CN103246869B - Method is monitored in crime based on recognition of face and behavior speech recognition - Google Patents
Method is monitored in crime based on recognition of face and behavior speech recognition Download PDFInfo
- Publication number
- CN103246869B CN103246869B CN201310139660.3A CN201310139660A CN103246869B CN 103246869 B CN103246869 B CN 103246869B CN 201310139660 A CN201310139660 A CN 201310139660A CN 103246869 B CN103246869 B CN 103246869B
- Authority
- CN
- China
- Prior art keywords
- face
- personnel
- alert notification
- monitoring
- characteristic information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Alarm Systems (AREA)
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The present invention provides a kind of crime based on face recognition technology and behavior speech recognition to monitor method, comprises the steps: step 1, by photographic head recorded video, and described video dimensionality reduction is become pictorial information set；Step 2, pictorial information set is carried out Activity recognition comparison according to intelligent behavior pattern, if comparison is passed through, send pre-alert notification and preserve video；Step 3, people's police on duty examine alert, are determined the position of described photographic head and neighbouring police strength situation by GPS locating and tracking, send alert information to neighbouring people's police；If without people's police on duty, then determine the position of described photographic head and neighbouring police strength situation automatically by GPS locating and tracking, send alert information to neighbouring people's police and operator on duty.The present invention arranges different intelligent behavior patterns according to the monitoring demand of different occasions, introduces and has purpose to monitor, it is achieved pre-alarm and prevention in advance, it is prevented that cases worsens further, shortens the time of solving a case, improves case-solving rate.
Description
Technical field
The present invention relates to a kind of crime based on recognition of face and behavior speech recognition and monitor method.
Background technology
Current face recognition technology is based on the face feature of people, facial image recognition or video flowing to input first determine whether whether it exists face, if there is face, then provide the positional information of the position of each face, size and each major facial organ further, then according to these information, extract the identity characteristic contained in each face further, and itself and known face are contrasted, thus identifying the identity of each face；Current recognition of face is based only on the recognition of face of big triangle, only completes the identification of face the first half, lacks the identification of the identification of face the latter half, contour feature, and when nozzle type changes, ratio change combines；So when expression changes, error is also bigger than normal, and discrimination is naturally on the low side.
Simultaneously, the method of existing monitoring criminal behavior is usually employing camera head monitor combination application, photographic head has been video record and real-time manual monitoring, cases is transferred video after occurring and is checked clue, big amount of video search efficiency is low, application in fighting crime is subjected to very big restriction, and " retrieval is slow, comparison is slow " has become the subject matter that restricting current grass-roots unit Video Applications effect plays afterwards.
Owing to the discrimination of existing face recognition technology is not high, cause utilizing camera head monitor crowd massing and the technology based on recognition of face such as hover for a long time to be used widely, therefore cannot to crowd massing, hover for a long time, frequently come in and go out, cross over circumference, abduct the behaviors such as child and carry out effective monitoring and early warning, cannot find that potential criminal behavior also just cannot accomplish pre-alarm and prevention in advance in time, be unfavorable for the maintenance of social order and stable.
Summary of the invention
The technical problem to be solved in the present invention, it is in that to provide a kind of crime based on recognition of face and behavior speech recognition to monitor method, its monitoring demand according to different occasions, different intelligent behavior patterns is set, purpose is had to monitor thus introducing, intelligence carries out behavior linguistic analysis comparison, thus realizing timely pre-alarm and prevention in advance, then again through artificial really alert, make clear and definite responding to process, prevent cases from worsening further, lock suspected target in time simultaneously, identification comparison again through face database, shorten and solve a case the time, improve case-solving rate, change the tracking afterwards of present stage, transfer the inefficient pattern of multitude of video retrieval investigation.
The present invention is achieved in that
Method is monitored in crime based on face recognition technology and behavior speech recognition, comprises the steps:
Step 1, by photographic head recorded video, described video dimensionality reduction is become pictorial information set；
Step 2, pictorial information set is carried out Activity recognition comparison according to intelligent behavior pattern, if comparison is passed through, send pre-alert notification and preserve video, if comparison is not passed through, preserve video；
Step 3, people's police on duty confirm described pre-alert notification, examine when being implicitly present in alert, are determined the position of described photographic head and the police strength situation near this camera position by GPS locating and tracking, send alert information to neighbouring people's police；
If confirming described pre-alert notification presetting really to warn in the time without people's police on duty, the position of described photographic head is then determined automatically by GPS locating and tracking, and the police strength situation near this camera position, send alert information to neighbouring people's police, send alert information to operator on duty simultaneously.
Further, described face recognition technology is particularly as follows: described face recognition technology is identified as basis with big triangle, in conjunction with one or more the combination in any in little triangle identification, six spacing identifications, cross-ratio identification, obtain face characteristic information with this, it is determined that the uniqueness of personnel；Described big triangle identification refers to and forms a big triangle with left eye centre position, right eye centre position, nose position, obtains three length of sides and big each corner dimension in triangle of big triangle；Described little triangle identification refers to and forms a little triangle hitting exactly depression points under chin high order end, chin low order end, lower lip, obtains each corner dimension in the ratio on little three limits of triangle and little triangle；Described six spacing identifications refer to the spacing and lower jaw branch profiled spaces that obtain between the minimum range of horizontal direction between two eyebrows, the width of forehead, the maximum intermalar distance of face mask and the width of face, nasion spacing, the wing of nose；Described cross-ratio identification refers to: the most left profile point of left face cheekbone, the right lower jaw branch profile point the cross-ratio value of spacing and the right face the rightest profile point of cheekbone, the spacing of left face lower jaw branch profile point.
Further, intelligent behavior pattern in described step 2 is: face catches monitoring mode, the face characteristic information of everyone, particularly as follows: picture each in pictorial information set is obtained the face characteristic information of everyone in picture by described face recognition technology, is then carried out the likelihood ratio pair with the face characteristic information in a face database by described step 2；If similarity exceedes the threshold value set by face acquisition mode, comparison is passed through, and sends pre-alert notification and preserves video, and otherwise, comparison is not passed through, and preserves video.
Further, intelligent behavior pattern in described step 2 is: step 2 described in crowd massing monitoring mode is particularly as follows: obtain the face characteristic information of everyone in picture by picture each in pictorial information set by described face recognition technology, determine the uniqueness of personnel, obtain the number in each picture, thus grasping the mobility scale of personnel amount in this shooting area in real time；Pre-set the higher limit of this shooting area number, if personnel amount exceedes this higher limit in this shooting area, crowd massing is then occurred to transfinite, send pre-alert notification, preserve video, the face characteristic information of everyone in this shooting area and a face database are carried out the likelihood ratio pair simultaneously, judging whether to exist in this shooting area the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal or missing child, if similarity exceedes the threshold value set by crowd massing monitoring mode, comparison is passed through, and again sends pre-alert notification；If personnel amount is not above this higher limit in this shooting area, preserve video.
Further, the intelligent behavior pattern in described step 2 is: crowd massing monitoring mode, and described step 2, particularly as follows: first delimit monitored area in the shooting area of photographic head, arranges upper limit threshold and the monitoring time period of personnel amount in this monitored area；Then, each picture in pictorial information set in this monitoring time period is obtained the face characteristic information in described monitored area by described face recognition technology, determine the uniqueness of personnel in this monitored area, obtain the number in this monitored area, thus grasping the mobility scale of monitored area personnel amount in this monitoring time period in real time；If personnel amount exceedes this upper limit threshold in this monitored area, crowd massing is then occurred to transfinite, send pre-alert notification, preserve video, the face characteristic information of everyone in this monitored area and a face database are carried out the likelihood ratio pair simultaneously, judge whether to exist in this monitored area the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal or missing child, if similarity exceedes the threshold value set by crowd massing monitoring mode, then again send pre-alert notification；If personnel amount is not above this higher limit in this monitored area, preserve video.
nullFurther，Intelligent behavior pattern in described step 2 is: monitoring mode of hovering for a long time，Described step 2 is particularly as follows: arrange threshold value residence time、Path and track determination time，First each picture in described pictorial information set is obtained the face characteristic information of everyone in picture by described face recognition technology，Determine the uniqueness of personnel，Thus everyone is positioned tracking，By the residence time of photographic head record everyone and motion track，Then judge whether everyone's motion track within the track determination time set meets described path，If there are the personnel met，Judge whether the residence time meeting personnel exceedes threshold value residence time again，If exceeding，Then send pre-alert notification、Preserve video，The face characteristic information producing the personnel of pre-alert notification is carried out the likelihood ratio pair with a face database simultaneously，Judge that whether these personnel are the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal or missing child，If similarity exceedes the threshold value hovered for a long time set by monitoring mode，Then again send pre-alert notification；If being absent from meeting the personnel of path, or the personnel meeting path are not above threshold value residence time its residence time, then preserve video.
Further, the path of described setting is " Z " type path or " M " type path or " O " type path.
nullFurther，Intelligent behavior pattern in described step 2 is: frequently come in and go out monitoring mode，First described step 2 particularly as follows: set at least one monitoring period and frequency threshold corresponding to each monitoring period，Then each picture in described pictorial information set is obtained the face characteristic information of everyone in picture by described face recognition technology，Determine the uniqueness of personnel，And everyone is uniquely identified，Record everyone occurs in the number of times of shooting area and the time of discrepancy every time，And calculate the discrepancy frequency of everyone，Then according to the monitoring period that each access time is corresponding，Obtain frequency threshold corresponding to this monitoring period，When the frequency that comes in and goes out exceedes the frequency threshold of correspondence，Send pre-alert notification、Preserve video，The face characteristic information producing the personnel of pre-alert notification is carried out the likelihood ratio pair with a face database simultaneously，Judge that whether these personnel are the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal or missing child，If similarity exceedes the threshold value frequently come in and gone out set by monitoring mode，Then again send pre-alert notification；If being absent from exceeding the personnel of each frequency threshold, then preserve video.
Further, intelligent behavior pattern in described step 2 is: circumference monitoring instruction pattern, first described step 2 particularly as follows: delimit warning region in the shooting area of photographic head, if described pictorial information set occurs mobile target, and this moves target and enters described warning region or when moving in warning region, send pre-alert notification, record this motion track moving target and preserve video, the mobile target producing pre-alert notification is obtained face characteristic information by face recognition technology simultaneously, then the likelihood ratio pair is carried out with a face database, whether judge that this moves target is the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal or missing child, if similarity exceedes the threshold value set by circumference monitoring instruction pattern, then again send pre-alert notification, if being absent from mobile target, then preserve video；Obtain the mobile target of face characteristic information for face recognition technology cannot be passed through, then no longer send pre-alert notification.
nullFurther，Intelligent behavior pattern in described step 2 is: abduct child's monitoring mode，Described step 2 is particularly as follows: described pre-alert notification includes one-level pre-alert notification、Two grades of pre-alert notification and three grades of pre-alert notification，Described one-level pre-alert notification the highest grade，Two grades of pre-alert notification grades are secondly，Three grades of pre-alert notification grades are minimum，First the child age district needing monitoring is set、The change frequency threshold of human face similarity degree threshold value and facial expression，Then each picture in pictorial information set is obtained the face characteristic information of everyone in picture by described face recognition technology，The age size of everyone is judged by face characteristic information，The child's face characteristic information falling into described child age district is carried out similarity identification with a missing child storehouse，Judge whether it is missing child，If similarity exceedes described human face similarity degree threshold value，Then send one-level pre-alert notification，Preserve video simultaneously，Then the face characteristic information of the adjacent personnel of this child is obtained further by described face recognition technology，The face characteristic information of adjacent personnel and a face database are carried out the likelihood ratio pair，Judge whether it is the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal，If similarity exceedes described human face similarity degree threshold value，Then again send one-level pre-alert notification，If being not above described human face similarity degree threshold value，Then again the face characteristic information of these adjacent personnel is carried out the likelihood ratio pair with the human face photo in resident population storehouse and population from other places storehouse，Thus obtaining the identity information of these adjacent personnel；
The child's face characteristic information falling into described child age district is carried out similarity identification with a missing child storehouse, judge whether it is missing child, if similarity is not above described human face similarity degree threshold value, then obtain the distance that this child is adjacent between personnel further, judge the extremity relation of adjacent personnel and this child, if not intimate contact relation, then preserve video；If intimate contact relation, the expression of this child is then judged further according to the frequent degree of this child's face characteristic information change, if the frequent degree of change exceedes described change frequency threshold, sound is gathered by voice collector, this sound and sound bank being compared, having analysed whether sob or threatening vocabulary, if having, then send two grades of pre-alert notification, preserve video；If the frequent degree of change is not above described change frequency threshold, gather sound by voice collector, this sound and sound bank are compared, analysing whether sob or threatening vocabulary, if having, then having sent three grades of pre-alert notification, preserving video, if not having, then not processing；Described voice collector and described photographic head are arranged at same place, and described sound bank includes the sob collected in advance and the threatening vocabulary being configured with language not of the same race.
Further, described face database includes public security and deploys to ensure effective monitoring and control of illegal activities personnel's photo library, the online escaped criminal's information bank in the whole nation and missing child storehouse.
Further, after the video that described photographic head is recorded clearly is processed by video, dimensionality reduction becomes pictorial information set.
Further, the sound that voice collector collects is by comparing with described sound bank after noise filtration treatment.
Further, the alert information in described step 3 comprises the personal information of camera position information and face database comparison and the motion track of similarity situation, the acquired personnel's photographic intelligence producing early warning and these personnel；The mode of the pre-alert notification in described step 3 is: pre-alert notification is pointed out in the way of page ejection prompting frame is in conjunction with alarm sound.
It is an advantage of the current invention that: the present invention passes through photographic head recorded video, video dimensionality reduction is become pictorial information set, demand further according to place, photographic head place, corresponding intelligent behavior pattern is set, purpose is had to monitor thus introducing, carry out intelligent behavior linguistic analysis comparison, realize face and catch monitoring, crowd massing, hover for a long time, frequently come in and go out, cross over circumference, abduct the monitoring effectively in time of the behaviors such as child, send pre-alert notification, thus realizing timely pre-alarm and prevention in advance, then again through artificial really alert, make clear and definite responding to process, prevent cases from worsening further, lock suspected target in time simultaneously, identification comparison again through face database, shorten and solve a case the time, improve case-solving rate, change the tracking afterwards of present stage, transfer the inefficient pattern of multitude of video retrieval investigation.
Accompanying drawing explanation
The present invention is further illustrated in conjunction with the embodiments with reference to the accompanying drawings.
Fig. 1 is face characteristic pattern of the present invention.
Fig. 2 is the schematic flow sheet of the present invention.
Detailed description of the invention
Refer to each circle in Fig. 1 and Fig. 2, Fig. 1 and represent that the characteristic point of face, the numeral in each circle are that characteristic point is numbered, convenient its position of explanation, Fig. 1 marks the position of 98 characteristic points, and has carried out label respectively, respectively label 1～label 98.
Method is monitored in crime based on face recognition technology and behavior speech recognition, described face recognition technology is particularly as follows: described face recognition technology is identified as basis with big triangle, in conjunction with one or more the combination in any in little triangle identification, six spacing identifications, cross-ratio identification, face characteristic information is obtained, it is determined that the uniqueness of personnel with this；Described big triangle identification refers to and forms a big triangle with left eye centre position (referring to label 96 in Fig. 1), right eye centre position (referring to label 97 in Fig. 1), nose position (referring to label 98 in Fig. 1), obtains three length of sides and big each corner dimension in triangle of big triangle；Described little triangle identification refers to and forms a little triangle hitting exactly depression points (referring to label 85 in Fig. 1) under chin high order end (referring to label 9 in Fig. 1), chin low order end (referring to label 11 in Fig. 1), lower lip, obtains each corner dimension in the ratio on little three limits of triangle and little triangle；Described six spacing identifications refer to the minimum range (i.e. the spacing of label 25 and label 30 in Fig. 1) obtaining horizontal direction between two eyebrows, the width (i.e. the spacing of label 1 and label 19 in Fig. 1) of forehead, the width (i.e. the spacing of label 3 and label 17 in Fig. 1) of the maximum intermalar distance of face mask and face, nasion spacing (i.e. the spacing of label 64 and label 75 in Fig. 1), spacing between the wing of nose (i.e. the spacing of label 67 and label 72 in Fig. 1) and lower jaw branch profiled spaces (i.e. the spacing of label 7 and label 13 in Fig. 1)；Described cross-ratio identification refers to the cross-ratio value of spacing and the right face the rightest profile point of cheekbone (label 17 referring in Fig. 1), the spacing of left face lower jaw branch profile point (label 7 referring in Fig. 1) of the most left profile point (label 3 referring in Fig. 1) of left face cheekbone, the right lower jaw branch profile point (label 13 referring in Fig. 1).Described face characteristic information includes: spacing between the minimum range of horizontal direction, the width of forehead, the maximum intermalar distance of face mask, nasion spacing, the wing of nose, lower jaw branch profiled spaces and cross-ratio value between each corner dimension, two eyebrows in each corner dimension in three length of sides of big triangle and big triangle, the ratio on three limits of little triangle and little triangle；
Described crime monitoring method comprises the steps:
Step 1, by photographic head recorded video, described video dimensionality reduction is become pictorial information set；In the present embodiment, after the video that described photographic head is recorded clearly is processed by video, dimensionality reduction becomes pictorial information set
Step 2, pictorial information set is carried out Activity recognition comparison according to intelligent behavior pattern, if comparison is passed through, send pre-alert notification and preserve video, if comparison is not passed through, preserve video；
Step 3, people's police on duty confirm described pre-alert notification, examine when being implicitly present in alert, are determined the position of described photographic head and the police strength situation near this camera position by GPS locating and tracking, send alert information to neighbouring people's police；
If confirming described pre-alert notification presetting really to warn in the time without people's police on duty, the position of described photographic head is then determined automatically by GPS locating and tracking, and the police strength situation near this camera position, send alert information to neighbouring people's police, send alert information to operator on duty simultaneously.In the present embodiment, described alert information comprises the personal information of camera position information and face database comparison and the motion track of similarity situation, the acquired personnel's photographic intelligence producing early warning and these personnel；The mode of described pre-alert notification is: pre-alert notification is pointed out in the way of page ejection prompting frame is in conjunction with alarm sound.
In the present embodiment, six kinds of intelligent behavior patterns of concrete statement, according to the place demand that photographic head is arranged, the intelligent behavior pattern adopted under this place is set accordingly.
Pattern one: face catches monitoring mode
The face characteristic information of everyone, particularly as follows: picture each in pictorial information set is obtained the face characteristic information of everyone in picture by described face recognition technology, is then carried out the likelihood ratio pair with the face characteristic information in a face database by described step 2；If similarity exceedes the threshold value set by face acquisition mode, comparison is passed through, and sends pre-alert notification and preserves video, and otherwise, comparison is not passed through, and preserves video.Described similarity, for big triangle identification technology, namely in three length of sides of the big triangle of a face and big triangle each corner dimension respectively accordingly with the similarity degree of each corner dimension in three length of sides of the big triangle of another face and big triangle.Described face database includes public security and deploys to ensure effective monitoring and control of illegal activities personnel's photo library, the online escaped criminal's information bank in the whole nation and missing child storehouse.
Described face is caught monitoring mode and be can be applicable to airport bayonet socket or gateway, railway station or gateway, bus station or gateway, megastore, catch face characteristic information, then comparison in real time, provides hint information effective, timely, reliable for arresting the personnel of deploying to ensure effective monitoring and control of illegal activities and fugitive personnel and rescue missing child.
Pattern two: crowd massing monitoring mode
Described step 2 particularly as follows:
Picture each in pictorial information set is obtained the face characteristic information of everyone in picture by described face recognition technology, it is determined that the uniqueness of personnel, it is thus achieved that the number in each picture, thus grasping the mobility scale of personnel amount in this shooting area in real time；Pre-set the higher limit of this shooting area number, if personnel amount exceedes this higher limit in this shooting area, crowd massing is then occurred to transfinite, send pre-alert notification, preserve video, the face characteristic information of everyone in this shooting area and described face database are carried out the likelihood ratio pair simultaneously, judging whether to exist in this shooting area the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal or missing child, if similarity exceedes the threshold value set by crowd massing monitoring mode, comparison is passed through, and again sends pre-alert notification；If personnel amount is not above this higher limit in this shooting area, preserve video.By the identification comparison with face database, a suspect is analyzed and processed in time, the time of solving a case can be shortened, improve case-solving rate, change the tracking afterwards of present stage, transfer the inefficient pattern of multitude of video retrieval investigation.
Another preferred embodiment of described crowd massing monitoring mode is: first delimit monitored area in the shooting area of photographic head, arranges upper limit threshold and the monitoring time period of personnel amount in this monitored area；Then, each picture in pictorial information set in this monitoring time period is obtained the face characteristic information in described monitored area by described face recognition technology, determine the uniqueness of personnel in this monitored area, obtain the number in this monitored area, thus grasping the mobility scale of monitored area personnel amount in this monitoring time period in real time；If personnel amount exceedes this upper limit threshold in this monitored area, crowd massing is then occurred to transfinite, send pre-alert notification, preserve video, the face characteristic information of everyone in this monitored area and described face database are carried out the likelihood ratio pair simultaneously, judge whether to exist in this monitored area the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal or missing child, if similarity exceedes the threshold value set by crowd massing monitoring mode, then again send pre-alert notification；If personnel amount is not above this higher limit in this monitored area, preserve video.By delimiting monitored area so that monitoring position is more accurate, and the pre-alert notification accuracy rate sent is higher.
Crowd massing monitoring mode can be applicable to before square or government bodies building or bank doorway, and give warning in advance illegal gathering that may be present or violent behavior of assembling a crowd.
Pattern three, hover monitoring mode for a long time
nullDescribed step 2 is particularly as follows: arrange threshold value residence time、Path and track determination time，First each picture in described pictorial information set is obtained the face characteristic information of everyone in picture by described face recognition technology，Determine the uniqueness of personnel，Thus everyone is positioned tracking，By the residence time of photographic head record everyone and motion track，Then judge whether everyone's motion track within the track determination time set meets described path，If there are the personnel met，Judge whether the residence time meeting personnel exceedes threshold value residence time again，If exceeding，Then send pre-alert notification、Preserve video，The face characteristic information producing the personnel of pre-alert notification is carried out the likelihood ratio pair with a face database simultaneously，Judge that whether these personnel are the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal or missing child，If similarity exceedes the threshold value hovered for a long time set by monitoring mode，Then again send pre-alert notification；If being absent from meeting the personnel of path, or the personnel meeting path are not above threshold value residence time its residence time, then preserve video.Described track determination time refers to: the track arranging in a period of time is analyzed, for instance, it is set to one minute, then one minute interior motion track of personnel is compared with described path.The path of described setting may is that " Z " type path or " M " type path, can set various path according to the behavior characteristics analyzing crime personnel herein.
Monitoring mode of hovering for a long time can be applicable to bank ATM and withdraws the money district, gives warning in advance and would be likely to occur crime behavior.
Pattern four, frequently come in and go out monitoring mode
nullFirst described step 2 particularly as follows: set at least one monitoring period and frequency threshold corresponding to each monitoring period，Then each picture in described pictorial information set is obtained the face characteristic information of everyone in picture by described face recognition technology，Determine the uniqueness of personnel，And everyone is uniquely identified，Record everyone occurs in the number of times of shooting area and the time of discrepancy every time，And calculate the discrepancy frequency of everyone，Then according to the monitoring period that each access time is corresponding，Obtain frequency threshold corresponding to this monitoring period，When the frequency that comes in and goes out exceedes the frequency threshold of correspondence，Send pre-alert notification、Preserve video，The face characteristic information producing the personnel of pre-alert notification is carried out the likelihood ratio pair with a face database simultaneously，Judge that whether these personnel are the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal or missing child，If similarity exceedes the threshold value frequently come in and gone out set by monitoring mode，Then again send pre-alert notification；If being absent from exceeding the personnel of each frequency threshold, then preserve video.
Described frequent discrepancy monitoring mode can be applicable to hotel's front door, bank gateway, government bodies gate etc..
Pattern five, circumference monitoring instruction pattern
First described step 2 particularly as follows: delimit warning region in the shooting area of photographic head, if described pictorial information set occurs mobile target, and this moves target and enters described warning region or when moving in warning region, send pre-alert notification, record this motion track moving target and preserve video, the mobile target producing pre-alert notification is obtained face characteristic information by face recognition technology simultaneously, then the likelihood ratio pair is carried out with a face database, whether judge that this moves target is the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal or missing child, if similarity exceedes the threshold value set by circumference monitoring instruction pattern, then again send pre-alert notification, if being absent from mobile target, then preserve video；Obtain the mobile target of face characteristic information for face recognition technology cannot be passed through, then no longer send pre-alert notification.
Described circumference monitoring instruction can be applicable to the important area in the internal important area of bank or government bodies building.
Pattern six, abduct child's monitoring mode
nullDescribed step 2 is particularly as follows: described pre-alert notification includes one-level pre-alert notification、Two grades of pre-alert notification and three grades of pre-alert notification，Described one-level pre-alert notification the highest grade，Two grades of pre-alert notification grades are secondly，Three grades of pre-alert notification grades are minimum，First the child age district needing monitoring is set、The change frequency threshold of human face similarity degree threshold value and facial expression，Then each picture in pictorial information set is obtained the face characteristic information of everyone in picture by described face recognition technology，The age size of everyone is judged by face characteristic information，The child's face characteristic information falling into described child age district is carried out similarity identification with a missing child storehouse，Judge whether it is missing child，If similarity exceedes described human face similarity degree threshold value，Then send one-level pre-alert notification，Preserve video simultaneously，Then the face characteristic information of the adjacent personnel of this child is obtained further by described face recognition technology，The face characteristic information of adjacent personnel and a face database are carried out the likelihood ratio pair，Judge whether it is the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal，If similarity exceedes described human face similarity degree threshold value，Then again send one-level pre-alert notification，If being not above described human face similarity degree threshold value，Then again the face characteristic information of these adjacent personnel is carried out the likelihood ratio pair with the human face photo in resident population storehouse and population from other places storehouse，Thus obtaining the identity information of these adjacent personnel；
The child's face characteristic information falling into described child age district is carried out similarity identification with a missing child storehouse, judge whether it is missing child, if similarity is not above described human face similarity degree threshold value, then obtain the distance that this child is adjacent between personnel further, judge the extremity relation of adjacent personnel and this child, if not intimate contact relation, then preserve video；If intimate contact relation, the expression of this child is then judged further according to the frequent degree of this child's face characteristic information change, if the frequent degree of change exceedes described change frequency threshold, sound is gathered by voice collector, the sound that voice collector collects, by comparing with sound bank after noise filtration treatment, has analysed whether sob or threatening vocabulary, if having, then send two grades of pre-alert notification, preserve video；If the frequent degree of change is not above described change frequency threshold, sound is gathered by voice collector, the sound that voice collector collects is by comparing with sound bank after noise filtration treatment, sob or threatening vocabulary are analysed whether, if having, then sending three grades of pre-alert notification, preserving video, if no, then not processing；Described voice collector and described photographic head are arranged at same place, and described sound bank includes the sob collected in advance and the threatening vocabulary being configured with language not of the same race.It is adjacent the whether intimate contact of the Distance Judgment between personnel according to child, can pass through to pre-set a distance and intimacy synopsis, then be made to determine whether intimate contact according to this table.
It is an advantage of the current invention that: the present invention passes through photographic head recorded video, video dimensionality reduction is become pictorial information set, demand further according to place, photographic head place, corresponding intelligent behavior pattern is set, purpose is had to monitor thus introducing, carry out intelligent behavior linguistic analysis comparison, realize face and catch monitoring, crowd massing, hover for a long time, frequently come in and go out, cross over circumference, abduct the monitoring effectively in time of the behaviors such as child, send pre-alert notification, thus realizing timely pre-alarm and prevention in advance, then again through artificial really alert, make clear and definite responding to process, prevent cases from worsening further, lock suspected target in time simultaneously, identification comparison again through face database, shorten and solve a case the time, improve case-solving rate, change the tracking afterwards of present stage, transfer the inefficient pattern of multitude of video retrieval investigation.
Although the foregoing describing the specific embodiment of the present invention; but those familiar with the art is to be understood that; we are merely exemplary described specific embodiment; rather than for the restriction to the scope of the present invention; those of ordinary skill in the art, in the equivalent modification made according to the spirit of the present invention and change, should be encompassed in the scope of the claimed protection of the present invention.
Claims (13)
1. monitor method based on the crime of face recognition technology and behavior speech recognition, it is characterised in that: comprise the steps:
Step 1, by photographic head recorded video, described video dimensionality reduction is become pictorial information set；
Step 2, pictorial information set is carried out Activity recognition comparison according to intelligent behavior pattern, if comparison is passed through, send pre-alert notification and preserve video, if comparison is not passed through, preserve video；Described intelligent behavior pattern include face catch monitoring mode, crowd massing monitoring mode, for a long time hover monitoring mode, frequently come in and go out monitoring mode, circumference monitoring instruction pattern and abduct child's monitoring mode；
Step 3, people's police on duty confirm described pre-alert notification, examine when being implicitly present in alert, are determined the position of described photographic head and the police strength situation near this camera position by GPS locating and tracking, send alert information to neighbouring people's police；
If confirming described pre-alert notification presetting really to warn in the time without people's police on duty, the position of described photographic head is then determined automatically by GPS locating and tracking, and the police strength situation near this camera position, send alert information to neighbouring people's police, send alert information to operator on duty simultaneously；
Described face recognition technology is particularly as follows: described face recognition technology is identified as basis with big triangle, in conjunction with one or more the combination in any in little triangle identification, six spacing identifications, cross-ratio identification, face characteristic information is obtained, it is determined that the uniqueness of personnel with this；Described big triangle identification refers to and forms a big triangle with left eye centre position, right eye centre position, nose position, obtains three length of sides and big each corner dimension in triangle of big triangle；Described little triangle identification refers to and forms a little triangle hitting exactly depression points under chin high order end, chin low order end, lower lip, obtains each corner dimension in the ratio on little three limits of triangle and little triangle；Described six spacing identifications refer to the spacing and lower jaw branch profiled spaces that obtain between the minimum range of horizontal direction between two eyebrows, the width of forehead, the maximum intermalar distance of face mask and the width of face, nasion spacing, the wing of nose；Described cross-ratio identification refers to: the most left profile point of left face cheekbone, the right lower jaw branch profile point the cross-ratio value of spacing and the right face the rightest profile point of cheekbone, the spacing of left face lower jaw branch profile point.
2. method is monitored in the crime based on face recognition technology and behavior speech recognition according to claim 1, it is characterized in that: the intelligent behavior pattern in described step 2 is: face catches monitoring mode, the face characteristic information of everyone, particularly as follows: picture each in pictorial information set is obtained the face characteristic information of everyone in picture by described face recognition technology, is then carried out the likelihood ratio pair with the face characteristic information in a face database by described step 2；If similarity exceedes the threshold value set by face acquisition mode, comparison is passed through, and sends pre-alert notification and preserves video, and otherwise, comparison is not passed through, and preserves video.
3. method is monitored in the crime based on face recognition technology and behavior speech recognition according to claim 1, it is characterized in that: the intelligent behavior pattern in described step 2 is: crowd massing monitoring mode, described step 2 is particularly as follows: obtain the face characteristic information of everyone in picture by picture each in pictorial information set by described face recognition technology, determine the uniqueness of personnel, obtain the number in each picture, thus grasping the mobility scale of personnel amount in shooting area in real time；Pre-set the higher limit of this shooting area number, if personnel amount exceedes this higher limit in this shooting area, crowd massing is then occurred to transfinite, send pre-alert notification, preserve video, the face characteristic information of everyone in this shooting area and a face database are carried out the likelihood ratio pair simultaneously, judging whether to exist in this shooting area the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal or missing child, if similarity exceedes the threshold value set by crowd massing monitoring mode, comparison is passed through, and again sends pre-alert notification；If personnel amount is not above this higher limit in this shooting area, preserve video.
4. method is monitored in the crime based on face recognition technology and behavior speech recognition according to claim 1, it is characterized in that: the intelligent behavior pattern in described step 2 is: crowd massing monitoring mode, described step 2, particularly as follows: first delimit monitored area in the shooting area of photographic head, arranges upper limit threshold and the monitoring time period of personnel amount in this monitored area；Then, each picture in pictorial information set in this monitoring time period is obtained the face characteristic information in described monitored area by described face recognition technology, determine the uniqueness of personnel in this monitored area, obtain the number in this monitored area, thus grasping the mobility scale of monitored area personnel amount in this monitoring time period in real time；If personnel amount exceedes this upper limit threshold in this monitored area, crowd massing is then occurred to transfinite, send pre-alert notification, preserve video, the face characteristic information of everyone in this monitored area and a face database are carried out the likelihood ratio pair simultaneously, judge whether to exist in this monitored area the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal or missing child, if similarity exceedes the threshold value set by crowd massing monitoring mode, then again send pre-alert notification；If personnel amount is not above this higher limit in this monitored area, preserve video.
null5. method is monitored in the crime based on face recognition technology and behavior speech recognition according to claim 1，It is characterized in that: the intelligent behavior pattern in described step 2 is: monitoring mode of hovering for a long time，Described step 2 is particularly as follows: arrange threshold value residence time、Path and track determination time，First each picture in described pictorial information set is obtained the face characteristic information of everyone in picture by described face recognition technology，Determine the uniqueness of personnel，Thus everyone is positioned tracking，By the residence time of photographic head record everyone and motion track，Then judge whether everyone's motion track within the track determination time set meets described path，If there are the personnel met，Judge whether the residence time meeting personnel exceedes threshold value residence time again，If exceeding，Then send pre-alert notification、Preserve video，The face characteristic information producing the personnel of pre-alert notification is carried out the likelihood ratio pair with a face database simultaneously，Judge that whether these personnel are the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal or missing child，If similarity exceedes the threshold value hovered for a long time set by monitoring mode，Then again send pre-alert notification；If being absent from meeting the personnel of path, or the personnel meeting path are not above threshold value residence time its residence time, then preserve video.
6. method is monitored in the crime based on face recognition technology and behavior speech recognition according to claim 5, it is characterised in that: the path of described setting is " Z " type path or " M " type path or " O " type path.
null7. method is monitored in the crime based on face recognition technology and behavior speech recognition according to claim 1，It is characterized in that: the intelligent behavior pattern in described step 2 is: frequently come in and go out monitoring mode，First described step 2 particularly as follows: set at least one monitoring period and frequency threshold corresponding to each monitoring period，Then each picture in described pictorial information set is obtained the face characteristic information of everyone in picture by described face recognition technology，Determine the uniqueness of personnel，And everyone is uniquely identified，Record everyone occurs in the number of times of shooting area and the time of discrepancy every time，And calculate the discrepancy frequency of everyone，Then according to the monitoring period that each access time is corresponding，Obtain frequency threshold corresponding to this monitoring period，When the frequency that comes in and goes out exceedes the frequency threshold of correspondence，Send pre-alert notification、Preserve video，The face characteristic information producing the personnel of pre-alert notification is carried out the likelihood ratio pair with a face database simultaneously，Judge that whether these personnel are the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal or missing child，If similarity exceedes the threshold value frequently come in and gone out set by monitoring mode，Then again send pre-alert notification；If being absent from exceeding the personnel of each frequency threshold, then preserve video.
null8. method is monitored in the crime based on face recognition technology and behavior speech recognition according to claim 1，It is characterized in that: the intelligent behavior pattern in described step 2 is: circumference monitoring instruction pattern，First described step 2 particularly as follows: delimit warning region in the shooting area of photographic head，If described pictorial information set occurs mobile target，And this moves target and enters described warning region or when moving in warning region，Send pre-alert notification、Record this motion track moving target and preserve video，The mobile target producing pre-alert notification is obtained face characteristic information by face recognition technology simultaneously，Then the likelihood ratio pair is carried out with a face database，Whether judge that this moves target is the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal or missing child，If similarity exceedes the threshold value set by circumference monitoring instruction pattern，Then again send pre-alert notification，If being absent from mobile target，Then preserve video；Obtain the mobile target of face characteristic information for face recognition technology cannot be passed through, then no longer send pre-alert notification.
null9. method is monitored in the crime based on face recognition technology and behavior speech recognition according to claim 1，It is characterized in that: the intelligent behavior pattern in described step 2 is: abduct child's monitoring mode，Described step 2 is particularly as follows: described pre-alert notification includes one-level pre-alert notification、Two grades of pre-alert notification and three grades of pre-alert notification，Described one-level pre-alert notification the highest grade，Two grades of pre-alert notification grades are secondly，Three grades of pre-alert notification grades are minimum，First the child age district needing monitoring is set、The change frequency threshold of human face similarity degree threshold value and facial expression，Then each picture in pictorial information set is obtained the face characteristic information of everyone in picture by described face recognition technology，The age size of everyone is judged by face characteristic information，The child's face characteristic information falling into described child age district is carried out similarity identification with a missing child storehouse，Judge whether it is missing child，If similarity exceedes described human face similarity degree threshold value，Then send one-level pre-alert notification，Preserve video simultaneously，Then the face characteristic information of the adjacent personnel of this child is obtained further by described face recognition technology，The face characteristic information of adjacent personnel and a face database are carried out the likelihood ratio pair，Judge whether it is the personnel of deploying to ensure effective monitoring and control of illegal activities or escaped criminal，If similarity exceedes described human face similarity degree threshold value，Then again send one-level pre-alert notification，If being not above described human face similarity degree threshold value，Then again the face characteristic information of these adjacent personnel is carried out the likelihood ratio pair with the human face photo in resident population storehouse and population from other places storehouse，Thus obtaining the identity information of these adjacent personnel；
The child's face characteristic information falling into described child age district is carried out similarity identification with a missing child storehouse, judge whether it is missing child, if similarity is not above described human face similarity degree threshold value, then obtain the distance that this child is adjacent between personnel further, judge the extremity relation of adjacent personnel and this child, if not intimate contact relation, then preserve video；If intimate contact relation, the expression of this child is then judged further according to the frequent degree of this child's face characteristic information change, if the frequent degree of change exceedes described change frequency threshold, sound is gathered by voice collector, this sound and sound bank being compared, having analysed whether sob or threatening vocabulary, if having, then send two grades of pre-alert notification, preserve video；If the frequent degree of change is not above described change frequency threshold, gather sound by voice collector, this sound and sound bank are compared, analysing whether sob or threatening vocabulary, if having, then having sent three grades of pre-alert notification, preserving video, if not having, then not processing；Described voice collector and described photographic head are arranged at same place, and described sound bank includes the sob collected in advance and the threatening vocabulary being configured with language not of the same race.
10. method is monitored in the crime based on face recognition technology and behavior speech recognition according to any one of claim 2 to 9, it is characterised in that: described face database includes public security and deploys to ensure effective monitoring and control of illegal activities personnel's photo library, the online escaped criminal's information bank in the whole nation and missing child storehouse.
11. method is monitored in the crime based on face recognition technology and behavior speech recognition according to claim 1, it is characterised in that: after the video that described photographic head is recorded clearly is processed by video, dimensionality reduction becomes pictorial information set.
12. method is monitored in the crime based on face recognition technology and behavior speech recognition according to claim 9, it is characterised in that: the sound that voice collector collects is by comparing with described sound bank after noise filtration treatment.
13. method is monitored in the crime based on face recognition technology and behavior speech recognition according to claim 1, it is characterised in that: the alert information in described step 3 comprises the personal information of camera position information and face database comparison and the motion track of similarity situation, the acquired personnel's photographic intelligence producing early warning and these personnel；The mode of the pre-alert notification in described step 3 is: pre-alert notification is pointed out in the way of page ejection prompting frame is in conjunction with alarm sound.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310139660.3A CN103246869B (en) | 2013-04-19 | 2013-04-19 | Method is monitored in crime based on recognition of face and behavior speech recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310139660.3A CN103246869B (en) | 2013-04-19 | 2013-04-19 | Method is monitored in crime based on recognition of face and behavior speech recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103246869A CN103246869A (en) | 2013-08-14 |
| CN103246869B true CN103246869B (en) | 2016-07-06 |
Family
ID=48926382
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310139660.3A Active CN103246869B (en) | 2013-04-19 | 2013-04-19 | Method is monitored in crime based on recognition of face and behavior speech recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103246869B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108364438A (en) * | 2017-12-28 | 2018-08-03 | 武汉烽火众智数字技术有限责任公司 | Method for early warning based on front end perception data and device |
Families Citing this family (86)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103530967A (en) * | 2013-11-03 | 2014-01-22 | 昆山志众欣诚信息技术有限公司 | Security alarm system for residential area |
| CN104866482A (en) * | 2014-02-21 | 2015-08-26 | 支录奎 | Wide area dynamic personnel real-time trace positioning information overlaying fuzzy quick inquiry system |
| CN103824064A (en) * | 2014-03-11 | 2014-05-28 | 深圳市中安视科技有限公司 | Huge-amount human face discovering and recognizing method |
| EP3121799A4 (en) * | 2014-03-19 | 2018-02-07 | Technomirai Co., Ltd | Digital loss-defense security system, method, and program |
| CN104980558B (en) * | 2014-04-01 | 2019-01-01 | 中国移动通信集团公司 | A kind of method, apparatus of information processing, mobile phone pet client and mobile terminal |
| EP3131076B1 (en) * | 2014-04-09 | 2018-12-19 | Technomirai Co., Ltd | Digital loss/accident defense system, method, and program |
| JP6597609B2 (en) | 2014-06-30 | 2019-10-30 | 日本電気株式会社 | Image processing apparatus, monitoring system, image processing method, and program |
| CN104135645A (en) * | 2014-07-31 | 2014-11-05 | 天津市亚安科技股份有限公司 | Video surveillance system and method for face tracking and capturing |
| CN104598021A (en) * | 2014-12-16 | 2015-05-06 | 天津三星电子有限公司 | Display equipment and display method |
| CN106033530A (en) * | 2015-03-11 | 2016-10-19 | 中兴通讯股份有限公司 | Identification method and apparatus for suspected person |
| CN107944434A (en) * | 2015-06-11 | 2018-04-20 | 广东欧珀移动通信有限公司 | A kind of alarm method and terminal based on rotating camera |
| CN105118228A (en) * | 2015-07-11 | 2015-12-02 | 深圳市前海安测信息技术有限公司 | Automatic alarm system and method used for monitoring children safety |
| CN105243780A (en) * | 2015-09-11 | 2016-01-13 | 中山大学 | Child safety monitoring method and system |
| CN105239897B (en) * | 2015-11-02 | 2018-02-09 | 张玥桐 | A kind of intelligent safety door |
| CN106960172A (en) * | 2016-01-08 | 2017-07-18 | 中兴通讯股份有限公司 | Personal identification processing method, apparatus and system |
| CN106131487A (en) * | 2016-07-12 | 2016-11-16 | 湖北誉恒科技有限公司 | Video monitoring system for ATM nacelle |
| KR101810190B1 (en) * | 2016-07-14 | 2017-12-18 | 김용상 | User authentication method and apparatus using face identification |
| CN107689021A (en) * | 2016-08-04 | 2018-02-13 | 张大海 | Intelligent criminal investigation net scheme |
| CN106341661B (en) * | 2016-09-13 | 2023-04-07 | 深圳市大道智创科技有限公司 | Patrol robot |
| CN106507037A (en) * | 2016-09-30 | 2017-03-15 | 北京中星微电子有限公司 | Intelligent control method and system that personnel invade/cross the border |
| CN108073577A (en) * | 2016-11-08 | 2018-05-25 | 中国电信股份有限公司 | A kind of alarm method and system based on recognition of face |
| WO2018119599A1 (en) * | 2016-12-26 | 2018-07-05 | 深圳前海达闼云端智能科技有限公司 | Method and device for searching for person and communication system |
| CN106971142B (en) * | 2017-02-07 | 2018-07-17 | 深圳云天励飞技术有限公司 | A kind of image processing method and device |
| CN106803934A (en) * | 2017-02-22 | 2017-06-06 | 王培博 | Ways for education monitoring system in a kind of children's safety based on monitor video |
| CN107016361A (en) * | 2017-03-29 | 2017-08-04 | 成都三零凯天通信实业有限公司 | Recognition methods and device based on video analysis |
| CN107016852B (en) * | 2017-04-25 | 2023-01-24 | 上海亦源智能科技有限公司 | Intelligent parking access control system and method with stress prevention function |
| CN107346371A (en) * | 2017-06-06 | 2017-11-14 | 北京云知声信息技术有限公司 | A kind of method for warming and device |
| CN107666652A (en) * | 2017-07-11 | 2018-02-06 | 珠海智加物联网络有限公司 | One kind is taken pictures localization method and system |
| CN107277459A (en) * | 2017-07-29 | 2017-10-20 | 安徽博威康信息技术有限公司 | A kind of camera views switching method recognized based on characteristics of human body with target following |
| CN107833033A (en) * | 2017-11-01 | 2018-03-23 | 深圳位置网科技有限公司 | A kind of method that registration card transaction illegal activities are docked with Skynet system |
| CN108537711A (en) * | 2018-03-27 | 2018-09-14 | 邱伟 | A kind of green road safety management system |
| CN108629322A (en) * | 2018-05-11 | 2018-10-09 | 中国联合网络通信集团有限公司 | Monitoring method, device and the computer readable storage medium of a suspect |
| CN108961495A (en) * | 2018-06-25 | 2018-12-07 | 北京蜂盒科技有限公司 | A kind of data-updating method and device based on recognition of face |
| CN109033973B (en) * | 2018-06-26 | 2020-11-03 | 北京旷视科技有限公司 | Monitoring and early warning method and device and electronic equipment |
| CN108806155B (en) * | 2018-06-29 | 2020-02-11 | 仁怀市云侠网络科技有限公司 | Security monitoring method based on Internet of things |
| CN110795963A (en) * | 2018-08-01 | 2020-02-14 | 深圳云天励飞技术有限公司 | Monitoring method, device and equipment based on face recognition |
| CN110795971B (en) * | 2018-08-02 | 2023-02-17 | 深圳云天励飞技术有限公司 | User behavior identification method, device, equipment and computer storage medium |
| CN109118626B (en) * | 2018-08-08 | 2022-09-13 | 腾讯科技（深圳）有限公司 | Lock control method and device, storage medium and electronic device |
| CN109040278B (en) * | 2018-08-20 | 2020-05-29 | 山东润一智能科技有限公司 | Hospital electrical and power system safety intelligent management cloud platform, method and system |
| CN109257417A (en) * | 2018-08-20 | 2019-01-22 | 山东润智能科技有限公司 | Boiler in hospital security stereo monitors cloud platform, system and method |
| CN109326073A (en) * | 2018-09-04 | 2019-02-12 | 重庆工业职业技术学院 | A kind of wireless sensor network safety defense monitoring system based on multipoint positioning |
| CN109672853B (en) * | 2018-09-25 | 2022-05-17 | 深圳壹账通智能科技有限公司 | Early warning method, device and equipment based on video monitoring and computer storage medium |
| CN109446936A (en) * | 2018-10-12 | 2019-03-08 | 银河水滴科技（北京）有限公司 | A kind of personal identification method and device for monitoring scene |
| CN109508645A (en) * | 2018-10-19 | 2019-03-22 | 银河水滴科技（北京）有限公司 | Personal identification method and device under monitoring scene |
| CN109190601A (en) * | 2018-10-19 | 2019-01-11 | 银河水滴科技（北京）有限公司 | Recongnition of objects method and device under a kind of monitoring scene |
| CN109543538A (en) * | 2018-10-23 | 2019-03-29 | 深圳壹账通智能科技有限公司 | Obtain method, apparatus, computer equipment and the storage medium of the track of alert object |
| CN109194931A (en) * | 2018-10-25 | 2019-01-11 | 哈尔滨海能达科技有限公司 | A kind of notification method and device of the target object based on video pictures |
| WO2020082293A1 (en) * | 2018-10-25 | 2020-04-30 | 哈尔滨海能达科技有限公司 | Method and device for notifying target object based on video image |
| CN109376672A (en) * | 2018-10-31 | 2019-02-22 | 银河水滴科技（北京）有限公司 | Target object localization method and device based on living things feature recognition |
| CN109359620A (en) * | 2018-10-31 | 2019-02-19 | 银河水滴科技（北京）有限公司 | A kind of method and device identifying suspect object |
| CN109389097A (en) * | 2018-10-31 | 2019-02-26 | 银河水滴科技（北京）有限公司 | A kind of recognition methods and identification device |
| CN109492604A (en) * | 2018-11-23 | 2019-03-19 | 北京嘉华科盈信息系统有限公司 | Faceform's characteristic statistics analysis system |
| CN109543638A (en) * | 2018-11-29 | 2019-03-29 | 与德科技有限公司 | A kind of face identification method, device, equipment and storage medium |
| CN109784177A (en) * | 2018-12-14 | 2019-05-21 | 深圳壹账通智能科技有限公司 | Missing crew's method for rapidly positioning, device and medium based on images match |
| CN109743541B (en) * | 2018-12-15 | 2023-04-18 | 深圳壹账通智能科技有限公司 | Intelligent monitoring method and device, computer equipment and storage medium |
| CN109858365B (en) * | 2018-12-28 | 2021-03-05 | 深圳云天励飞技术有限公司 | Special crowd gathering behavior analysis method and device and electronic equipment |
| CN109784274B (en) * | 2018-12-29 | 2021-09-14 | 杭州励飞软件技术有限公司 | Method for identifying trailing and related product |
| CN109815839B (en) * | 2018-12-29 | 2021-10-08 | 深圳云天励飞技术有限公司 | Loitering person identification method under micro-service architecture and related product |
| CN109872482A (en) * | 2019-01-21 | 2019-06-11 | 广东鑫诺安保安服务有限公司 | Wisdom security protection monitoring and managing method, system and storage medium |
| CN109902575B (en) * | 2019-01-24 | 2024-03-15 | 平安科技（深圳）有限公司 | Anti-walking method and device based on unmanned vehicle and related equipment |
| CN109886196A (en) * | 2019-02-21 | 2019-06-14 | 中水北方勘测设计研究有限责任公司 | Personnel track traceability system and method based on BIM plus GIS video monitoring |
| CN109961038A (en) * | 2019-03-20 | 2019-07-02 | 苏州浪潮智能科技有限公司 | A kind of children's safety monitoring method and system based on computerized algorithm |
| CN110046606B (en) * | 2019-04-26 | 2020-08-18 | 上海云从企业发展有限公司 | Early warning method and device |
| CN110113515B (en) * | 2019-05-13 | 2021-01-12 | Oppo广东移动通信有限公司 | Photographing control method and related product |
| CN110298254A (en) * | 2019-05-30 | 2019-10-01 | 罗普特科技集团股份有限公司 | A kind of analysis method and system for personnel's abnormal behaviour |
| CN110336976A (en) * | 2019-06-13 | 2019-10-15 | 长江大学 | A kind of intelligent monitoring probe and system |
| CN110176117B (en) * | 2019-06-17 | 2023-05-19 | 广东翔翼科技信息有限公司 | Monitoring device and monitoring method based on behavior recognition technology |
| CN110245630B (en) * | 2019-06-18 | 2023-02-17 | 广东中安金狮科创有限公司 | Monitoring data processing method and device and readable storage medium |
| CN110334120A (en) * | 2019-06-28 | 2019-10-15 | 深圳市商汤科技有限公司 | Records application method and device, storage medium |
| CN110555964B (en) * | 2019-07-25 | 2021-11-09 | 武汉烽火众智数字技术有限责任公司 | Multi-data fusion key area early warning system and method |
| CN110867041B (en) * | 2019-10-31 | 2023-07-18 | 上海水业设计工程有限公司 | System and method for analyzing crime of stolen non-motor vehicle based on video analysis |
| CN112773262A (en) * | 2019-11-08 | 2021-05-11 | 珠海市一微半导体有限公司 | Security control method based on sweeping robot, sweeping robot and chip |
| CN110991289B (en) * | 2019-11-25 | 2023-09-05 | 达闼机器人股份有限公司 | Abnormal event monitoring method and device, electronic equipment and storage medium |
| CN111179520B (en) * | 2019-12-06 | 2021-08-31 | 广州烽火众智数字技术有限公司 | Telecommunication fraud early warning method, system and storage medium |
| CN111291216B (en) * | 2020-02-28 | 2022-06-14 | 罗普特科技集团股份有限公司 | Method and system for analyzing foothold based on face structured data |
| CN111212274A (en) * | 2020-03-03 | 2020-05-29 | 江苏中数智慧信息科技有限公司 | Intelligent building security system |
| CN111460907B (en) * | 2020-03-05 | 2023-06-20 | 浙江大华技术股份有限公司 | Malicious behavior identification method, system and storage medium |
| CN111402532A (en) * | 2020-03-26 | 2020-07-10 | 海南鸿达盛创网络信息科技有限公司 | Comprehensive security video management control system |
| CN111654665A (en) * | 2020-05-15 | 2020-09-11 | 合肥极光科技股份有限公司 | Intelligent tail money cabinet monitoring system |
| CN111461581A (en) * | 2020-05-17 | 2020-07-28 | 商志营 | Intelligent early warning management system and implementation method |
| CN111645632B (en) * | 2020-06-14 | 2021-06-25 | 广东传奇机动车服务有限公司 | Automobile anti-theft and positioning tracking system based on intelligent identification and big data |
| CN111883130A (en) * | 2020-08-03 | 2020-11-03 | 上海茂声智能科技有限公司 | Fusion type voice recognition method, device, system, equipment and storage medium |
| CN112784712B (en) * | 2021-01-08 | 2023-08-18 | 重庆创通联智物联网有限公司 | Missing child early warning implementation method and device based on real-time monitoring |
| CN112597976B (en) * | 2021-03-02 | 2021-06-25 | 北京数字政通科技股份有限公司 | Intelligent prevention and control method and intelligent prevention and control system for target object |
| CN113205056B (en) * | 2021-05-12 | 2023-08-22 | 延边国泰新能源汽车有限公司 | Active image recognition early warning system based on public safety problem and processing method thereof |
| CN114666546B (en) * | 2022-03-24 | 2023-06-23 | 中国铁塔股份有限公司江苏省分公司 | Monitoring method and device for communication iron tower and communication iron tower |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101192307A (en) * | 2006-11-17 | 2008-06-04 | 鸿富锦精密工业（深圳）有限公司 | Point cloud triangular topological relations construction method |
| CN102682284A (en) * | 2012-04-05 | 2012-09-19 | 广州中国科学院先进技术研究所 | Early-warning scheduling framework based on cloud and robbing detection method in framework |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009053914A (en) * | 2007-08-27 | 2009-03-12 | Seiko Epson Corp | Image processor and image processing method |
-
2013
- 2013-04-19 CN CN201310139660.3A patent/CN103246869B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101192307A (en) * | 2006-11-17 | 2008-06-04 | 鸿富锦精密工业（深圳）有限公司 | Point cloud triangular topological relations construction method |
| CN102682284A (en) * | 2012-04-05 | 2012-09-19 | 广州中国科学院先进技术研究所 | Early-warning scheduling framework based on cloud and robbing detection method in framework |
Non-Patent Citations (2)
| Title |
|---|
| Facial point detection using boosted regression and graph models;Michel Valstar等;《Computer Vision and Pattern Recognition (CVPR), 2010 IEEE Conference on》;IEEE;20100618;第2729-2736页 * |
| 基于改进特征点定位算法的人脸自动识别系统研究;费俊琳;《中国优秀硕士学位论文全文数据库信息科技辑》;20080615(第6期);第1-63页 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108364438A (en) * | 2017-12-28 | 2018-08-03 | 武汉烽火众智数字技术有限责任公司 | Method for early warning based on front end perception data and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103246869A (en) | 2013-08-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103246869B (en) | Method is monitored in crime based on recognition of face and behavior speech recognition | |
| Sánchez et al. | Revisiting crowd behaviour analysis through deep learning: Taxonomy, anomaly detection, crowd emotions, datasets, opportunities and prospects | |
| CN109765539B (en) | Indoor user behavior monitoring method and device, electrical equipment and home monitoring system | |
| CN106203458B (en) | Crowd video analysis method and system | |
| CN110674772A (en) | Intelligent safety control auxiliary system and method for electric power operation site | |
| CN104915655A (en) | Multi-path monitor video management method and device | |
| CN110942545A (en) | Dense person entrance guard control system and method based on face recognition and video fence | |
| CN108537922A (en) | Visitor's method for early warning based on recognition of face and system | |
| CN104463174A (en) | Multi-angle dynamic people recognition and behavior prediction system | |
| CN108197575A (en) | A kind of abnormal behaviour recognition methods detected based on target detection and bone point and device | |
| CN111564224A (en) | Intelligent monitoring system with health monitoring function and implementation method thereof | |
| CN107230267A (en) | Intelligence In Baogang Kindergarten based on face recognition algorithms is registered method | |
| WO2022165685A1 (en) | Method for determining dangerousness of person, apparatus, system and storage medium | |
| CN112396658A (en) | Indoor personnel positioning method and positioning system based on video | |
| US20210364356A1 (en) | System and method for using artificial intelligence to enable elevated temperature detection of persons using commodity-based thermal cameras | |
| Yang et al. | Detection of non‐suicidal self‐injury based on spatiotemporal features of indoor activities | |
| CN111510675A (en) | Intelligent security system based on face recognition and big data analysis | |
| CN112132048A (en) | Community patrol analysis method and system based on computer vision | |
| CN112613449A (en) | Safety helmet wearing detection and identification method and system based on video face image | |
| Hao et al. | An end-to-end human abnormal behavior recognition framework for crowds with mentally disordered individuals | |
| WO2020167155A1 (en) | Method and system for detecting troubling events during interaction with a self-service device | |
| CN112800975A (en) | Behavior identification method in security check channel based on image processing | |
| CN107920224B (en) | Abnormity warning method, abnormity warning equipment and video monitoring system | |
| CN110633681A (en) | Bed-leaving detection method based on video | |
| CN116778657A (en) | Method and system for intelligently identifying intrusion behavior |
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 |