CN114299448B - A method and system for identifying the behavior of people illegally bringing mobile phones into sensitive places - Google Patents
A method and system for identifying the behavior of people illegally bringing mobile phones into sensitive places Download PDFInfo
- Publication number
- CN114299448B CN114299448B CN202111616228.XA CN202111616228A CN114299448B CN 114299448 B CN114299448 B CN 114299448B CN 202111616228 A CN202111616228 A CN 202111616228A CN 114299448 B CN114299448 B CN 114299448B
- Authority
- CN
- China
- Prior art keywords
- mobile phone
- sensitive place
- image
- person
- personnel
- 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
Landscapes
- Mobile Radio Communication Systems (AREA)
- Alarm Systems (AREA)
- Telephonic Communication Services (AREA)
Abstract
The invention provides a behavior recognition method and a system for personnel to carry a mobile phone into a sensitive place in violation, and belongs to the field of safety monitoring. The method comprises the steps of preprocessing a monitoring video of a sensitive place to obtain an image to be detected, carrying out personnel positioning on the image to be detected and forming a single image set through segmentation, judging whether the image enters the sensitive place by utilizing a pedestrian track, if so, entering the next step, ending monitoring, judging whether a mobile phone enters the sensitive place according to a mobile phone detection result, if so, recording illegal behaviors, if not, entering the next step, judging whether the mobile phone is stored by utilizing the overlapping area of a hand area and a mobile phone storage area, if so, judging that illegal behaviors do not exist, if not, recording illegal behaviors, and repeating the steps until monitoring of all personnel is completed. The invention can ensure higher accuracy while simplifying the judging process, and effectively solves the problem that people cannot find out when carrying the mobile phone illegally into a sensitive place.
Description
Technical Field
The invention belongs to the field of safety monitoring, and particularly relates to a method and a system for identifying behaviors of personnel carrying mobile phones into sensitive places in a violation manner.
Background
Part of sensitive sites are used for guaranteeing information safety and equipment safety, and mobile phones are generally required to be forbidden to be used, so that supervision is required to be carried into the sensitive sites by personnel if the personnel carry the mobile phones illegally. At present, the behavior recognition method for people to carry mobile phones into sensitive places in violation mainly comprises on-site staring, intelligent mobile phone storage and remote viewing and video monitoring.
The method has the advantages that the method is high in identification precision and obvious in effect, but special personnel are required to be arranged at each entrance and exit for a long time to watch, the efficiency is extremely low, a large amount of manpower resources are occupied, the method is not suitable for large-area popularization, and meanwhile the defect that post verification cannot be performed exists. For intelligent mobile phone storage cabinets, CN111364875A discloses an intelligent secret cabinet and an intelligent management system, and CN211950099U discloses an intelligent secret cabinet with face recognition function, but the mode lacks strong constraint, and has the problem of poor supervision effect. In the remote video monitoring, a monitor can browse multiple paths of videos at the same time in a monitoring room, but in the face of massive videos, visual fatigue is easy to occur for the monitor, the high concentration of attention is difficult to keep for a long time, and the problem of unstable recognition accuracy exists. Although a mobile phone detection model exists in the prior art, whether a mobile phone appears in a monitoring picture can be detected to judge whether personnel are illegal to collect and substitute into a sensitive place, the technology has higher false detection rate and omission rate and cannot be practically applied.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide a behavior recognition method and a system for a person to carry a mobile phone into a sensitive place in violation, and aims to solve the problems of low efficiency, poor constraint force and poor supervision effect of the existing judgment method.
In order to achieve the above purpose, the invention provides a behavior recognition method for a person to carry a mobile phone illegally into a sensitive place, which comprises the following steps:
s1, preprocessing a monitoring video of a sensitive place to obtain an image to be detected;
s2, personnel positioning is carried out on the image to be detected obtained in the step S1, and segmentation is carried out on the image to be detected according to the personnel positioning result so as to form a single image set;
s3, carrying out target tracking, mobile phone detection and human body key point detection on a single person image set of the current person to obtain a pedestrian track, a mobile phone detection result and a human body posture;
S4, judging whether the current personnel enter a sensitive place or not by utilizing the pedestrian track, if so, entering a step S5, and if not, ending the monitoring of the current personnel;
s5, selecting mobile phone detection results of a sensitive place entrance and a sensitive place exit according to the pedestrian track to judge whether current personnel enter the sensitive place with mobile phones, if so, recording illegal behaviors, otherwise, entering a step S6;
S6, selecting the human body posture of the mobile phone storage cabinet according to the pedestrian track, judging whether to store the mobile phone or not by utilizing the overlapping area of the hand area and the mobile phone storage area, if yes, judging that no illegal behaviors exist for the current personnel, and if no, recording the illegal behaviors;
and S7, repeating the steps S3-S6 until monitoring of all personnel is completed.
As a further preferred embodiment, in step S1, the monitoring video is decoded and transcoded to convert the video stream into single-frame pictures, and a preset number of single-frame pictures are extracted by a frame extraction technique to obtain the image to be detected.
As a further preferable mode, in step S2, the neural network model is used to perform personnel location on the image to be detected, and the following matrix is output:
Where (x i1,yi1) is the upper left corner coordinate of the ith person on the image to be detected, (x i2,yi2) is the lower right corner coordinate of the ith person on the image to be detected, the ith person is completely located inside the rectangular bounding box formed by the two points, score i E [0,1] represents the confidence level of the ith person, and the score i is directly eliminated for the person less than the confidence level threshold.
More preferably, in step S2, the confidence threshold is 0.3 to 0.4.
As a further preferred aspect, in step S3, the single person image set is sent to a mobile phone detection model to detect whether a mobile phone exists in the image.
As a further preferred option, in step S3, the single person image set is sent to a human body key point detection model to obtain a human body posture.
As a further preferable aspect, in step S5, the mobile phone position coordinates and the confidence level are recorded while the violation is recorded.
As a further preferred, step S6 comprises the sub-steps of:
s61, determining wrist key points according to the human body posture of a mobile phone storage cabinet, and taking the wrist key points as the center, and making squares with the side length of 50-150 pixels as hand areas;
S62, calculating the overlapping area of the hand area and the mobile phone storage area;
S63, judging whether the overlapping area is larger than an area threshold, if so, judging that the mobile phone is stored by the current personnel, and if not, recording the illegal action.
Further preferably, in step S63, the area threshold is 200 pixels to 300 pixels.
According to the invention, the behavior recognition system for the people carrying the mobile phone illegally into the sensitive place is provided, and comprises a monitoring device, a video processing device and an alarm device, wherein the monitoring device is used for acquiring a monitoring video of the sensitive place, the video processing device is used for realizing the behavior recognition method for the people carrying the mobile phone illegally into the sensitive place, and the alarm device is used for alarming when the illegal behavior is recognized.
In general, the above technical solutions conceived by the present invention have the following beneficial effects compared with the prior art:
1. According to the invention, the monitoring video is processed by utilizing a computer vision technology, firstly, whether the current person enters a sensitive place is judged by utilizing a pedestrian track, so that the judging process is simplified, only the person entering the sensitive place is identified, then whether illegal actions exist or not is primarily judged by whether a mobile phone exists in images of an entrance of the sensitive place and an exit of the sensitive place, if the illegal actions do not exist, whether the action of storing the mobile phone is further judged by the overlapping area of a hand area at a mobile phone storage cabinet and a mobile phone storage area, so that whether the illegal actions exist or not can be more accurately judged, the judging process is simplified, higher accuracy can be ensured, the problem that the person is difficult to find and find when the mobile phone is illegally carried into the sensitive place is effectively solved, and manpower and material resources are effectively saved while the identifying accuracy is greatly improved;
2. Particularly, the invention utilizes the neural network model to perform personnel positioning on the image to be detected so as to obtain a single image set, and can improve the judging efficiency while improving the judging precision;
3. In addition, the invention defines the hand area by taking the wrist key point as the center when judging whether the mobile phone is stored, judges whether the action of storing the mobile phone exists or not through the overlapping area of the hand area and the storage area, simplifies the judging process and can ensure higher accuracy.
Drawings
Fig. 1 is a flow chart of a behavior recognition method for a person carrying a mobile phone into a sensitive place in violation, which is provided by the embodiment of the invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
As shown in fig. 1, the invention provides a behavior recognition method for a person to carry a mobile phone into a sensitive place in violation, which comprises the following steps:
S1, preprocessing a monitoring video of a sensitive place, wherein the monitoring video can cover an inlet of the sensitive place, an outlet of the sensitive place and a mobile phone storage cabinet, can be obtained in real time or can be cached historical monitoring data, converts a video stream into single-frame pictures through decoding and transcoding in the preprocessing process, and extracts one picture every certain frame through a frame extraction technology, so that a preset number of single-frame pictures are obtained and are used as images to be detected;
s2, personnel positioning is carried out on the image to be detected obtained in the step S1, and segmentation is carried out on the image to be detected according to the personnel positioning result so as to form a single image set;
s3, carrying out target tracking, mobile phone detection and human body key point detection on a single person image set of the current person to obtain a pedestrian track, a mobile phone detection result and a human body posture;
s4, judging whether the current personnel enter a sensitive place or not by utilizing the track of the pedestrians, if so, entering a step S5, and if not, ending the monitoring of the current personnel;
S5, selecting mobile phone detection results of a sensitive place entrance and a sensitive place exit according to the track of the pedestrian to judge whether current personnel enter the sensitive place with mobile phones, if so, recording illegal behaviors, recording time, camera ID, illegal places, current video frame numbers, illegal personnel, mobile phone position coordinates and confidence, and if not, entering step S6;
S6, selecting the human body posture of the mobile phone storage cabinet according to the track of the pedestrian, judging whether to store the mobile phone or not by utilizing the overlapping area of the hand area and the mobile phone storage area, if yes, judging that no illegal action exists in the current personnel, if no, recording the illegal action, and recording time, camera ID, illegal places, current video frame number and illegal personnel;
and S7, repeating the steps S3-S6 until monitoring of all personnel is completed.
In step S2, a neural network model is used to locate the person in the image to be detected, and a matrix represented by the following formula is output:
Wherein each row represents one person in the image to be detected, (x i1,yi1) is the left upper corner coordinate of the ith person on the image to be detected, (x i2,yi2) is the right lower corner coordinate of the ith person on the image to be detected, the ith person is completely positioned inside a rectangular boundary frame formed by the two points, score i epsilon [0,1] represents the confidence of the ith person, the person with score i smaller than a confidence threshold is directly removed, the confidence threshold is preferably 0.3-0.4, so that the calculated amount can be reduced to improve the processing speed, and the image to be detected is segmented according to the result of personnel positioning to form a single person image set.
In step S3, a neural network model is utilized to track the target, a single image set is sent to a mobile phone detection model to detect whether a mobile phone exists in the image, wherein the mobile phone detection model can be a target detection model in the prior art, whether a specific target of the mobile phone exists in the detection image can be accurately distinguished through training, and meanwhile, the single image set is sent to a human body key point detection model to obtain the human body gesture.
Further, step S6 includes the following sub-steps:
s61, determining wrist key points according to the human body posture of the mobile phone storage cabinet, taking the wrist key points as the center, making squares with the side length of 50-150 pixels, and taking the squares as a hand area, wherein other key points in the human body posture can be taken as auxiliary basis for judging the position relationship between the human body and the mobile phone storage cabinet, so that the hand positioning accuracy is further improved;
S62, calculating the overlapping area of the hand area and the mobile phone storage area;
And S63, judging whether the overlapping area is larger than an area threshold, if so, indicating that the current personnel interact with the mobile phone storage area, judging that the current personnel does not have illegal behaviors, and if not, indicating that the current personnel does not store mobile phones, and recording the illegal behaviors.
Further, in step S63, the area threshold is 200 pixels to 300 pixels, so that the accuracy of the judgment can be ensured.
According to the invention, the system comprises a monitoring device, a video processing device and an alarm device, wherein the monitoring device is used for acquiring a monitoring video of a sensitive place and can cover an entrance of the sensitive place, an exit of the sensitive place and a mobile phone storage cabinet, the mobile phone storage cabinet is used for storing mobile phones before entering the sensitive place, the existing common mobile phone storage cabinet is utilized, the system has the advantage of low cost and easy arrangement, the video processing device is used for realizing the method for identifying the behavior of the personnel carrying the mobile phones illegally into the sensitive place, target tracking, mobile phone detection and human body key point detection can be carried out, and the alarm device is used for alarming when the illegal behaviors are identified and can push information to responsible persons.
It will be readily appreciated by those skilled in the art that the foregoing description is merely a preferred embodiment of the invention and is not intended to limit the invention, but any modifications, equivalents, improvements or alternatives falling within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (9)
1. A behavior recognition method for personnel to carry a mobile phone into a sensitive place in violation is characterized by comprising the following steps:
s1, preprocessing a monitoring video of a sensitive place to obtain an image to be detected;
s2, personnel positioning is carried out on the image to be detected obtained in the step S1, and segmentation is carried out on the image to be detected according to the personnel positioning result so as to form a single image set;
s3, carrying out target tracking, mobile phone detection and human body key point detection on a single person image set of the current person to obtain a pedestrian track, a mobile phone detection result and a human body posture;
S4, judging whether the current personnel enter a sensitive place or not by utilizing the pedestrian track, if so, entering a step S5, and if not, ending the monitoring of the current personnel;
s5, selecting mobile phone detection results of a sensitive place entrance and a sensitive place exit according to the pedestrian track to judge whether current personnel enter the sensitive place with mobile phones, if so, recording illegal behaviors, otherwise, entering a step S6;
S6, selecting the human body posture of the mobile phone storage cabinet according to the pedestrian track, judging whether to store the mobile phone by utilizing the overlapping area of the hand area and the mobile phone storage area, if yes, judging that no illegal behaviors exist for the current personnel, and if no, recording the illegal behaviors, wherein the specific steps are as follows:
s61, determining wrist key points according to the human body posture of a mobile phone storage cabinet, and taking the wrist key points as the center, and making squares with the side length of 50-150 pixels as hand areas;
S62, calculating the overlapping area of the hand area and the mobile phone storage area;
s63, judging whether the overlapping area is larger than an area threshold, if so, judging that the mobile phone is stored by the current personnel, and if not, recording the illegal action;
and S7, repeating the steps S3-S6 until monitoring of all personnel is completed.
2. The method for identifying behavior of a person carrying a mobile phone illegally into a sensitive place according to claim 1, wherein in step S1, a surveillance video is decoded and transcoded to convert a video stream into single-frame pictures, and a preset number of single-frame pictures are extracted by a frame extraction technique to obtain an image to be detected.
3. The method for identifying behavior of a person carrying a mobile phone into a sensitive place according to claim 1, wherein in step S2, the person is positioned on an image to be detected by using a neural network model, and the following matrix is output:
In the formula, Is the firstThe upper left corner of the person on the image to be detected,Is the firstThe lower right angular position of the person on the image to be detected, the firstThe individual is located entirely inside the rectangular bounding box formed by these two points,Represent the firstConfidence of individual person forPeople less than the confidence threshold are rejected directly.
4. The method for identifying behavior of a person carrying a mobile phone into a sensitive place according to claim 3, wherein in step S2, the confidence threshold is 0.3-0.4.
5. The method for identifying behavior of a person carrying a mobile phone illegally into a sensitive place as claimed in claim 1, wherein in step S3, a single person image set is sent to a mobile phone detection model to detect whether a mobile phone exists in the image.
6. The method for identifying behavior of a person carrying a mobile phone into a sensitive place according to claim 1, wherein in step S3, a single person image set is sent to a human body key point detection model to obtain a human body posture.
7. The method for identifying behavior of a person carrying a mobile phone into a sensitive place according to claim 1, wherein in step S5, the position coordinates and the confidence level of the mobile phone are recorded while the behavior of the person carrying the mobile phone is recorded.
8. The method for identifying behavior of a person carrying a mobile phone into a sensitive place according to claim 1, wherein in step S63, the area threshold is 200 pixels to 300 pixels.
9. The system is characterized by comprising a monitoring device, a video processing device and an alarm device, wherein the monitoring device is used for acquiring a monitoring video of a sensitive place, the video processing device is used for realizing the method for identifying the behavior of the personnel carrying the mobile phone into the sensitive place, and the alarm device is used for alarming when the illegal behavior is identified.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111616228.XA CN114299448B (en) | 2021-12-27 | 2021-12-27 | A method and system for identifying the behavior of people illegally bringing mobile phones into sensitive places |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111616228.XA CN114299448B (en) | 2021-12-27 | 2021-12-27 | A method and system for identifying the behavior of people illegally bringing mobile phones into sensitive places |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114299448A CN114299448A (en) | 2022-04-08 |
| CN114299448B true CN114299448B (en) | 2024-12-13 |
Family
ID=80970225
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111616228.XA Active CN114299448B (en) | 2021-12-27 | 2021-12-27 | A method and system for identifying the behavior of people illegally bringing mobile phones into sensitive places |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114299448B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114913452A (en) * | 2022-04-20 | 2022-08-16 | 哈尔滨工业大学（深圳） | Office place-based violation detection system and method |
| CN115393563A (en) * | 2022-09-15 | 2022-11-25 | 杭州萤石软件有限公司 | Package detection method and system and electronic equipment |
| CN116563745A (en) * | 2023-04-10 | 2023-08-08 | 长沙海信智能系统研究院有限公司 | Method, device and electronic equipment for identifying personnel holding a knife |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20130138973A (en) * | 2012-06-12 | 2013-12-20 | 오유미 | Method for displaying motion drawing based on smart-phone, and smart-phone with motion drawing display function |
| CN110991261A (en) * | 2019-11-12 | 2020-04-10 | 苏宁云计算有限公司 | Interactive behavior recognition method and device, computer equipment and storage medium |
-
2021
- 2021-12-27 CN CN202111616228.XA patent/CN114299448B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20130138973A (en) * | 2012-06-12 | 2013-12-20 | 오유미 | Method for displaying motion drawing based on smart-phone, and smart-phone with motion drawing display function |
| CN110991261A (en) * | 2019-11-12 | 2020-04-10 | 苏宁云计算有限公司 | Interactive behavior recognition method and device, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114299448A (en) | 2022-04-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113269091A (en) | Personnel trajectory analysis method, equipment and medium for intelligent park | |
| CN113903081B (en) | Hydropower plant image visual identification artificial intelligence alarm method and device | |
| CN114299448B (en) | A method and system for identifying the behavior of people illegally bringing mobile phones into sensitive places | |
| CN114926781B (en) | Multi-person time-space domain abnormal behavior positioning method and system supporting real-time monitoring scene | |
| CN111091098A (en) | Training method and detection method of detection model and related device | |
| CN109658554A (en) | A kind of intelligent residential district security system based on big data | |
| CN111985331B (en) | Detection method and device for preventing trade secret from being stolen | |
| CN112183219A (en) | Public safety video monitoring method and system based on face recognition | |
| RU2713876C1 (en) | Method and system for detecting alarm events when interacting with self-service device | |
| CN118942159A (en) | A park visitor safety warning method, system and medium based on digital twin | |
| CN110246292A (en) | Domestic video monitoring method, apparatus and storage medium | |
| CN116189052A (en) | Security method, system, intelligent terminal and storage medium based on video stream analysis | |
| CN117998039A (en) | Video data processing method, device, equipment and storage medium | |
| CN113743380B (en) | Active tracking method based on video image dynamic monitoring | |
| CN118821019A (en) | Smart community personnel track recognition method and system | |
| CN115205581A (en) | Fishing detection method, fishing detection device and computer readable storage medium | |
| CN118470645A (en) | Visual detection-based intelligent pen test monitoring system and method | |
| CN107516076A (en) | Portrait identification method and device | |
| CN109215150A (en) | Face is called the roll and method of counting and its system | |
| CN110689694A (en) | Intelligent monitoring system and method based on image processing | |
| CN114979567B (en) | Object and region interaction method and system applied to video intelligent monitoring | |
| CN113468985B (en) | A locking method for persons carrying suspicious radiation sources | |
| Dumoulin et al. | UNICITY: A depth maps database for people detection in security airlocks | |
| CN113158933A (en) | Method, system, device and storage medium for identifying lost personnel | |
| CN116580514A (en) | Intelligent security method, system, medium and electronic equipment based on Internet of things |
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 |