CN103353940A - Identification method and system for dynamically adjusting comparison sequence based on probability of occurrence - Google Patents
Identification method and system for dynamically adjusting comparison sequence based on probability of occurrence Download PDFInfo
- Publication number
- CN103353940A CN103353940A CN2013101796574A CN201310179657A CN103353940A CN 103353940 A CN103353940 A CN 103353940A CN 2013101796574 A CN2013101796574 A CN 2013101796574A CN 201310179657 A CN201310179657 A CN 201310179657A CN 103353940 A CN103353940 A CN 103353940A
- Authority
- CN
- China
- Prior art keywords
- guarded region
- face
- people
- personnel
- probability
- 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.)
- Granted
Links
Images
Abstract
The invention provides an identification method and a system for dynamically adjusting a comparison sequence based on the probability of occurrence. The method comprises the steps of: (1) analyzing a law of time when a person enters a monitoring area so as to determine the probability of occurrence of the person enters the monitoring area in each time period; (2) analyzing a law of time when the person leaves the monitoring area so as to determine the probability of occurrence of the person leaves the monitoring area in each time period; (3) determining a comparison sequence C according to the probability of occurrence from high to low on the basis of the time period when the person enters the monitoring area; (4) comparing according to the comparison sequence C when identification is carried out on faces acquired by an entrance video monitoring system of the monitoring area; (5) determining a comparison sequence D according to the probability of occurrence from high to low on the basis of the time period when the person leaves the monitoring area; and (6) comparing according to the comparison sequence D when identification is carried out on faces acquired by an exit video monitoring system of the monitoring area.
Description
Technical field
The present invention relates to the regional protection and monitor field of intelligent and real-time, especially relate to a kind of recognition methods and system that dynamically adjusts the comparison order based on probability of occurrence.
Background technology
Prior art adopts the Static Human Face data, adopts the fixing face database of people's face data when guarded region entrance video monitoring system is carried out recognition of face; The fixing face database of the same people of employing face data when guarded region outlet video monitoring system is carried out recognition of face; And essentially identical face database is adopted in the recognition of face of guarded region entrance and exit video monitoring.
Face alignment in the face recognition process is unordered at random, or is the order that depends on people's face data in the face database.System as shown in Figure 1 has people's face data of N people and storage order unordered at random in the guarded region entrance recognition of face data, have people's face data of M people and storage order unordered at random in the guarded region outlet recognition of face database.
The shortcoming of prior art is: the fixing face database of people's face data is all adopted in the recognition of face of guarded region entrance and exit video monitoring, and these two databases are generally identical.Carried out unnecessary recognition of face comparison, wasted system time, recognition speed is slow, and invalid computation has been wasted the energy.
Summary of the invention
In order to address the above problem, the invention provides and a kind ofly dynamically adjust the recognition methods of comparison order based on probability of occurrence, the method comprises:
Step (1) passes in and out the temporal regularity that the concrete personnel of time history record automatic analysis advance people's guarded region according to the guarded region personnel and enters the probability of occurrence of guarded region in each time period to determine these personnel;
Step (2) passes in and out the temporal regularity that the concrete personnel of time history record automatic analysis leave guarded region according to the guarded region personnel and leaves the probability of occurrence of guarded region in each time period to determine these personnel;
Step (3) enters time period of guarded region according to concrete personnel, the face alignment order C when determining in this time period that according to probability of occurrence order from high to low the facial image that enters guarded region carried out recognition of face;
When people's face that step (4) is obtained guarded region entrance video monitoring system is identified, people's face data and the people's face data in the current monitored area entrance face database obtained are compared by the above-mentioned order C that compares;
Step (5) is left time period of guarded region according to concrete personnel, the face alignment order D when determining in this time period that according to probability of occurrence order from high to low the facial image that leaves guarded region carried out recognition of face;
When people's face that step (6) is obtained guarded region outlet video monitoring system is identified, the people's face data in people's face data of obtaining and the current monitored area outlet video monitoring system face database are compared sequentially by above-mentioned that D compares.
Further, described step (4) comprising: the personnel that will enter guarded region delete from guarded region entrance face database, and this people's relevant people face data are added in the guarded region outlet video monitoring face database.
Further, step (6) comprising: people's face data of this relating to persons of deletion from guarded region outlet face database, add people's face data of this relating to persons to guarded region entrance face database.
The present invention also provides a kind of and has dynamically adjusted comparison recognition system sequentially based on probability of occurrence, and this system comprises:
Pass in and out the module of temporal regularity to determine that these personnel enter the probability of occurrence of guarded region in each time period that the concrete personnel of time history record automatic analysis advance people's guarded region according to the guarded region personnel;
Pass in and out the temporal regularity that the concrete personnel of time history record automatic analysis leave guarded region according to the guarded region personnel and leave the module of the probability of occurrence of guarded region in each time period to determine these personnel;
Guarded region entrance dynamic human face identification database enters time period of guarded region according to concrete personnel, the face alignment order C when determining in this time period that according to probability of occurrence order from high to low the facial image that enters guarded region carried out recognition of face;
Guarded region entrance video monitoring system for people's face data of obtaining the personnel of entering, is only compared people's face data and the people's face data in the current monitored area entrance dynamic human face database obtained when identifying by the above-mentioned order C that compares;
Guarded region outlet dynamic human face identification database leaves time period of guarded region according to concrete personnel, the face alignment order D when determining in this time period that according to probability of occurrence order from high to low the facial image that leaves guarded region carried out recognition of face;
Guarded region outlet video monitoring system is used for obtaining people's face data of the personnel of leaving, only the people's face data in people's face data of obtaining and the current monitored area outlet dynamic human face database is compared sequentially by above-mentioned that D compares when identifying.
Further, after guarded region entrance video monitoring system is being compared end, the personnel that enter guarded region are deleted from guarded region entrance face database, and this people's relevant people face data are added in the guarded region outlet video monitoring face database.
Further, after comparison finished, the relevant people's face data of personnel were left in deletion from guarded region outlet face database, add people's face data of this relating to persons to guarded region entrance face database in guarded region outlet video monitoring system.
Further, this system comprises that the guarded region personnel pass in and out database of record, is used for the storage personnel and passes in and out record.
Description of drawings
Fig. 1 is the structural representation that illustrates according to the system of prior art.
Fig. 2 illustrates the view that enters guarded region according to of the present invention.
Fig. 3 illustrates the view of leaving guarded region according to of the present invention.
Fig. 4 is the module map that illustrates according to system of the present invention.
Embodiment
For above-mentioned purpose of the present invention, feature and advantage are become apparent more, the present invention is further detailed explanation below in conjunction with the drawings and specific embodiments:
As shown in Figure 2, described in the solution of the present invention, according to probability of occurrence people's face comparing arranged sequentially order from high to low, the constitutional diagram when detecting someone A and enter into guarded region.
As shown in Figure 3, described in the solution of the present invention, according to probability of occurrence people's face comparing arranged sequentially order from high to low, the constitutional diagram when detecting someone A and leave guarded region.
As shown in Figure 4, the scheme that adopts the dynamic human face database according to the present invention and sequentially carry out recognition of face based on probability of occurrence adjustment comparison has been described:
Pass in and out the temporal regularity that the concrete personnel of time history record automatic analysis advance people's guarded region according to the guarded region personnel and enter the probability of occurrence of guarded region in each time period to determine these personnel.For example, entering the guarded region number of times and always enter the probability of occurrence that number of times calculates this time period in this time period of 8:00 to 9:00 according to personnel A is 80%, personnel A is 10% at the probability of occurrence of this time period of 9:00 to 10:00, and the probability of occurrence of other times section is 10%.
Pass in and out the temporal regularity that the concrete personnel of time history record automatic analysis leave guarded region according to the guarded region personnel and leave the probability of occurrence of guarded region in each time period to determine these personnel.For example, leave the guarded region number of times and total to leave the probability of occurrence that number of times calculates this time period be 85% according to personnel A in this time period of 17:00 to 18:00, personnel A is 10% at the probability of occurrence of this time period of 18:00 to 19:00, and the probability of occurrence of other times section is 5%.
Guarded region entrance dynamic human face database enters time period of guarded region according to concrete personnel, the face alignment order C when determining in this time period that according to probability of occurrence order from high to low the facial image that enters guarded region carried out recognition of face.
Guarded region outlet dynamic human face database leaves time period of guarded region according to concrete personnel, the face alignment order D when determining in this time period that according to probability of occurrence order from high to low the facial image that leaves guarded region carried out recognition of face.
When people's face that guarded region entrance video monitoring system is obtained is identified, only people's face data and the people's face data in the current monitored area entrance dynamic human face database obtained are compared by the above-mentioned order C that compares, the personnel that enter guarded region are deleted from guarded region entrance dynamic human face database, when the follow-up people's face that obtains of entrance video monitoring system is identified, need not the people's face data that will obtain with before entered should the zone people's face data (deleting in the database) compare, preferentially people's face data of the high people of probability of occurrence compared, thereby accelerate the recognition of face speed of porch, find fast and will know others, and add this people's relevant people face data in the guarded region outlet video monitoring face database (the probability decision that occurs in this time period according to this people is added on it position that exports in the face database).
When people's face that guarded region outlet video monitoring system is obtained is identified, only the people's face data in people's face data of obtaining and the current monitored area outlet dynamic human face database are compared sequentially by above-mentioned that D compares, preferentially people's face data of the high people of probability of occurrence compared, thereby accelerate the recognition of face speed in exit, find fast and will know others, and from guarded region outlet face database people's face data of this relating to persons of deletion, add people's face data of this relating to persons to guarded region entrance face database (probability that occurs at entrance according to this people determines it is added on the position in the entrance face database) within this time period.
This system comprises that further the guarded region personnel pass in and out database of record, is used for the storage personnel and passes in and out record.
The effect that the method according to this invention and system can obtain is: the dynamic human face database is all adopted in the recognition of face of guarded region entrance and exit video monitoring, pass in and out guarded region according to personnel and automatically adjust face database, determine the priority of face alignment with the height of probability of occurrence, find fast people's face data of matching image, avoid carrying out unnecessary recognition of face, save computing time, accelerated recognition speed, and avoided invalid computation to waste energy.
More than be the detailed description that the preferred embodiments of the present invention are carried out, but those of ordinary skill in the art should be appreciated that within the scope of the present invention, and guided by the spirit various improvement, interpolation and replacement all are possible.These are all in the protection domain that claim of the present invention limits.
Claims (7)
1. dynamically adjust the recognition methods of comparison order based on probability of occurrence for one kind, it is characterized in that the method comprises:
Step (1) passes in and out the temporal regularity that the concrete personnel of time history record automatic analysis advance people's guarded region according to the guarded region personnel and enters the probability of occurrence of guarded region in each time period to determine these personnel;
Step (2) passes in and out the temporal regularity that the concrete personnel of time history record automatic analysis leave guarded region according to the guarded region personnel and leaves the probability of occurrence of guarded region in each time period to determine these personnel;
Step (3) enters time period of guarded region according to concrete personnel, the face alignment order C when determining in this time period that according to probability of occurrence order from high to low the facial image that enters guarded region carried out recognition of face;
When people's face that step (4) is obtained guarded region entrance video monitoring system is identified, people's face data and the people's face data in the current monitored area entrance face database obtained are compared by the above-mentioned order C that compares;
Step (5) is left time period of guarded region according to concrete personnel, the face alignment order D when determining in this time period that according to probability of occurrence order from high to low the facial image that leaves guarded region carried out recognition of face;
When people's face that step (6) is obtained guarded region outlet video monitoring system is identified, the people's face data in people's face data of obtaining and the current monitored area outlet video monitoring system face database are compared sequentially by above-mentioned that D compares.
2. method according to claim 1 is characterized in that:
Step (4) further comprises: the personnel that will enter guarded region delete from guarded region entrance face database, and this people's relevant people face data are added in the guarded region outlet video monitoring face database.
3. method according to claim 1 and 2 is characterized in that:
Step (6) further comprises: people's face data of this relating to persons of deletion from guarded region outlet face database, add people's face data of this relating to persons to guarded region entrance face database.
4. dynamically adjust the recognition system of comparison order based on probability of occurrence for one kind, it is characterized in that this system comprises:
Pass in and out the module of temporal regularity to determine that these personnel enter the probability of occurrence of guarded region in each time period that the concrete personnel of time history record automatic analysis advance people's guarded region according to the guarded region personnel;
Pass in and out the temporal regularity that the concrete personnel of time history record automatic analysis leave guarded region according to the guarded region personnel and leave the module of the probability of occurrence of guarded region in each time period to determine these personnel;
Guarded region entrance dynamic human face identification database enters time period of guarded region according to concrete personnel, the face alignment order C when determining in this time period that according to probability of occurrence order from high to low the facial image that enters guarded region carried out recognition of face;
Guarded region entrance video monitoring system for people's face data of obtaining the personnel of entering, is only compared people's face data and the people's face data in the current monitored area entrance dynamic human face database obtained when identifying by the above-mentioned order C that compares;
Guarded region outlet dynamic human face identification database leaves time period of guarded region according to concrete personnel, the face alignment order D when determining in this time period that according to probability of occurrence order from high to low the facial image that leaves guarded region carried out recognition of face;
Guarded region outlet video monitoring system is used for obtaining people's face data of the personnel of leaving, only the people's face data in people's face data of obtaining and the current monitored area outlet dynamic human face database is compared sequentially by above-mentioned that D compares when identifying.
5. system according to claim 4 is characterized in that:
Guarded region entrance video monitoring system is deleted the personnel that enter guarded region after comparison finishes from guarded region entrance face database, and this people's relevant people face data are added in the guarded region outlet video monitoring face database.
6. it is characterized in that according to claim 4 or 5 described systems:
Guarded region outlet video monitoring system is after comparison finishes, and the relevant people's face data of personnel are left in deletion from guarded region outlet face database, add people's face data of this relating to persons to guarded region entrance face database.
7. system according to claim 6 is characterized in that:
This system comprises that further the guarded region personnel pass in and out database of record, is used for the storage personnel and passes in and out record.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310179657.4A CN103353940B (en) | 2013-05-15 | 2013-05-15 | A kind of recognition methods dynamically adjusting comparison sequence based on probability of occurrence and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310179657.4A CN103353940B (en) | 2013-05-15 | 2013-05-15 | A kind of recognition methods dynamically adjusting comparison sequence based on probability of occurrence and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103353940A true CN103353940A (en) | 2013-10-16 |
| CN103353940B CN103353940B (en) | 2016-12-07 |
Family
ID=49310307
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310179657.4A Expired - Fee Related CN103353940B (en) | 2013-05-15 | 2013-05-15 | A kind of recognition methods dynamically adjusting comparison sequence based on probability of occurrence and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103353940B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105868413A (en) * | 2016-04-28 | 2016-08-17 | 南京信息职业技术学院 | Video retrieval method for quickly locating sudden case |
| CN105913037A (en) * | 2016-04-26 | 2016-08-31 | 广东技术师范学院 | Face identification and radio frequency identification based monitoring and tracking system |
| CN107292228A (en) * | 2017-05-05 | 2017-10-24 | 珠海数字动力科技股份有限公司 | A kind of method for accelerating face recognition search speed |
| WO2021063037A1 (en) * | 2019-09-30 | 2021-04-08 | 华为技术有限公司 | Person database partitioning method, and device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102625081A (en) * | 2012-02-20 | 2012-08-01 | 华焦宝 | Method and system for tracking target in visit place |
| CN102622580A (en) * | 2012-02-20 | 2012-08-01 | 华焦宝 | Face detection and recognition method and system |
-
2013
- 2013-05-15 CN CN201310179657.4A patent/CN103353940B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102625081A (en) * | 2012-02-20 | 2012-08-01 | 华焦宝 | Method and system for tracking target in visit place |
| CN102622580A (en) * | 2012-02-20 | 2012-08-01 | 华焦宝 | Face detection and recognition method and system |
Non-Patent Citations (1)
| Title |
|---|
| 宋红等: "基于人脸检测与跟踪的智能监控系统", 《北京理工大学学报》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105913037A (en) * | 2016-04-26 | 2016-08-31 | 广东技术师范学院 | Face identification and radio frequency identification based monitoring and tracking system |
| CN105868413A (en) * | 2016-04-28 | 2016-08-17 | 南京信息职业技术学院 | Video retrieval method for quickly locating sudden case |
| CN105868413B (en) * | 2016-04-28 | 2019-09-20 | 南京信息职业技术学院 | A kind of video retrieval method of quick positioning burst merit |
| CN107292228A (en) * | 2017-05-05 | 2017-10-24 | 珠海数字动力科技股份有限公司 | A kind of method for accelerating face recognition search speed |
| WO2021063037A1 (en) * | 2019-09-30 | 2021-04-08 | 华为技术有限公司 | Person database partitioning method, and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103353940B (en) | 2016-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110855935B (en) | Personnel track generation system and method based on multiple cameras | |
| CN103279740B (en) | A kind of method and system utilizing dynamic data base to accelerate intelligent monitoring recognition of face | |
| US20190325229A1 (en) | Target object identifying device, target object identifying method and target object identifying program | |
| CN106203458B (en) | Crowd video analysis method and system | |
| CN110222640B (en) | Method, device and method for identifying suspect in monitoring site and storage medium | |
| CN106780250B (en) | Intelligent community security event processing method and system based on Internet of things technology | |
| CN103530912A (en) | Attendance recording system having emotion identification function, and method thereof | |
| CN107067504A (en) | A kind of recognition of face safety-protection system and a suspect's detection and method for early warning | |
| WO2021180060A1 (en) | Channel gate control method, apparatus and system | |
| CN105427421A (en) | Entrance guard control method based on face recognition | |
| US20070183634A1 (en) | Auto Individualization process based on a facial biometric anonymous ID Assignment | |
| CN103353940A (en) | Identification method and system for dynamically adjusting comparison sequence based on probability of occurrence | |
| CN111460940B (en) | Method and system for studying and judging stranger footfall points | |
| CN108960156B (en) | Face detection and recognition method and device | |
| CN110334564A (en) | A kind of permanent resident population's recognition methods and system based on target following | |
| CN110751675B (en) | Urban pet activity track monitoring method based on image recognition and related equipment | |
| CN110969215A (en) | Clustering method and device, storage medium and electronic device | |
| CN106327619A (en) | In/out management method and system thereof | |
| CN109757395A (en) | A kind of pet behavioral value monitoring system and method | |
| CN112634561A (en) | Safety alarm method and system based on image recognition | |
| CN107992958A (en) | Population super-limit prewarning method based on ARMA | |
| CN116386120B (en) | A noninductive control management system for wisdom campus dormitory | |
| CN112800975A (en) | Behavior identification method in security check channel based on image processing | |
| CN108615286A (en) | A kind of cell intelligent management method and system | |
| CN101847204A (en) | Abnormal face detecting method for embedded digital video recorder (DVR) system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20161207 Termination date: 20190515 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |