the technical field of video monitoring, in particular to a method, a device and a system for monitoring illegal photographing behaviors.
Background
Existing video-based motion recognition algorithms generally perform motion determination based on consecutive frames. One of the common methods is to extract an overall feature value from each frame of picture by using a Convolutional Neural Network (CNN), and then classify a feature value sequence by using a Recurrent Neural Network (RNN) algorithm. This method is easily affected by the scene, and the background picture and the shooting angle have a great influence on the determination result. Reducing this effect requires a large number of training samples to increase diversity, which is costly for practical engineering implementations.
The existing motion recognition technology of the Convolutional Neural Network (CNN) + long and short term memory network (LSTM) adopts a method for extracting features of the whole picture by the CNN, and the method is applied to the scene of illegal photographing monitoring. The only method for reducing the interference of the environmental factors is to improve the diversity of the samples and shoot a large number of sample videos in different environments for training. But not only does this increase the cost of preparing the sample, due to the diversity and unpredictability of the actual work scenario, but the results may still not meet the diversity requirements in the actual work.
The action of shooting by lifting the mobile phone can have a plurality of different postures and different shooting angles. The shooting action is very similar to other daily actions, and is particularly easy to be confused in a two-dimensional picture. Simply using the CNN network to extract the features of the picture cannot accurately capture the difference between the two features, which may cause a large amount of misjudgments. Although the general deep learning algorithm can reduce the false positive rate by increasing the amount of training samples, the acquisition cost of the samples is high, and the result may still not meet the requirement of practical work.
Disclosure of Invention
The invention mainly solves the technical problem of providing a method, a device and a system for monitoring illegal photographing behaviors, which can reduce the interference of environmental factors on action recognition, reduce the misjudgment rate and reduce the amount of training samples.
In order to achieve the above object, the first technical solution adopted by the present invention is: a method for monitoring illegal photographing behavior, wherein the photographing behavior is completed by a person using a mobile phone to photograph a target, the method is characterized by comprising the following steps:
judging a photographing behavior by adopting the characteristic value of the person, the characteristic value of the mobile phone and the characteristic value of the target, and alarming the photographing behavior judged to be illegal, wherein the characteristic values are position information and image information;
forming a network video stream by a camera video picture, and reading the video frames from the video stream according to 5 second intervals;
the YOLO frame model receives the video frames, detects the person, the mobile phone and the target in each video frame picture, extracts the position information of the person, the position information of the mobile phone and the position information of the target and combines the position information into a position information sequence according to the time sequence; extracting the image information of the person, the image information of the mobile phone and the image information of the target and combining the image information into an image information sequence according to a time sequence;
calculating the moving distance of the mobile phone relative to the video frame picture based on the position information sequence of the mobile phone, if the moving distance of the mobile phone in the horizontal direction is 5-20% of the width of the video frame picture or the moving distance in the vertical direction is 5-20% of the height of the video frame picture, judging that the photographing behavior is an abnormal photographing behavior, inputting the position information sequence corresponding to the abnormal photographing behavior into a recurrent neural network model, otherwise, judging that the photographing behavior is a normal photographing behavior, and ending the processing;
the recurrent neural network model analyzes the position information sequence corresponding to the abnormal photographing behavior and outputs a result, if the result is abnormal, the image information corresponding to the abnormal photographing behavior is input into a classifier, otherwise, the photographing behavior is judged to be a normal photographing behavior, and the processing is finished;
the classifier analyzes the input image information and outputs a result, if the result is abnormal, the illegal photographing behavior of the photographing behavior is judged, and an alarm signal is sent out; otherwise, judging the photographing behavior to be a normal photographing behavior, and ending the processing.
The second technical scheme adopted by the invention is as follows: an illegal act of taking a picture monitoring device, the act of taking a picture by aiming at a target with a mobile phone is accomplished, comprising:
the judging module judges the photographing behavior by adopting the characteristic value of the person, the characteristic value of the mobile phone and the characteristic value of the target, and alarms the photographing behavior judged to be illegal, wherein the characteristic values are position information and image information; the video frame module adopts a monitoring camera video picture to form a network video stream, and reads the video frames from the video stream according to 5-second intervals;
a YOLO frame model module which adopts a YOLO frame model to receive the video frames, detects the people, the mobile phone and the targets in each video frame picture, extracts the position information of the people, the position information of the mobile phone and the position information of the targets and combines the position information into a position information sequence according to the time sequence; extracting the image information of the person, the image information of the mobile phone and the image information of the target and combining the image information into an image information sequence according to a time sequence;
a mobile phone position judging module, which calculates the moving distance of the mobile phone relative to the video frame picture based on the position information sequence of the mobile phone, if the moving distance of the mobile phone in the horizontal direction is 5-20% of the width of the video frame picture or the moving distance in the vertical direction is 5-20% of the height of the video frame picture, then judges that the photographing behavior is an abnormal photographing behavior, and inputs the position information sequence corresponding to the abnormal photographing behavior into a recurrent neural network model, otherwise, judges that the photographing behavior is a normal photographing behavior, and ends the processing;
the cyclic neural network model module is used for analyzing the position information sequence corresponding to the abnormal photographing behavior and outputting a result, if the result is abnormal, the image information corresponding to the abnormal photographing behavior is input into the classifier module, otherwise, the photographing behavior is judged to be a normal photographing behavior, and the processing is finished;
the classifier module analyzes the input image information and outputs a result, and if the result is abnormal, the illegal photographing behavior of the photographing behavior is judged and an alarm signal is sent out; otherwise, judging the photographing behavior to be a normal photographing behavior, and ending the processing.
The third technical scheme adopted by the invention is as follows: an illegal photographing monitoring system is characterized by comprising an illegal photographing behavior monitoring device in the technical scheme II.
The fourth technical scheme adopted by the invention is as follows: a computer-readable storage medium storing computer instructions, wherein the computer instructions are operated to execute the illegal photographing behavior monitoring method according to the first technical aspect.
The fifth technical scheme adopted by the invention is as follows: a computer device comprising a processor and a memory, the memory storing computer instructions, wherein:
the processor operates the computer instructions to execute the illegal photographing behavior monitoring method of the first technical scheme.
The invention has the beneficial effects that: the interference of environmental factors on motion recognition is reduced, the misjudgment rate is reduced, the training sample amount is reduced, and the sample preparation cost is reduced.
Drawings
FIG. 1 is a schematic diagram of a framework of an illegal photographing behavior monitoring method according to the present invention;
FIG. 2 is a schematic diagram of an illegal photographing behavior monitoring method according to the present invention;
fig. 3 is a schematic structural diagram of an illegal photographing behavior monitoring apparatus according to the present invention.
Detailed Description
The following detailed description of the preferred embodiments of the present invention, taken in conjunction with the accompanying drawings, will make the advantages and features of the invention easier to understand by those skilled in the art, and thus will clearly and clearly define the scope of the invention.
With the development of smart phone technology, people often use a mobile phone to shoot a selected target, and besides normal shooting, some shooting behaviors inconsistent with the environment also occur.
The invention mainly identifies the behaviors shot by the mobile phone in the indoor office environment based on the combination of the monitoring camera and the deep learning technology, and gives an alarm for some illegal shooting behaviors.
Fig. 1 is a schematic diagram of a framework of an illegal photographing behavior monitoring method, which mainly includes five parts, namely video signal acquisition, target detection and feature extraction, a cyclic neural network model, a convolutional neural network and classification.
The video signal acquisition mainly aims at image pictures shot by a monitoring camera in an office. A YOLO (you Only Look one) frame is adopted for target detection and feature extraction, a large amount of video picture information and data information are collected firstly, the YOLO frame is input, a YOLO frame model is formed through training, and then the method is used for the method. The cyclic neural network model is also used for acquiring a large amount of abnormal data and normal data, inputting the abnormal data and the normal data into the cyclic neural network, training the abnormal data and the normal data to form the cyclic neural network model and then applying the cyclic neural network model to the invention. The classifier in the invention is a convolutional neural network model, and firstly collects a large number of pictures shot by a monitoring camera, wherein the pictures comprise people, mobile phones and shot targets (such as computer screens). The pictures are cut into the most suitable pictures (mainly comprising people, mobile phones and targets) through a YOLO frame, a large number of cut pictures are divided into two types of normal pictures and abnormal pictures, then the normal pictures and the abnormal pictures are input into a convolutional neural network, a classifier is formed through training, and then the method is applied to the invention.
The following describes in detail the process of the illegal photographing behavior monitoring method according to the present invention with reference to fig. 2 as an example:
collecting video pictures shot by a monitoring camera in an office environment, transmitting the video pictures through a network according to a time sequence to form a network video stream, and reading video frames from the video stream according to 5-second intervals. Through the statistics of the photographing action completion time, a normal continuous photographing action time takes about 5 seconds, and thus 5 seconds is adopted for the present invention.
Inputting the video frames into a YOLO (you Only Look one) frame model, detecting people, mobile phones and targets in each video frame picture, extracting position information of the people, the mobile phones and the targets, and combining the position information of the people, the mobile phones and the targets into a position information sequence according to a time sequence; and extracting the image information of the person, the image information of the mobile phone and the image information of the target and combining the image information into an image information sequence according to the time sequence. The purpose of this step is to find people, cell phones and targets in the picture.
And calculating the moving distance of the mobile phone relative to the video frame picture according to the position information sequence of the mobile phone, if the moving distance of the mobile phone in the horizontal direction is 5-20% of the width of the video frame picture or the moving distance in the vertical direction is 5-20% of the height of the video frame picture, judging that the photographing behavior is an abnormal photographing behavior, inputting the corresponding position information sequence into a recurrent neural network model, otherwise, judging that the photographing behavior is a normal photographing behavior, and ending the processing. According to the statistical research of technicians, the range of 5-20% is manually set, and when the range is less than 5%, the mobile phone does not move or the moving distance is very small in most cases, so that the photographing behavior cannot exist; and when the content exceeds 20 percent, the objects in the hands of the people are not true mobile phones, and can be mice or tablet computers, and the situation that the people hold the mobile phones to move fast (non-photographing action) is also caused.
For abnormal photographing behaviors with suspected photographing, analyzing the corresponding position information sequence by adopting a recurrent neural network model and outputting a result, if the result is abnormal, inputting image information corresponding to the suspected abnormal photographing behavior into a classifier, otherwise, judging the photographing behavior to be normal, and ending the processing;
the classifier analyzes the input image information and outputs a result, if the result is abnormal, the illegal photographing behavior of the photographing behavior is judged, and an alarm signal is sent out; otherwise, judging the photographing behavior to be a normal photographing behavior, and ending the processing.
The illegal photographing behavior monitoring device comprises six parts, as shown in fig. 3, which are respectively a judgment module, a video frame module, a YOLO frame model module, a mobile phone position judgment module, a recurrent neural network model module and a classifier module. The operation is carried out according to the flow of the illegal photographing behavior monitoring method.
The illegal photographing behavior monitoring device can also form an illegal photographing behavior monitoring system with other devices (such as a computer terminal, a mobile phone, a tablet personal computer and the like), so that personnel obtaining checking and monitoring authorization can master the occurrence condition of the photographing behavior in time.
The various illustrative logics, logical blocks, and modules described in connection with the illegal photo-activity monitoring method of the present invention may be implemented or performed with 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, or any combination thereof designed to perform the illegal photo-activity monitoring method herein. A general purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.
The illegal photographing behavior monitoring method can be directly implemented in hardware, in a software module executed by a processor or in a combination of the two. A software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an Application Specific Integrated Circuit (ASIC). The ASIC may reside in a user terminal. In the alternative, the processor and the storage medium may reside as discrete components in a user terminal.
The application of the invention only identifies and analyzes people, mobile phones and photographed targets in the monitored image picture, removes the interference of other environmental factors in the picture on action identification, greatly improves the adaptability of the monitoring method in different scenes, reduces the requirement on sample acquisition, reduces the amount of training samples and reduces the sample preparation cost. The invention reduces the misjudgment rate through three judgment processes and improves the effective rate of the monitoring method.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all equivalent structural changes made by using the contents of the present specification and the drawings, or applied directly or indirectly to other related technical fields, are included in the scope of the present invention.
Claims (5)
1. A method for monitoring illegal photographing behavior, wherein the photographing behavior is completed by a person using a mobile phone to photograph a target, the method is characterized by comprising the following steps:
judging a photographing behavior by adopting the characteristic value of the person, the characteristic value of the mobile phone and the characteristic value of the target, and alarming the photographing behavior judged to be illegal, wherein the characteristic values are position information and image information;
forming a network video stream by a camera video picture, and reading video frames from the video stream according to 5 second intervals;
the YOLO frame model receives the video frames, detects the person, the mobile phone and the target in each video frame picture, extracts the position information of the person, the position information of the mobile phone and the position information of the target and combines the position information into a position information sequence according to the time sequence; extracting the image information of the person, the image information of the mobile phone and the image information of the target and combining the image information into an image information sequence according to a time sequence;
calculating the moving distance of the mobile phone relative to the video frame picture based on the position information sequence of the mobile phone, if the moving distance of the mobile phone in the horizontal direction is 5-20% of the width of the video frame picture or the moving distance in the vertical direction is 5-20% of the height of the video frame picture, judging that the photographing behavior is an abnormal photographing behavior, inputting the position information sequence corresponding to the abnormal photographing behavior into a recurrent neural network model, otherwise, judging that the photographing behavior is a normal photographing behavior, and ending the processing;
the recurrent neural network model analyzes the position information sequence corresponding to the abnormal photographing behavior and outputs a result, if the result is abnormal, the image information corresponding to the abnormal photographing behavior is input into a classifier, otherwise, the photographing behavior is judged to be a normal photographing behavior, and the processing is finished;
the classifier analyzes the input image information and outputs a result, if the result is abnormal, the illegal photographing behavior of the photographing behavior is judged, and an alarm signal is sent out; otherwise, judging the photographing behavior to be a normal photographing behavior, and ending the processing.
2. An illegal act of taking a picture monitoring device, the act of taking a picture by aiming at a target with a mobile phone is accomplished, comprising:
the judging module judges the photographing behavior by adopting the characteristic value of the person, the characteristic value of the mobile phone and the characteristic value of the target, and alarms the photographing behavior judged to be illegal, wherein the characteristic values are position information and image information;
the video frame module adopts a monitoring camera video picture to form a network video stream and reads video frames from the video stream according to 5-second intervals;
a YOLO frame model module which adopts a YOLO frame model to receive the video frames, detects the people, the mobile phone and the targets in each video frame picture, extracts the position information of the people, the position information of the mobile phone and the position information of the targets and combines the position information into a position information sequence according to the time sequence; extracting the image information of the person, the image information of the mobile phone and the image information of the target and combining the image information into an image information sequence according to a time sequence;
a mobile phone position judging module, which calculates the moving distance of the mobile phone relative to the video frame picture based on the position information sequence of the mobile phone, if the moving distance of the mobile phone in the horizontal direction is 5-20% of the width of the video frame picture or the moving distance in the vertical direction is 5-20% of the height of the video frame picture, then judges that the photographing behavior is an abnormal photographing behavior, and inputs the position information sequence corresponding to the abnormal photographing behavior into a recurrent neural network model, otherwise, judges that the photographing behavior is a normal photographing behavior, and ends the processing;
the cyclic neural network model module is used for analyzing the position information sequence corresponding to the abnormal photographing behavior and outputting a result, if the result is abnormal, the image information corresponding to the abnormal photographing behavior is input into the classifier module, otherwise, the photographing behavior is judged to be a normal photographing behavior, and the processing is finished;
the classifier module analyzes the input image information and outputs a result, and if the result is abnormal, the illegal photographing behavior of the photographing behavior is judged and an alarm signal is sent out; otherwise, judging the photographing behavior to be a normal photographing behavior, and ending the processing.
3. An illegal photographing monitoring system comprising the illegal photographing behavior monitoring apparatus of claim 2.
4. A computer readable storage medium storing computer instructions, wherein the computer instructions are operable to perform the illegal photo act monitoring method of claim 1.
5. A computer device comprising a processor and a memory, the memory storing computer instructions, wherein: the processor operates the computer instructions to perform the illegal photo act monitoring method of claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910907087.3A CN110572618B (en) | 2019-09-24 | 2019-09-24 | Illegal photographing behavior monitoring method, device and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910907087.3A CN110572618B (en) | 2019-09-24 | 2019-09-24 | Illegal photographing behavior monitoring method, device and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110572618A CN110572618A (en) | 2019-12-13 |
| CN110572618B true CN110572618B (en) | 2021-03-16 |
Family
ID=68782101
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910907087.3A Active CN110572618B (en) | 2019-09-24 | 2019-09-24 | Illegal photographing behavior monitoring method, device and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110572618B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111225187A (en) * | 2020-01-16 | 2020-06-02 | 厦门博海中天信息科技有限公司 | Intelligent monitoring method and device for recognizing photo taking prevention of mobile phone and readable medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106998422A (en) * | 2016-01-26 | 2017-08-01 | 陈征 | A kind of automatic identification is taken on the sly the apparatus and method of camera lens |
| CN108234741A (en) * | 2017-11-16 | 2018-06-29 | 吴英 | Intelligent mobile phone control method based on action recognition |
| CN109271814A (en) * | 2018-10-09 | 2019-01-25 | 裕利年电子南通有限公司 | Anti- take pictures display system and anti-photographic method based on computer vision |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006267687A (en) * | 2005-03-24 | 2006-10-05 | Fujitsu Ltd | Information processing device, unauthorized photography preventing method and unauthorized photography preventing program |
| US20100321493A1 (en) * | 2008-03-07 | 2010-12-23 | Thomson Licensing | Apparatus and method for remote monitoring |
| US20130057693A1 (en) * | 2011-09-02 | 2013-03-07 | John Baranek | Intruder imaging and identification system |
| JP6056323B2 (en) * | 2012-09-24 | 2017-01-11 | 富士通株式会社 | Gaze detection device, computer program for gaze detection |
| CN104969239A (en) * | 2012-12-04 | 2015-10-07 | 光壳科技股份有限公司 | Device and methods for detecting a camera |
| CN107257449B (en) * | 2016-06-27 | 2018-02-09 | 杨春 | A kind of remote supervision system based on cloud computing |
| CN107784653A (en) * | 2017-11-06 | 2018-03-09 | 山东浪潮云服务信息科技有限公司 | A kind of Anti-sneak-shooting system and method |
| CN110096972A (en) * | 2019-04-12 | 2019-08-06 | 重庆科芮智能科技有限公司 | Data guard method, apparatus and system |
| CN110163100B (en) * | 2019-04-17 | 2022-04-01 | 中国电子科技网络信息安全有限公司 | Anti-photographing display |
-
2019
- 2019-09-24 CN CN201910907087.3A patent/CN110572618B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106998422A (en) * | 2016-01-26 | 2017-08-01 | 陈征 | A kind of automatic identification is taken on the sly the apparatus and method of camera lens |
| CN108234741A (en) * | 2017-11-16 | 2018-06-29 | 吴英 | Intelligent mobile phone control method based on action recognition |
| CN109271814A (en) * | 2018-10-09 | 2019-01-25 | 裕利年电子南通有限公司 | Anti- take pictures display system and anti-photographic method based on computer vision |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110572618A (en) | 2019-12-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10534957B2 (en) | Eyeball movement analysis method and device, and storage medium | |
| Shao et al. | Deep convolutional dynamic texture learning with adaptive channel-discriminability for 3D mask face anti-spoofing | |
| CN110609920B (en) | Pedestrian hybrid search method and system in video monitoring scene | |
| WO2020094091A1 (en) | Image capturing method, monitoring camera, and monitoring system | |
| WO2020083111A1 (en) | Liveness detection method and device, electronic apparatus, storage medium and related system using the liveness detection method | |
| WO2020094088A1 (en) | Image capturing method, monitoring camera, and monitoring system | |
| Avgerinakis et al. | Recognition of activities of daily living for smart home environments | |
| CN109492550B (en) | Living body detection method, living body detection device and related system applying living body detection method | |
| CN111553231B (en) | Face snapshot and deduplication system, method, terminal and medium based on information fusion | |
| CN110782433A (en) | Dynamic information violent parabolic detection method and device based on time sequence and storage medium | |
| Raya et al. | Analysis realization of Viola-Jones method for face detection on CCTV camera based on embedded system | |
| CN207233038U (en) | Face is called the roll and number system | |
| CN114885119A (en) | Intelligent monitoring alarm system and method based on computer vision | |
| CN110572618B (en) | Illegal photographing behavior monitoring method, device and system | |
| CN114764895A (en) | Abnormal behavior detection device and method | |
| CN210072642U (en) | Crowd abnormal behavior detection system based on video monitoring | |
| Varghese et al. | Video anomaly detection in confined areas | |
| Jahlan et al. | Detecting Violence in Video Based on Deep Features Fusion Technique | |
| Chen et al. | Head-shoulder detection using joint HOG features for people counting and video surveillance in library | |
| CN111144260A (en) | Detection method, device and system of crossing gate | |
| Mustafah et al. | Face detection system design for real time high resolution smart camera | |
| CN111091089B (en) | Face image processing method and device, electronic equipment and storage medium | |
| CN111814613A (en) | Face recognition method, face recognition equipment and computer readable storage medium | |
| Wu et al. | An improved CornerNet-lite method for pedestrian detection of unmanned aerial vehicle images | |
| Wu et al. | Intelligent monitoring system based on Hi3531 for recognition of human falling action |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |