CN109376639B - Accompanying personnel early warning system and method based on portrait recognition - Google Patents
Accompanying personnel early warning system and method based on portrait recognition Download PDFInfo
- Publication number
- CN109376639B CN109376639B CN201811199697.4A CN201811199697A CN109376639B CN 109376639 B CN109376639 B CN 109376639B CN 201811199697 A CN201811199697 A CN 201811199697A CN 109376639 B CN109376639 B CN 109376639B
- Authority
- CN
- China
- Prior art keywords
- accompanying
- portrait
- person
- target person
- target
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/53—Recognition of crowd images, e.g. recognition of crowd congestion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
- G08B21/24—Reminder alarms, e.g. anti-loss alarms
Abstract
The invention provides a system and a method for early warning accompanying personnel based on portrait recognition, which belong to the technical field of portrait recognition application, and aim to solve the problem that accompanying personnel analysis generally adopts a posterior analysis method and cannot prevent the accompanying personnel in advance. The invention comprises the following steps: the portrait recognition module is used for detecting the portrait of each point location in real time; an accompanying rule setting module for setting an accompanying rule and/or inputting a user-defined target person; the logic judgment module is used for judging whether the detected portrait is a target person, recording the space-time trajectory of the target person when the detected portrait is determined to be the target person, and caching a portrait feature set of the target person in an effective accompanying time range of each space-time point; when the space-time trajectory of the target person reaches the accompanying detection condition, acquiring the accompanying person of the target person according to the set and the accompanying rule; the early warning output module is used for outputting the early warning of accompanying personnel; the invention can remind related management personnel to intervene before a case, thereby achieving the prevention effect.
Description
Technical Field
The invention relates to an early warning system and an early warning method, in particular to a system and a method for early warning accompanying personnel based on portrait recognition, and belongs to the technical field of portrait recognition application.
Background
In the field of face recognition, automatic feature point extraction and comparison are carried out on snap pictures of a field camera and internal personnel pictures registered in a system, two pictures with similarity exceeding a threshold value are regarded as the same person through setting the threshold value, and output is identity identification, so that public security case solving work such as face arrangement and control alarm is completed.
However, in the practical public security and government anti-terrorism, stability maintenance and public security management work, the most important work is to prevent and timely discover abnormal persons or abnormal behaviors, further intervene early, take measures early, discover and control suspects as much as possible in the early stage of a case and even before the case, and avoid 'sheep death and reinforcement'.
Therefore, the mode of simply adopting the picture alarm of the control personnel cannot meet the requirements, all cases cannot know who the criminal is in advance, and the target clear face picture cannot be taken necessarily, so that the case cannot be really used by front-line management personnel in actual work.
Disclosure of Invention
Aiming at the problem that follow-up person analysis generally adopts a post analysis method and cannot prevent in advance, the invention provides a follow-up person early warning system and method based on portrait recognition.
A system for accompanying person early warning based on portrait recognition, the system comprising:
the portrait recognition module is used for detecting the portrait of each point location in real time;
an accompanying rule setting module for setting an accompanying rule and/or inputting a user-defined target person; the setting of the accompanying rule is to set one or more characteristic conditions of the accompanying person except for the portrait characteristics;
the logic judgment module is used for comparing the detected portrait extraction features with a feature set in a target person library, judging through a similarity threshold, recording the space-time trajectory of the target person when the target person is determined, and caching the portrait feature set of the target person in the effective accompanying time range of each space-time point; the system is also used for acquiring the accompanying personnel of the target personnel according to the set and the accompanying rule when the space-time trajectory of the target personnel reaches the accompanying detection condition;
the early warning output module is used for outputting the early warning of accompanying personnel;
the portrait track library is used for storing portrait characteristics, tracks and accompanying rules of all point positions;
and the target person library is used for storing the portrait feature set of the target person.
Preferably, the accompanying detection conditions are:
whether the space-time trajectory of the target person exceeds a threshold value of the occurrence space-time point location or the occurrence times.
Preferably, the logic determining module obtains the accompanying person of the target person, and includes:
when the space-time trajectory of the target person reaches the accompanying detection condition, performing cluster analysis on the portrait features in the set corresponding to the target person to obtain the space-time trajectory of the accompanying person;
and when the space-time trajectory of a certain accompanying person meets the accompanying space-time frequency setting condition of the target person, determining that the accompanying person is the accompanying person of the target person.
Preferably, the characteristic conditions in the accompanying rule setting module are accompanying person identity definition, accompanying time before and after, same person trajectory merging period, number of times of simultaneous occurrence of the accompanying person and the target person, sex, age, clothing or clothing.
Preferably, the logic judgment module is further configured to determine that the target person is a single trip when it is determined that there is no accompanying person meeting the accompanying rule;
and the early warning output module is also used for outputting the early warning of the target personnel who go out independently.
Preferably, the logic judgment module is configured to compare the detected portrait features with features in a target person library, and includes:
comparing the detected portrait characteristics with any characteristic or characteristics in a target person library, wherein the similarity exceeds a threshold value, and judging that the portrait is the portrait of the target person;
and comparing the detected portrait characteristics with all characteristics in the target person library, wherein the similarity is lower than a threshold value, and judging that the portrait is a non-target person portrait.
The invention also comprises an accompanying personnel early warning method based on portrait recognition, which is characterized by comprising the following steps:
s1, detecting the portrait of each point in real time;
s2, setting an accompanying rule and/or inputting a target person defined by a user; the setting of the accompanying rule is to set one or more characteristic conditions of the accompanying person except for the portrait characteristics;
storing the portrait characteristics, the track and the accompanying rules of each point location into a portrait track library;
s3, comparing the detected portrait extraction features with feature sets in a target person library, judging through a similarity threshold, recording the space-time trajectory of the target person when the target person is determined, and caching the portrait feature sets of the target person in the effective accompanying time range of each space-time point; the system is also used for acquiring the accompanying personnel of the target personnel according to the set and the accompanying rule when the space-time trajectory of the target personnel reaches the accompanying detection condition;
the target personnel base stores a characteristic set of the target personnel;
and S4, outputting the accompanying personnel early warning.
Preferably, the accompanying detection conditions are:
whether the space-time trajectory of the target person exceeds a threshold value of the occurrence space-time point location or the occurrence times.
Preferably, in S3, the acquiring the accompanying person of the target person includes:
when the space-time trajectory of the target person reaches the accompanying detection condition, performing cluster analysis on the portrait features in the set corresponding to the target person to obtain the space-time trajectory of the accompanying person;
and when the space-time trajectory of a certain accompanying person meets the accompanying space-time frequency setting condition of the target person, determining that the accompanying person is the accompanying person of the target person.
Preferably, in S3, the characteristic condition is an accompanying person identity definition, an accompanying time, a same person trajectory merging period, the number of times of simultaneous occurrence of the accompanying person and the target person, a sex, an age, clothing or clothing.
Preferably, the S3 further includes: when the target person is determined to have no accompanying person meeting accompanying rules, the target person is a single trip;
the S4 further includes: and carrying out early warning output on target personnel who travel alone.
Preferably, in S3, the comparing the detected portrait characteristics with the characteristics in the target person library includes:
comparing the detected portrait characteristics with any characteristic or characteristics in a target person library, wherein the similarity exceeds a threshold value, and judging that the portrait is the portrait of the target person;
and comparing the detected portrait characteristics with all characteristics in the target person library, wherein the similarity is lower than a threshold value, and judging that the portrait is a non-target person portrait.
The invention has the beneficial effects that based on the portrait detection and comparison identification technology, all the persons passing through the point positions are analyzed in real time, the front and rear person data are cached, the accompanying state analysis is carried out on the target person according to the portrait identification result, and therefore, the real-time early warning output is carried out on the person reaching the accompanying state set by the user, and the related management personnel can be reminded to intervene before the case, so that the early warning and prevention effects are achieved.
Drawings
FIG. 1 is a schematic diagram of a person identification-based accompanying person early warning system according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The invention is further described with reference to the following drawings and specific examples, which are not intended to be limiting.
As shown in fig. 1, the accompanying person early warning system based on portrait recognition according to this embodiment includes:
the portrait recognition module is used for detecting the portrait of each point location in real time;
an accompanying rule setting module for setting an accompanying rule and/or inputting a user-defined target person; the setting of the accompanying rule is to set one or more characteristic conditions of the accompanying person except for the portrait characteristics;
the logic judgment module is used for comparing the detected portrait extraction features with a feature set in a target person library, performing merging period judgment through similarity threshold judgment when the detected portrait extraction features are determined to be target persons, recording the space-time trajectory of the target persons, and caching the portrait feature set of the target persons in the effective accompanying time range of each space-time point; the system is also used for acquiring the accompanying personnel of the target personnel according to the set and the accompanying rule when the space-time trajectory of the target personnel reaches the accompanying detection condition;
the early warning output module is used for outputting the early warning of accompanying personnel;
the portrait track library is used for storing the portrait, the track and the accompanying rule of each point;
and the target person library is used for storing the characteristic set of the target person.
The characteristics of the target person library storage of the present embodiment include: portrait features and high-level features, the high-level features including accompanying person identity definition, accompanying time before and after, same person trajectory merging period, number of times that the accompanying person and the target person appear simultaneously, gender, age, clothing and/or clothing; the portrait trajectory library of the present embodiment is used for storing massive portrait trajectories and accompanying rules at certain locations (point locations) within a certain time range, and the portrait trajectories are used for acquiring spatiotemporal trajectories of target persons or accompanying persons, that is: all the personnel passing through the point location are stored, and follow-up judgment and early warning are facilitated.
In the embodiment, the same personnel track merging period is set through the accompanying rule setting module, and is used for merging the occurrence time points and the occurrence places of the personnel within the set time, so that the phenomenon that one person unconsciously outputs repeatedly for many times within a short period is avoided, and if one person wanders in front of the camera, the period is counted as one time of space-time point location.
The embodiment utilizes portrait detection, feature recognition and face recognition technology to compare the features of target personnel (such as children and the old) meeting certain physical and morphological features with the features of key personnel pre-stored in a target personnel library, confirm the identity, record the space-time track of the confirmed target personnel, cache the face feature records of the confirmed target personnel in a period of time before and after each appearance point, logically perform real-time monitoring and recording, when the track of the confirmed target personnel meets the accompanying detection condition, acquire the personnel space-time track information appearing simultaneously with the track, acquire the accompanying personnel of the target personnel according to the set accompanying rule, and output early warning. The embodiment analyzes all the personnel passing through the point location, caches the personnel data before and after the point location, and performs accompanying state analysis on the target personnel according to the portrait recognition result, so that the accompanying state personnel set by the user are output in a real-time early warning manner, and related management personnel can be reminded to intervene before the case, and the effects of early warning and prevention are achieved.
In a preferred embodiment, the accompanying detection conditions of the present embodiment are:
whether the space-time trajectory of the target person exceeds a threshold value of the occurrence space-time point location or the occurrence times.
In this embodiment, the number of times of occurrence at a certain point location/points or at a certain point location/points is used as a threshold value according to the space-time trajectory of the target person to determine whether the accompanying person of the target person needs to be detected.
In a preferred embodiment, the acquiring, by the logic determining module of this embodiment, the accompanying person of the target person includes:
when the space-time trajectory of the target personnel reaches the accompanying detection condition, performing cluster analysis on the portrait features in the set corresponding to the target personnel, and performing merging cycle judgment to obtain the space-time trajectory of the accompanying personnel;
and when the space-time trajectory of a certain accompanying person meets the accompanying space-time frequency setting condition of the target person, determining that the accompanying person is the accompanying person of the target person.
In the present embodiment, the spatiotemporal trajectory of the accompanying person is compared with the spatiotemporal trajectory of the target person, and the accompanying spatiotemporal frequency satisfies the setting conditions, for example: along with the times, the embodiment obtains the space-time trajectory information of the fellow persons according to the clustering result, outputs the persons meeting the condition of setting the along space-time frequency, realizes the gathering early warning of the key target persons, and can also realize the early warning of the trailing target persons.
In a preferred embodiment, the characteristic conditions in the association rule setting module are the accompanying person identity definition, the accompanying time, the same person trajectory merging period setting, the number of times of appearance of the accompanying person and the target person, the sex, the age, clothing (color, texture), or clothing (hat, glasses, mask).
The method and the device are used for narrowing the range optionally, and are combined with or combined with the human face features, so that the early warning range or accuracy is improved.
In a preferred embodiment, the logic determining module of this embodiment is further configured to determine that the target person is a single trip when it is determined that the target person does not have an accompanying person meeting the accompanying rule;
and the early warning output module is also used for outputting the early warning of the target personnel who go out independently.
According to the embodiment, the early warning of the target person during the unattended independent trip is carried out according to the set conditions.
In a preferred embodiment, the logic determining module of this embodiment is configured to compare the detected portrait features with features in the target person library, and includes:
comparing the detected portrait characteristics with any characteristic or characteristics in a target person library, wherein the similarity exceeds a threshold value, and judging that the portrait is the portrait of the target person;
and comparing the detected portrait characteristics with all characteristics in the target person library, wherein the similarity is lower than a threshold value, and judging that the portrait is a non-target person portrait.
The embodiment supports identity confirmation of the personnel, confirms that external personnel are not in the target personnel bank, confirms that key personnel are key personnel in the target personnel bank, and can enable the user to flexibly realize the identity confirmation through configuration so as to meet the actual service requirement. The identity of the accompanying person is confirmed in the same manner as the identity of the target person.
The embodiment also provides a method for early warning accompanying personnel based on portrait recognition, which comprises the following steps:
s1, detecting the portrait of each point in real time;
s2, setting an accompanying rule and/or inputting a target person defined by a user; the setting of the accompanying rule is to set one or more characteristic conditions of the accompanying person except for the portrait characteristics;
storing the portrait, the track and the accompanying rule of each point to a portrait track library;
s3, comparing the detected portrait extraction features with feature sets in a target person library, judging through a similarity threshold, recording the space-time trajectory of the target person when the target person is determined, and caching the portrait feature sets of the target person in the effective accompanying time range of each space-time point; the system is also used for acquiring the accompanying personnel of the target personnel according to the set and the accompanying rule when the space-time trajectory of the target personnel reaches the accompanying detection condition;
the target personnel base stores a characteristic set of the target personnel;
and S4, outputting the accompanying personnel early warning.
In a preferred embodiment, the accompanying detection conditions of the present embodiment are:
whether the space-time trajectory of the target person exceeds a threshold value of the occurrence space-time point location or the occurrence times.
In a preferred embodiment, in S3 of the present embodiment, the acquiring the accompanying person of the target person includes:
when the space-time trajectory of the target personnel reaches the accompanying detection condition, performing cluster analysis on the portrait features in the set corresponding to the target personnel, and performing merging cycle judgment to obtain the space-time trajectory of the accompanying personnel;
and when the space-time trajectory of a certain accompanying person meets the accompanying space-time frequency setting condition of the target person, determining that the accompanying person is the accompanying person of the target person.
In a preferred embodiment, in S3, the characteristic condition is an accompanying person identity definition, an accompanying time, the number of times of the accompanying person and the target person appearing simultaneously, a sex, an age, clothing or clothing.
In a preferred embodiment, S2 of the present embodiment further includes: when the target person is determined to have no accompanying person meeting accompanying rules, the target person is a single trip;
the S4 further includes: and carrying out early warning output on target personnel who travel alone.
In a preferred embodiment, the step S3 of this embodiment of comparing the detected portrait features with features in the target person library includes:
comparing the detected portrait characteristics with any characteristic or characteristics in a target person library, wherein the similarity exceeds a threshold value, and judging that the portrait is the portrait of the target person;
and comparing the detected portrait characteristics with all characteristics in the target person library, wherein the similarity is lower than a threshold value, and judging that the portrait is a non-target person portrait.
The specific embodiment is as follows: as shown in fig. 2, the accompanying person early warning method based on portrait recognition in this embodiment includes the following steps:
step 1: detecting the portrait of each point in real time;
step 2: carrying out feature recognition on each appearing portrait target, comparing the face features with a target personnel library, matching high-level features after the comparison is successful, and judging whether the portrait target is a target personnel, wherein the high-level features in the step comprise: accompanying personnel identity definition, accompanying time, accompanying times, sex, age, clothing and/or wearing objects of the accompanying personnel and the target personnel, if matching is successful, turning to the step 3, and if not, storing the portrait and the track characteristics into a portrait track library;
and step 3: recording the ID of the target person, comparing the ID with a portrait track library, checking whether the same person track merging period condition is met, if so, merging, otherwise, counting as an independent space-time node, and counting the occurrence frequency of the ID person;
and 4, step 4: if the accumulated occurrence times of the id personnel exceed the threshold value, the step 5 is carried out, and if not, the id portrait and the track characteristics of the target personnel are stored in a portrait track library;
and 5: recording and counting all person portrait feature sets of the id person in the accompanying time range of each occurrence point;
step 6: carrying out personnel clustering on the portrait feature set, checking whether the same personnel track merging period condition is met, if so, merging, and if not, counting as an independent space-time node to obtain a space-time track of each id;
and 7: comparing the space-time trajectory of each id obtained in the step 6 with the space-time trajectory of the target person id, if the number of times of the accompanying is up to a threshold value, determining that the corresponding id is the id of the accompanying person, and turning to the step 8, otherwise, ending the thread;
and 8: and judging whether the identity of the accompanying personnel of the id accords with an accompanying rule, if so, outputting the early warning of the accompanying personnel, recording the position where the accompanying personnel appear, and if not, ending the thread.
Although the invention herein has been described with reference to particular embodiments, it is to be understood that these embodiments are merely illustrative of the principles and applications of the present invention. It is therefore to be understood that numerous modifications may be made to the illustrative embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that features described in different dependent claims and herein may be combined in ways different from those described in the original claims. It is also to be understood that features described in connection with individual embodiments may be used in other described embodiments.
Claims (12)
1. An accompanying person early warning system based on portrait recognition, the system comprising:
the portrait recognition module is used for detecting the portrait of each point location in real time;
an accompanying rule setting module for setting an accompanying rule and/or inputting a user-defined target person; the setting of the accompanying rule is to set one or more characteristic conditions of the accompanying person except for the portrait characteristics;
the logic judgment module is used for comparing the detected portrait extraction features with a feature set in a target person library, judging through a similarity threshold, recording the space-time trajectory of the target person when the target person is determined, and caching the portrait feature set of the target person in the effective accompanying time range of each space-time point; the system is also used for acquiring the accompanying personnel of the target personnel according to the set and the accompanying rule when the space-time trajectory of the target personnel reaches the accompanying detection condition;
the early warning output module is used for outputting the early warning of accompanying personnel;
the portrait track library is used for storing portrait characteristics, tracks and accompanying rules of all point positions;
and the target person library is used for storing the characteristic set of the target person.
2. The system of claim 1, wherein the accompanying detection conditions are:
whether the space-time trajectory of the target person exceeds a threshold value of the occurrence space-time point location or the occurrence times.
3. The accompanying person early warning system based on face recognition as claimed in claim 1 or 2, wherein the logic determination module obtains the accompanying person of the target person, and comprises:
when the space-time trajectory of the target person reaches the accompanying detection condition, performing cluster analysis on the portrait features in the set corresponding to the target person to obtain the space-time trajectory of the accompanying person;
and when the space-time trajectory of a certain accompanying person meets the accompanying space-time frequency setting condition of the target person, determining that the accompanying person is the accompanying person of the target person.
4. The accompanying person early warning system based on portrait recognition according to claim 3, wherein the characteristic conditions in the accompanying rule setting module are accompanying person identity definition, accompanying time before and after, same person track merging period, number of times of simultaneous appearance of the accompanying person and the target person, gender, age, clothing or clothing.
5. The accompanying personnel early warning system based on portrait recognition as claimed in claim 1, wherein the logic judgment module is further configured to determine that the target personnel is a single trip when it is determined that the target personnel does not have an accompanying personnel meeting accompanying rules;
and the early warning output module is also used for outputting the early warning of the target personnel who go out independently.
6. The system of claim 1, wherein the logic determination module is configured to compare the detected portrait characteristics with characteristics in a target people library, and comprises:
comparing the detected portrait characteristics with any characteristic or characteristics in a target person library, wherein the similarity exceeds a threshold value, and judging that the portrait is the portrait of the target person;
and comparing the detected portrait characteristics with all characteristics in the target person library, wherein the similarity is lower than a threshold value, and judging that the portrait is a non-target person portrait.
7. An accompanying person early warning method based on portrait recognition is characterized by comprising the following steps:
s1, detecting the portrait of each point in real time;
s2, setting an accompanying rule and/or inputting a target person defined by a user; the setting of the accompanying rule is to set one or more characteristic conditions of the accompanying person except for the portrait characteristics;
storing the portrait characteristics, the track and the accompanying rules of each point location into a portrait track library;
s3, comparing the detected portrait extraction features with feature sets in a target person library, judging through a similarity threshold, recording the space-time trajectory of the target person when the target person is determined, and caching the portrait feature sets of the target person in the effective accompanying time range of each space-time point; the system is also used for acquiring the accompanying personnel of the target personnel according to the set and the accompanying rule when the space-time trajectory of the target personnel reaches the accompanying detection condition; the target personnel base stores a characteristic set of the target personnel;
and S4, outputting the accompanying personnel early warning.
8. The accompanying person early warning method based on portrait recognition according to claim 7, wherein the accompanying detection conditions are:
whether the space-time trajectory of the target person exceeds a threshold value of the occurrence space-time point location or the occurrence times.
9. The accompanying person early warning method based on portrait recognition according to claim 7 or 8, wherein in the step S3, acquiring the accompanying person of the target person includes:
when the space-time trajectory of the target person reaches the accompanying detection condition, performing cluster analysis on the portrait features in the set corresponding to the target person to obtain the space-time trajectory of the accompanying person;
and when the space-time trajectory of a certain accompanying person meets the accompanying space-time frequency setting condition of the target person, determining that the accompanying person is the accompanying person of the target person.
10. The accompanying person early warning method based on portrait recognition as claimed in claim 9, wherein in S3, the characteristic conditions are accompanying person identity definition, accompanying time before and after, same person track merging period, number of times of simultaneous occurrence of the accompanying person and the target person, gender, age, clothing or clothing.
11. The portrait recognition-based accompanying person early warning method according to claim 7, wherein the S3 further includes: when the target person is determined to have no accompanying person meeting accompanying rules, the target person is a single trip;
the S4 further includes: and carrying out early warning output on target personnel who travel alone.
12. The method for pre-warning accompanying people based on portrait recognition according to claim 7, wherein the step of S3, for comparing the detected portrait features with features in a target people library, comprises:
comparing the detected portrait characteristics with any characteristic or characteristics in a target person library, wherein the similarity exceeds a threshold value, and judging that the portrait is the portrait of the target person;
and comparing the detected portrait characteristics with all characteristics in the target person library, wherein the similarity is lower than a threshold value, and judging that the portrait is a non-target person portrait.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811199697.4A CN109376639B (en) | 2018-10-16 | 2018-10-16 | Accompanying personnel early warning system and method based on portrait recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811199697.4A CN109376639B (en) | 2018-10-16 | 2018-10-16 | Accompanying personnel early warning system and method based on portrait recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109376639A CN109376639A (en) | 2019-02-22 |
| CN109376639B true CN109376639B (en) | 2021-12-17 |
Family
ID=65399851
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811199697.4A Active CN109376639B (en) | 2018-10-16 | 2018-10-16 | Accompanying personnel early warning system and method based on portrait recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109376639B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109947793B (en) * | 2019-03-20 | 2022-05-31 | 深圳市北斗智能科技有限公司 | Method and device for analyzing accompanying relationship and storage medium |
| CN109977108B (en) * | 2019-04-03 | 2021-04-27 | 深圳市甲易科技有限公司 | Behavior trajectory library-based multi-trajectory collision analysis method |
| CN110276272A (en) * | 2019-05-30 | 2019-09-24 | 罗普特科技集团股份有限公司 | Confirm method, apparatus, the storage medium of same administrative staff's relationship of label personnel |
| CN110569720B (en) * | 2019-07-31 | 2022-06-07 | 安徽四创电子股份有限公司 | Audio and video intelligent identification processing method based on audio and video processing system |
| CN110674749B (en) * | 2019-09-24 | 2022-10-25 | 北京明智和术科技有限公司 | Fake-licensed vehicle identification method and device and storage medium |
| CN111222404A (en) * | 2019-11-15 | 2020-06-02 | 北京市商汤科技开发有限公司 | Method, device and system for detecting co-pedestrian, electronic equipment and storage medium |
| CN112653990A (en) * | 2020-09-18 | 2021-04-13 | 武汉爱迪科技股份有限公司 | Screening algorithm and system for close contact personnel |
| CN112364176A (en) * | 2020-10-26 | 2021-02-12 | 青岛海信网络科技股份有限公司 | Method, equipment and system for constructing personnel action track |
| CN114863637B (en) * | 2022-03-30 | 2023-10-17 | 财上门科技(北京)有限公司 | Hotel security management system and method |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103400148A (en) * | 2013-08-02 | 2013-11-20 | 上海泓申科技发展有限公司 | Video analysis-based bank self-service area tailgating behavior detection method |
| CN104183039A (en) * | 2014-08-19 | 2014-12-03 | 花韬 | Anti-following system with function of human body biologic characteristic identification |
| CN105022999A (en) * | 2015-07-12 | 2015-11-04 | 上海微桥电子科技有限公司 | Man code company real-time acquisition system |
| CN106778655A (en) * | 2016-12-27 | 2017-05-31 | 华侨大学 | A kind of entrance based on human skeleton is trailed and enters detection method |
| CN106875679A (en) * | 2017-03-14 | 2017-06-20 | 东软集团股份有限公司 | Recognize the method and device of escort vehicle |
| CN107292907A (en) * | 2017-07-14 | 2017-10-24 | 灵动科技（北京）有限公司 | A kind of method to following target to be positioned and follow equipment |
| CN107861979A (en) * | 2017-09-27 | 2018-03-30 | 汉柏科技有限公司 | A kind of adjoint personnel's acquisition methods and system based on recognition of face |
| CN108122013A (en) * | 2017-12-29 | 2018-06-05 | 重庆锐纳达自动化技术有限公司 | One kind, which follows, to be excluded non-to follow mesh calibration method in movement |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003088157A1 (en) * | 2002-04-08 | 2003-10-23 | Newton Security Inc. | Tailgating and reverse entry detection, alarm, recording and prevention using machine vision |
| US9142106B2 (en) * | 2012-05-23 | 2015-09-22 | Honeywell International, Inc. | Tailgating detection |
| US10235854B2 (en) * | 2014-08-19 | 2019-03-19 | Sensormatic Electronics, LLC | Tailgating detection in frictionless access control system |
-
2018
- 2018-10-16 CN CN201811199697.4A patent/CN109376639B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103400148A (en) * | 2013-08-02 | 2013-11-20 | 上海泓申科技发展有限公司 | Video analysis-based bank self-service area tailgating behavior detection method |
| CN104183039A (en) * | 2014-08-19 | 2014-12-03 | 花韬 | Anti-following system with function of human body biologic characteristic identification |
| CN105022999A (en) * | 2015-07-12 | 2015-11-04 | 上海微桥电子科技有限公司 | Man code company real-time acquisition system |
| CN106778655A (en) * | 2016-12-27 | 2017-05-31 | 华侨大学 | A kind of entrance based on human skeleton is trailed and enters detection method |
| CN106875679A (en) * | 2017-03-14 | 2017-06-20 | 东软集团股份有限公司 | Recognize the method and device of escort vehicle |
| CN107292907A (en) * | 2017-07-14 | 2017-10-24 | 灵动科技（北京）有限公司 | A kind of method to following target to be positioned and follow equipment |
| CN107861979A (en) * | 2017-09-27 | 2018-03-30 | 汉柏科技有限公司 | A kind of adjoint personnel's acquisition methods and system based on recognition of face |
| CN108122013A (en) * | 2017-12-29 | 2018-06-05 | 重庆锐纳达自动化技术有限公司 | One kind, which follows, to be excluded non-to follow mesh calibration method in movement |
Non-Patent Citations (1)
| Title |
|---|
| 基于双层异构深度神经网络模型的人脸识别关键技术研究;李晋等;《电信工程技术与标准化》;20170615(第06期);第29-34页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109376639A (en) | 2019-02-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109376639B (en) | Accompanying personnel early warning system and method based on portrait recognition | |
| CN106203458B (en) | Crowd video analysis method and system | |
| US11704936B2 (en) | Object tracking and best shot detection system | |
| CN106997629B (en) | Access control method, apparatus and system | |
| CN110648352B (en) | Abnormal event detection method and device and electronic equipment | |
| JP6854881B2 (en) | Face image matching system and face image search system | |
| WO2014132841A1 (en) | Person search method and platform occupant search device | |
| CN105868689A (en) | Cascaded convolutional neural network based human face occlusion detection method | |
| CN106355154B (en) | Method for detecting frequent passing of people in surveillance video | |
| CN108537922A (en) | Visitor's method for early warning based on recognition of face and system | |
| CN113223046B (en) | Method and system for identifying prisoner behaviors | |
| CN112183162A (en) | Face automatic registration and recognition system and method in monitoring scene | |
| KR101957677B1 (en) | System for learning based real time guidance through face recognition and the method thereof | |
| JP7201072B2 (en) | Surveillance device, suspicious object detection method, and program | |
| CN109615256A (en) | Wisdom garden security protection risk control method, storage medium, electronic equipment and system | |
| Cuevas et al. | Detection of stationary foreground objects using multiple nonparametric background-foreground models on a finite state machine | |
| CN112565676A (en) | Method, system and equipment for identifying violation of on-duty personnel based on intelligent camera | |
| KR101515214B1 (en) | Identification method using face recognition and entrance control system and method thereof using the identification method | |
| CN110717357B (en) | Early warning method and device, electronic equipment and storage medium | |
| CN108898042A (en) | A kind of detection method applied to user's abnormal behaviour in ATM machine cabin | |
| CN110674761A (en) | Regional behavior early warning method and system | |
| CN111291596A (en) | Early warning method and device based on face recognition | |
| CN111462417A (en) | Multi-information verification system and multi-information verification method for unmanned bank | |
| Yuganthini et al. | Activity tracking of employees in industries using computer vision | |
| CN113269127A (en) | Face recognition and pedestrian re-recognition monitoring method and system for real-time automatic database building |
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 |