CN102254169A - Multi-camera-based face recognition method and multi-camera-based face recognition system - Google Patents
Multi-camera-based face recognition method and multi-camera-based face recognition system Download PDFInfo
- Publication number
- CN102254169A CN102254169A CN 201110242426 CN201110242426A CN102254169A CN 102254169 A CN102254169 A CN 102254169A CN 201110242426 CN201110242426 CN 201110242426 CN 201110242426 A CN201110242426 A CN 201110242426A CN 102254169 A CN102254169 A CN 102254169A
- Authority
- CN
- China
- Prior art keywords
- video camera
- target
- camera
- face
- monitoring
- 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 discloses a multi-camera-based face recognition method and a multi-camera-based face recognition system. The method comprises the following steps of: 1, positioning and tracking a target in a preset monitoring sub-region of one camera by the camera, and discriminating face attitude parameters of the target; 2, selecting the proper camera to build a working group according to the position information of the target and the face attitude parameters and acquiring a multi-viewing-angle face image; 3, identifying the target by using a multi-viewing-angle face recognition method; and 4, when the target enters the preset monitoring sub-region of another camera, executing the step 1. The invention has the advantages that: a plurality of cameras finish the positioning and tracking of the target one by one; the method and the system are applicable to monitoring of irregular, blocked and super-large regions; the multi-viewing-angle face image of the target is acquired by using the working group, dynamically built, of the camera, so the availability of the acquired face image is improved, and the maximum efficiency of the camera is achieved; and a monitoring mode without a main camera has expandability and robustness.
Description
Technical field
The present invention relates to irregular, have block and the super large zone in based on the face identification method and the system of multiple-camera, belong to the face recognition technology field of single target.
Background technology
In existing numerous biometrics identification technologies, face recognition technology is subjected to extensive concern in recent years because of advantages such as its distinctive initiative, non-infringement and user friendly always.At present, video monitoring system is a face recognition technology main flow direction in actual applications.Usually, video monitoring system adopts a plurality of video cameras to cooperate and finishes the recognition of face task, and its benefit is: 1) under the situation that does not need people's target to cooperate, can obtain more each visual angle facial image; 2) a plurality of shooting functions cover guarded region better, the blind area do not occur monitoring.Video monitoring system is mainly used in estate management, customs's frontier inspection, safety precaution, public security field such as deploy to ensure effective monitoring and control of illegal activities.
Existing face identification system based on the multiple-camera cooperation adopts a wide-angle (or comprehensive) video camera as main camera people's target to be positioned tracking usually.For example, U.S. patent of invention US20070092245 adopts a wide-angle imaging machine to utilize low-resolution image that object appearing in the whole guarded region is detected, locatees, follows the tracks of, after target enters certain fixing camera supervised subregion in narrow angle, this narrow angle video camera uses high-definition picture to extract face characteristic, this face identification system will be asked for help target need be positive in the face of video camera array, therefore is applicable to Mag ﹠ Bag Entrance etc.Chinese invention patent CN101236599 adopts a wide-angle (or comprehensive) video camera that object appearing in the whole guarded region is detected, locatees and follows the tracks of, and a plurality of ball-shaped camera rotational alignment targets are carried out man face image acquiring.
These are by the face identification system of main camera location people's target, and its limitation is that main camera must can cover whole guarded region, otherwise the monitoring blind area will occur.But, modern architecture usually has a lot of irregular areas (for example Japanese museum in the World's Fair of Shanghai, French museum, Britain museum), these irregular guarded regions are because the space structure complexity can't be provided with the main camera that can cover whole guarded region; A lot of shelters (for example putting the odeum and the museum of a lot of showcases) are perhaps arranged in the regular domain, and in this class guarded region, main camera can produce the monitoring blind area owing to shelter; Perhaps super large zone (for example airport, railway station, gymnasium), in this class guarded region, main camera can can't cover whole guarded region because the visual field is not enough.
Because above-mentioned zone mostly is the public domain, the security problem counts for much, and is related to the country and people's the security of the lives and property.Therefore, research at irregular, have the face identification method and the system of blocking that important meaning is arranged with the super large zone.
Summary of the invention
The objective of the invention is to, face identification method and system based on multiple-camera are provided, it utilizes a plurality of video cameras that people's target is carried out that relay-type location is followed the tracks of and dynamically sets up video camera working group and gathers the various visual angles facial image jointly and carry out recognition of face, avoided irregular, have block with the super large zone in the phenomenon of monitoring blind area appears, thereby better safeguard one's country and the people's lives and property safety.
For solving the problems of the technologies described above, the present invention adopts following technical scheme: the face identification method based on multiple-camera may further comprise the steps:
S1, video camera positions tracking to appearing at its target that presets in the monitoring subregion, and differentiates the human face posture parameter of target; S2 selects suitable video camera building work group and gathers the various visual angles facial image according to target position information and human face posture parameter;
S3 utilizes the various visual angles face identification method to come recognition objective;
S4 presets monitoring during subregion when what target entered another video camera, forwards S1 to.
Video camera described in the preceding method adopts initiatively video camera, i.e. PTZ (Pan-Tilt-Zoom) video camera, thus can obtain the image of clear people's face.
In the preceding method, the video camera that is used to locate tracking in the described working group is motionless in the course of the work, other video camera rotates gathers the various visual angles facial image, the video camera that wherein is used to locate tracking maintains static, can obtain the high-resolution human face image at a certain visual angle available in the recognition of face, other video cameras rotate, and can obtain more various visual angles facial image; In addition, adopt this working method to overcome that the video camera of each in the working group all maintains static in the prior art, a plurality of video camera colocated tracking targets and need carry out the registration on time, the space and need that more overlapping covered problem is arranged between each video camera to target.
Video camera described in the preceding method comprises: imageing sensor, processing unit and communication unit, imageing sensor are used to obtain the image information of scene; Processing unit is used for the scene object appearing is detected, locatees, follows the tracks of processing; Communication unit is used for keeping communicating by letter with center processor.
Among the step S2, along with the variation of the human face posture that moves of people's target, the member's dynamic change working group in, thus when monitoring, gathers relay tracking the various visual angles facial image of people's target timely, be convenient to accurate recognition people target more.
Aforesaid face identification method based on multiple-camera among the step S1, uses six-freedom degree parameter [x, y, z, α, β, γ] represent the position and the human face posture of target, x wherein, y, z is illustrated respectively in three coordinate axis in the volume coordinate, α represents the plane internal rotation angle degree of people's face, and β represents the pitching anglec of rotation of people's face, and γ represents the degree of depth anglec of rotation outside the plane of people's face.
Aforesaid face identification method based on multiple-camera, the method for the building work group described in the step S2 may further comprise the steps:
S10, video camera obtain the positional information and the human face posture parameter of current goal;
S20, the table of comparisons in the queries dictionary, determine to be suitable for capturing the adjacent camera of target facial image, and to they transmission instructions, invite their building work groups, wherein, the described table of comparisons is meant the installation site according to space structure and each video camera of guarded region, the positional information of the target that is suitable for taking of adjacent each video camera of listing of each video camera and the table of comparisons of human face posture parameter value scope, when the building work group, search dictionary, the facial image that obtains is screened in advance, saved the step that the later stage screens once more to the facial image that collects, make the facial image that obtains to be used for recognition of face substantially, so through such screening, video camera is not had the destination to call, but is used effectively, has brought into play the maximum efficiency of video camera.
Aforesaid face identification method based on multiple-camera, the layout of a plurality of video cameras in the following way in the described method: whole guarded region is divided into a plurality of monitoring subregions that preset, each video camera is preset one of them subregion of monitoring, the described visibility region that presets the monitoring subregion less than each video camera, thereby guarantee that video camera can monitor whole guarded region, the blind area does not appear monitoring, and respectively preset zero lap between the monitoring subregion, thereby the difference that presets the monitoring subregion that can effectively enter according to people's target is assigned different video cameras and is positioned tracking, makes that whole monitor procedure is in good order.
Aforesaid face identification method based on multiple-camera, step S4 also comprises: S5, positional information and electronic chart according to target, determine to be responsible for the video camera of location, target region, tracing task, clear, and, send the instruction that rotational alignment targets are captured to other video cameras with this video camera composition working group to the instruction that this video camera transmission recovers preset condition.
Aforesaid face identification method based on multiple-camera, utilization various visual angles face identification method described in the step S3 comes recognition objective to be meant: according to face database, use various visual angles recognition of face program that a plurality of visual angles facial image that obtains is carried out recognition of face.
Realize the face identification system based on multiple-camera of preceding method, it comprises: a center processor and a plurality of video camera, form local area network by wired or wireless mode between center processor and each video camera and center processor can communicate with any video camera in the supervisory system, described center processor is used for selecting suitable video camera building work group and receiving the various visual angles facial image that video camera is gathered in the working group according to the positional information and the human face posture parameter of target; Utilize the various visual angles face identification method to come recognition objective; Described a plurality of camera performance is identical, is used for positioning tracking to entering its target that presets the monitoring subregion, and differentiates the human face posture parameter of target; Gather the various visual angles facial image.
Video camera described in the aforementioned system adopts initiatively video camera, i.e. PTZ (Pan-Tilt-Zoom) video camera, thus can obtain the image of clear people's face.
In the aforementioned system, the video camera that is used to locate tracking in the described working group is motionless in the course of the work, other video camera rotates gathers the various visual angles facial image, the video camera that wherein is used to locate tracking maintains static, can obtain the high-resolution human face image at a certain visual angle available in the recognition of face, other video cameras rotate, and can obtain more various visual angles facial image; In addition, adopt this working method to overcome that the video camera of each in the working group all maintains static in the prior art, a plurality of video camera colocated tracking targets and need carry out the registration on time, the space and need that more overlapping covered problem is arranged between each video camera to target.
Video camera described in the aforementioned system comprises: imageing sensor, processing unit and communication unit, imageing sensor are used to obtain the image information of scene; Processing unit is used for the scene object appearing is detected, locatees, follows the tracks of processing; Communication unit is used for keeping communicating by letter with center processor.
Working group described in the aforementioned system, variation along with the human face posture that moves of people's target, member's dynamic change in the working group, thus the various visual angles facial image of people's target in the relay tracking monitoring, gathered timely, be convenient to accurate recognition people target more.
In the aforesaid system, use six-freedom degree parameter [x, y, z, α, beta, gamma] represent the position and the human face posture of target, x wherein, y, z are illustrated respectively in three coordinate axis in the volume coordinate, and α represents the plane internal rotation angle degree of people's face, β represents the pitching anglec of rotation of people's face, and γ represents the degree of depth anglec of rotation outside the plane of people's face.
In the aforementioned system, the method for building work group may further comprise the steps:
S10, video camera obtain the positional information and the human face posture parameter of current goal;
S20, the table of comparisons in the queries dictionary, determine to be suitable for capturing the adjacent camera of target facial image, and to they transmission instructions, invite their building work groups, wherein, the described table of comparisons is meant the installation site according to space structure and each video camera of guarded region, the positional information of the target that is suitable for taking of adjacent each video camera of listing of each video camera and the table of comparisons of human face posture parameter value scope, when the building work group, search dictionary, the facial image that obtains is screened in advance, saved the step that the later stage screens once more to the facial image that collects, make the facial image that obtains to be used for recognition of face substantially, so through such screening, video camera is not had the destination to call, but is used effectively, has brought into play the maximum efficiency of video camera.
In the aforementioned system, the layout of a plurality of video cameras is in the following way: whole guarded region is divided into a plurality of monitoring subregions that preset, each video camera is preset one of them subregion of monitoring, the described visibility region that presets the monitoring subregion less than each video camera, thereby guarantee that video camera can monitor whole guarded region, the blind area does not appear monitoring, and respectively preset zero lap between the monitoring subregion, thereby the difference that presets the monitoring subregion that can effectively enter according to people's target is assigned different video cameras and is positioned tracking, makes that whole monitor procedure is in good order.
In the aforesaid face identification system based on multiple-camera, described center processor comprises: dictionary module, be used to store according to the positional information of the target that is suitable for taking of listed adjacent each video camera of each video camera in the installation site of the space structure of guarded region and each video camera and the table of comparisons of human face posture parameter value scope, target carried out the various visual angles man face image acquiring thereby make things convenient for center processor to select suitable video camera to form working group according to target current position information and human face posture parameter.
In the aforesaid face identification system based on multiple-camera, described center processor comprises: electric map module, be used for storing the space structure that presets the monitoring subregion and in the position of respectively presetting the video camera installed of monitoring subregion, thereby center processor can be assigned corresponding video camera target is positioned tracking according to the monitoring subregion that presets that target enters, and utilize the adjacent video camera of this video camera to form working group to target collection various visual angles facial image, clear.
In the aforesaid face identification system based on multiple-camera, described center processor comprises: face database and face recognition module, described face recognition module is used to store the recognition of face program.
Compared with prior art, the invention has the beneficial effects as follows: 1) method for supervising and the supervisory system of the present invention's proposition do not need main camera that target is positioned tracking, but finish the location of people's target is followed the tracks of by a plurality of video camera relays, be applicable to that monitoring is irregular, block and the super large zone to have extensibility and robustness; 2) it is motionless in the course of the work to be used to locate the video camera of tracking in the working group of the present invention, other video camera rotates gathers the various visual angles facial image, the video camera that wherein is used to locate tracking maintains static, can obtain the high-resolution human face image at a certain visual angle available in the recognition of face, other video cameras rotate, and can obtain more various visual angles facial image; In addition, adopt this working method to overcome that the video camera of each in the working group all maintains static in the prior art, a plurality of video camera colocated tracking targets and need carry out the registration on time, the space and need that more overlapping covered problem is arranged between each video camera to target; 3) the present invention searches dictionary when the building work group, the facial image that obtains is screened in advance, saved the step that the later stage screens once more to the facial image that collects, make the facial image that obtains to be used for recognition of face substantially, so through such screening, video camera is not had the destination to call, but is used effectively, has brought into play the maximum efficiency of video camera; 4) a plurality of video cameras adopt following layout type among the present invention: whole guarded region is divided into a plurality of monitoring subregions that preset, each video camera is preset one of them subregion of monitoring, the described visibility region that presets the monitoring subregion less than each video camera, thereby guarantee that video camera can monitor whole guarded region, the blind area does not appear monitoring, and respectively preset zero lap between the monitoring subregion, thereby the difference that presets the monitoring subregion that can effectively enter according to people's target is assigned different video cameras and is positioned tracking, makes that whole monitor procedure is in good order.
Description of drawings
Fig. 1 is the workflow diagram of a kind of embodiment of the present invention;
Fig. 2 is the structural representation of a kind of embodiment of the present invention;
Fig. 3 is the structural representation of video camera among a kind of embodiment of the present invention;
Fig. 4 is the true layout of museum's irregular area;
Fig. 5 is the layout that a rule has occlusion area.
Reference numeral: 1~20-video camera, 21-center processor, 22-dividing plate.
The present invention is further illustrated below in conjunction with the drawings and specific embodiments.
Embodiment
Embodiments of the invention:
Embodiment 1, based on the face identification method of multiple-camera, as shown in Figure 1, may further comprise the steps:
S1, video camera positions tracking to appearing at its target that presets in the monitoring subregion, and differentiates the human face posture parameter of target;
S2 selects suitable video camera building work group and gathers the various visual angles facial image according to target position information and human face posture parameter;
S3 utilizes the various visual angles face identification method to come recognition objective;
S4 presets monitoring during subregion when what target entered another video camera, forwards S1 to.
In the said method, described video camera adopts initiatively video camera, i.e. PTZ (Pan-Tilt-Zoom) video camera, thus can obtain the image of clear people's face.
In the said method, the video camera that is used to locate tracking in the described working group is motionless in the course of the work, other video camera rotates gathers the various visual angles facial image, the video camera that wherein is used to locate tracking maintains static, can obtain the high-resolution human face image at a certain visual angle available in the recognition of face, other video cameras rotate, and can obtain more various visual angles facial image; In addition, adopt this working method to overcome that the video camera of each in the working group all maintains static in the prior art, a plurality of video camera colocated tracking targets and need carry out the registration on time, the space and need that more overlapping covered problem is arranged between each video camera to target.
In the said method, described video camera comprises: imageing sensor, processing unit and communication unit, imageing sensor are used to obtain the image information of scene; Processing unit is used for the scene object appearing is detected, locatees, follows the tracks of processing; Communication unit is used for keeping communicating by letter with center processor.
Among the step S2, along with the variation of the human face posture that moves of people's target, the member's dynamic change working group in, thus when monitoring, gathers relay tracking the various visual angles facial image of people's target timely, be convenient to accurate recognition people target more.
In the said method, described step S1 uses six-freedom degree parameter [x, y, z, α, β, γ] represent the position and the human face posture of target, x wherein, y, z is illustrated respectively in three coordinate axis in the volume coordinate, α represents the plane internal rotation angle degree of people's face, and β represents the pitching anglec of rotation of people's face, and γ represents the degree of depth anglec of rotation outside the plane of people's face.
In the said method, the method for the described building work group of step S2 may further comprise the steps:
S10, video camera obtain the positional information and the human face posture parameter of current goal;
S20, the table of comparisons in the queries dictionary, determine to be suitable for capturing the adjacent camera of target facial image, and to they transmission instructions, invite their building work groups, wherein, the described table of comparisons is meant the installation site according to space structure and each video camera of guarded region, the positional information of the target that is suitable for taking of adjacent each video camera of listing of each video camera and the table of comparisons of human face posture parameter value scope, when the building work group, search dictionary, the facial image that obtains screened in advance, such as human face posture parameter display-object back to video camera a, face video camera b, just have only b not have a so in the dictionary, the facial image of acquisition is positive, utilizable substantially, saved the step that the later stage screens once more to the facial image that collects, make the facial image that obtains to be used for recognition of face substantially, so through such screening, video camera is not had the destination to call, but used effectively, brought into play the maximum efficiency of video camera.
In the said method, the layout of a plurality of video cameras is in the following way: whole guarded region is divided into a plurality of monitoring subregions that preset, each video camera is preset one of them subregion of monitoring, the described visibility region that presets the monitoring subregion less than each video camera, thereby guarantee that video camera can monitor whole guarded region, the blind area does not appear monitoring, and respectively preset zero lap between the monitoring subregion, thereby the difference that presets the monitoring subregion that can effectively enter according to people's target is assigned different video cameras and is positioned tracking, makes that whole monitor procedure is in good order.
In the said method, step S4 also comprises:
S5, positional information and electronic chart according to target, determine to be responsible for the video camera of location, target region, tracing task, clear, and, send the instruction that rotational alignment targets are captured to other video cameras with this video camera composition working group to the instruction that this video camera transmission recovers preset condition.
In the said method, the described utilization various visual angles of step S3 face identification method comes recognition objective to be meant: according to face database, use various visual angles recognition of face program that a plurality of visual angles facial image that obtains is carried out recognition of face.
Realize the face identification system based on multiple-camera of said method, as shown in Figure 2, it comprises: a center processor and a plurality of video camera, form local area network by wired or wireless mode between center processor and each video camera and center processor can communicate with any video camera in the supervisory system, described center processor is used for selecting suitable video camera building work group and receiving the various visual angles facial image that video camera is gathered in the working group according to the positional information and the human face posture parameter of target; Utilize the various visual angles face identification method to come recognition objective; Described a plurality of camera performance is identical, is used for positioning tracking to entering its target that presets the monitoring subregion, and differentiates the human face posture parameter of target; Gather the various visual angles facial image.
In the said system, described video camera adopts initiatively video camera, i.e. PTZ (Pan-Tilt-Zoom) video camera, thus can obtain the image of clear people's face.
In the said system, the video camera that is used to locate tracking in the described working group is motionless in the course of the work, other video camera rotates gathers the various visual angles facial image, the video camera that wherein is used to locate tracking maintains static, can obtain the high-resolution human face image at a certain visual angle available in the recognition of face, other video cameras rotate, and can obtain more various visual angles facial image; In addition, adopt this working method to overcome that the video camera of each in the working group all maintains static in the prior art, a plurality of video camera colocated tracking targets and need carry out the registration on time, the space and need that more overlapping covered problem is arranged between each video camera to target.
In the said system, described video camera (as shown in Figure 3) comprising: imageing sensor, processing unit and communication unit, imageing sensor are used to obtain the image information of scene; Processing unit is used for the scene object appearing is detected, locatees, follows the tracks of processing; Communication unit is used for keeping communicating by letter with center processor.
Working group described in the said system, variation along with the human face posture that moves of people's target, member's dynamic change in the working group, thus the various visual angles facial image of people's target in the relay tracking monitoring, gathered timely, be convenient to accurate recognition people target more.
In the said system, use six-freedom degree parameter [x, y, z, α, beta, gamma] represent the position and the human face posture of target, x wherein, y, z are illustrated respectively in three coordinate axis in the volume coordinate, and α represents the plane internal rotation angle degree of people's face, β represents the pitching anglec of rotation of people's face, and γ represents the degree of depth anglec of rotation outside the plane of people's face.
In the said system, the method for building work group may further comprise the steps:
S10, video camera obtain the positional information and the human face posture parameter of current goal;
S20, the table of comparisons in the queries dictionary, determine to be suitable for capturing the adjacent camera of target facial image, and to they transmission instructions, invite their building work groups, wherein, the described table of comparisons is meant the installation site according to space structure and each video camera of guarded region, the positional information of the target that is suitable for taking of adjacent each video camera of listing of each video camera and the table of comparisons of human face posture parameter value scope, when the building work group, search dictionary, the facial image that obtains screened in advance, such as human face posture parameter display-object back to video camera a, face video camera b, just have only b not have a so in the dictionary, the facial image of acquisition is positive, utilizable substantially, saved the step that the later stage screens once more to the facial image that collects, make the facial image that obtains to be used for recognition of face substantially, so through such screening, video camera is not had the destination to call, but used effectively, brought into play the maximum efficiency of video camera.
In the said system, the layout of a plurality of video cameras is in the following way: whole guarded region is divided into a plurality of monitoring subregions that preset, each video camera is preset one of them subregion of monitoring, the described visibility region that presets the monitoring subregion less than each video camera, thereby guarantee that video camera can monitor whole guarded region, the blind area does not appear monitoring, and respectively preset zero lap between the monitoring subregion, thereby the difference that presets the monitoring subregion that can effectively enter according to people's target is assigned different video cameras and is positioned tracking, makes that whole monitor procedure is in good order.
In the said system, described center processor comprises: dictionary module, be used to store according to the positional information of the target that is suitable for taking of listed adjacent each video camera of each video camera in the installation site of the space structure of guarded region and each video camera and the table of comparisons of human face posture parameter value scope, target carried out the various visual angles man face image acquiring thereby make things convenient for center processor to select suitable video camera to form working group according to target current position information and human face posture parameter.
In the said system, described center processor comprises:
Electric map module, be used for storing the space structure that presets the monitoring subregion and in the position of respectively presetting the video camera installed of monitoring subregion, thereby center processor can be assigned corresponding video camera target is positioned tracking according to the monitoring subregion that presets that target enters, and utilize the adjacent video camera of this video camera to form working group to target collection various visual angles facial image, clear.
In the said system, described center processor comprises:
Face database and face recognition module, described face recognition module is used to store the recognition of face program.
Store specific jumbo face database and the high-performance recognition of face program of a cover in the center processor 21, also store one one for selecting the used dictionary of suitable video camera building work group and one to comprise the electronic chart that each video camera presets the distribution of monitoring subregion.The task of center processor 21 has: 1) receive positional information, the human face posture parameter of people's target of being transmitted by video camera, and preserve the various visual angles facial image that collects; 2) according to the positional information and the human face posture parameter of people's target, search dictionary, select suitable video camera building work group; 3), determine to be responsible for the video camera of location, target region, tracing task according to the position and the electronic chart of people's target; 4) send instruction to video camera, comprise recover preset condition, the rotational alignment target is captured; 5), use various visual angles recognition of face program that a plurality of visual angles facial image that obtains is carried out recognition of face according to face database; 6) take corresponding action according to recognition result, as reporting to the police etc.
Video camera 1~20 adopts initiatively video camera, i.e. PTZ (Pan-Tilt-Zoom) video camera is to obtain the image of clear people's face.Each video camera comprises imageing sensor, processing unit and communication unit.
The layout type of video camera is as follows:
Whole monitoring space is divided into a plurality of monitoring subregions that preset, and each video camera is preset one of them subregion of monitoring.The guarded region that presets of respectively numbering video camera among Fig. 4 promptly is the subregion by the identical numbering of dotted line mark, promptly video camera 1 to preset the monitoring subregion be subregion 1.All preset the monitoring subregion can cover whole guarded region, and each presets zero lap between the monitoring subregion.
It is different with the visibility region of video camera that the video camera that defines among the present invention presets the monitoring subregion.Generally speaking, the visibility region of video camera is greater than its preset monitoring subregion, so has certain overlapping region between the visibility region of each video camera.
The cooperation principle of a plurality of video cameras is as follows:
Each video camera uses high-definition picture to monitor preset monitoring subregion separately in the normal state.After people's target occurs, it is motionless that the corresponding video camera of monitoring subregion is preset at people's target place, be responsible for people's target is positioned tracking, and positional information and the human face posture parameter of transmitting people's target are given center processor, center processor is selected suitable adjacent camera building work group according to the information of transmitting and the dictionary that prestores, and the interior video camera rotational alignment people's target of while command group is also gathered the facial image of people's target.Preset monitoring during subregion when what target entered other video camera, center processor specifies corresponding video camera to recover preset condition, be responsible for people's target is positioned tracking, thereby finished with the video camera relay monitoring of goal activities regional change and the various visual angles man face image acquiring of dynamic duty group.
With the single target is example, and the concrete cooperation scheme of a plurality of video cameras is as follows:
(1-1) each video camera 1~20 of center processor 21 initialization enters and presets monitor state.Use preset monitoring parameter (comprising monitoring position, focal length, resolution, bias light etc.), monitoring separately preset the monitoring subregion, and utilize time-domain difference and Face Detection method to detect whether to occur people's target in this subregion;
(1-2) when presetting of video camera 15 people's target occurred in the monitoring subregion 15, this video camera began the target localization to the people, and utilizes the Mean-Shift algorithm that target is followed the tracks of, and uses template matching method to differentiate the human face posture parameter of people's target simultaneously;
(1-3) video camera 15 sends the positional information and the human face posture parameter of people's target to center processor 21.Location estimation and human face posture according to 15 pairs of people's targets of video camera are differentiated, center processor 21 sends instruction to the video camera that (as video camera 12,13,14,16) adjacent with video camera 15 is suitable for gathering the target facial image, invite their building work groups, and to the positional information of their sender's targets;
(1-4) in working group, video camera 15 is responsible for people's target is positioned, follows the tracks of, and high-resolution monitoring image is provided; The positional information of people's target that other video camera provides according to video camera 15, rotational alignment people's target face takes, and gathers the facial image of more various visual angles;
(1-5) facial image of the various visual angles that each video camera will collect in the working group sends center processor 21 to, center processor 21 uses the AdaBoost sorter and carries out the various visual angles recognition of face based on the decision level fusion method of Bayesian inference, mates with face database and discerns people's target;
(1-6) preset monitoring during subregion when what processor 21 finder target locations in center had entered video camera 13, center processor 21 sends instruction to video camera 13, orders it to enter and presets monitor state.Afterwards, video camera 13 replaces video camera 15, repeats (1-2) to (1-5) step, finishes the relay-type monitoring to people's target.
In the said method, step (1-3) comprises following each step according to the method for the differentiation of people's target position information and human face posture being initiated the building work group:
(2-1) at each video camera, the present invention uses six-freedom degree parameter [x, y, z, α, beta, gamma] represent the position and the human face posture of people's target, x wherein, y, z are illustrated respectively in three coordinate axis in the volume coordinate, and α represents the plane internal rotation angle degree of people's face, β represents the pitching anglec of rotation of people's face, and γ represents the degree of depth anglec of rotation outside the plane of people's face.If [x
a, y
a, z
a, α
a, β
a, γ
a] be detected positional information and human face posture parameter in the video camera 15 when forefathers' target;
(2-2) store a dictionary in the center processor 21 in advance, this dictionary is according to the installation site of space structure and each video camera of guarded region, positional information and human face posture parameter [x that adjacent each video camera of each video camera and they are suitable for people's target of taking have separately been listed, y, z, α, beta, gamma] table of comparisons of span.
(2-3) positional information and the human face posture parameter [x that obtains according to video camera 15 when forefathers' target
a, y
a, z
a, α
a, β
a, γ
a], center processor 21 is inquired about the dictionary that prestores, the adjacent camera of determining to be suitable for capturing people's target facial image, and, invite their building work groups to their transmission instructions.Along with the variation of the human face posture that moves of people's target, group member's dynamic change.
According to the practical use difference, the face database of storage in the center processor 21 both can be to allow the face database that enters or pass through, also can be the face database that is prohibited from entering or orders to arrest.
During the museum opened, each video camera was initially taked common function mode, separately monitoring separately preset the monitoring subregion, and transmit high-resolution monitoring image to center processor 21.When center processor 21 or monitor staff judge that unusual target appears in certain zone, start the working method that the present invention proposes.In addition, during closing the shop, system will be set to the working method that the present invention proposes.
Claims (10)
1. based on the face identification method of multiple-camera, it is characterized in that, may further comprise the steps:
S1, video camera positions tracking to appearing at its target that presets in the monitoring subregion, and differentiates the human face posture parameter of target;
S2 selects suitable video camera building work group and gathers the various visual angles facial image according to target position information and human face posture parameter;
S3 utilizes the various visual angles face identification method to come recognition objective;
S4 presets monitoring during subregion when what target entered another video camera, forwards S1 to.
2. the face identification method based on multiple-camera according to claim 1, it is characterized in that, the video camera that is used to locate tracking in the working group described in the step S2 is motionless in the course of the work, and other video camera rotates gathers the various visual angles facial image.
3. the face identification method based on multiple-camera according to claim 1 is characterized in that, the method for the building work group described in the step S2 may further comprise the steps:
S10, video camera obtain the positional information and the human face posture parameter of current goal;
S20, the table of comparisons in the queries dictionary, determine to be suitable for capturing the adjacent camera of target facial image, and to they transmission instructions, invite their building work groups, the described table of comparisons is meant the installation site according to space structure and each video camera of guarded region, the positional information of the target that is suitable for taking of adjacent each video camera of listing of each video camera and the table of comparisons of human face posture parameter value scope.
4. the face identification method based on multiple-camera according to claim 1, it is characterized in that, the layout of a plurality of video cameras in the following way in the described method: whole guarded region is divided into a plurality of monitoring subregions that preset, each video camera is preset one of them subregion of monitoring, and each presets zero lap between the monitoring subregion.
5. the face identification method based on multiple-camera according to claim 1 is characterized in that step S4 also comprises:
S5 according to the positional information and the electronic chart of target, determines to be responsible for the video camera of location, target region, tracing task.
6. the face identification method based on multiple-camera according to claim 1, it is characterized in that, utilization various visual angles face identification method described in the step S3 comes recognition objective to be meant: according to face database, use various visual angles recognition of face program that a plurality of visual angles facial image that obtains is carried out recognition of face.
7. realize the face identification system based on multiple-camera of the described method of claim 1~6, it is characterized in that, it comprises: a center processor and a plurality of video camera, form local area network by wired or wireless mode between center processor and each video camera and center processor can communicate with any video camera in the supervisory system, described center processor is used for selecting suitable video camera building work group and receiving the various visual angles facial image that video camera is gathered in the working group according to the positional information and the human face posture parameter of target; Utilize the various visual angles face identification method to come recognition objective; Described a plurality of camera performance is identical, is used for positioning tracking to entering its target that presets the monitoring subregion, and differentiates the human face posture parameter of target; Gather the various visual angles facial image.
8. the face identification system based on multiple-camera according to claim 7 is characterized in that, described center processor comprises:
Dictionary module is used to store according to the positional information of the target that is suitable for taking of listed adjacent each video camera of each video camera in the installation site of the space structure of guarded region and each video camera and the table of comparisons of human face posture parameter value scope.
9. the face identification system based on multiple-camera according to claim 7 is characterized in that, described center processor comprises:
Electric map module is used for storing the space structure that presets the monitoring subregion and in the position of respectively presetting the video camera installed of monitoring subregion.
10. the face identification system based on multiple-camera according to claim 7 is characterized in that, described center processor comprises:
Face database and face recognition module, described face recognition module is used to store the recognition of face program.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2011102424264A CN102254169B (en) | 2011-08-23 | 2011-08-23 | Multi-camera-based face recognition method and multi-camera-based face recognition system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2011102424264A CN102254169B (en) | 2011-08-23 | 2011-08-23 | Multi-camera-based face recognition method and multi-camera-based face recognition system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102254169A true CN102254169A (en) | 2011-11-23 |
CN102254169B CN102254169B (en) | 2012-08-22 |
Family
ID=44981423
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2011102424264A Expired - Fee Related CN102254169B (en) | 2011-08-23 | 2011-08-23 | Multi-camera-based face recognition method and multi-camera-based face recognition system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102254169B (en) |
Cited By (36)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103310219A (en) * | 2012-03-08 | 2013-09-18 | 佳能株式会社 | Method and equipment for precision evaluation of object shape registration, and method and equipment for registration |
CN103824045A (en) * | 2012-11-16 | 2014-05-28 | 中兴通讯股份有限公司 | Face recognition and tracking method and face recognition and tracking system |
CN103856762A (en) * | 2014-01-26 | 2014-06-11 | 孙培梁 | Multi-camera intelligent selection and video priority judgment system and selection method |
CN103986917A (en) * | 2014-06-03 | 2014-08-13 | 中科融通物联科技无锡有限公司 | Multi-angle thermal image monitoring system |
CN104615997A (en) * | 2015-02-15 | 2015-05-13 | 四川川大智胜软件股份有限公司 | Human face anti-fake method based on multiple cameras |
CN104615996A (en) * | 2015-02-15 | 2015-05-13 | 四川川大智胜软件股份有限公司 | Multi-view two-dimension facial feature point automatic positioning method |
CN105474232A (en) * | 2013-06-17 | 2016-04-06 | 匡特莫格公司 | System and method for biometric identification |
CN105512602A (en) * | 2014-10-16 | 2016-04-20 | 南京索酷信息科技有限公司 | Method of applying face recognition based on global and local features to smart community |
CN105857252A (en) * | 2015-12-07 | 2016-08-17 | 乐视网信息技术(北京)股份有限公司 | Automobile anti-theft tracing method and system |
CN106127911A (en) * | 2016-08-25 | 2016-11-16 | 济南新吉纳远程测控股份有限公司 | Gate inhibition's tracking identification method and device, recognition and tracking system, gate control system |
CN103310219B (en) * | 2012-03-08 | 2016-11-30 | 佳能株式会社 | The precision assessment method of registration with objects shape and equipment, the method and apparatus of registration |
CN106530357A (en) * | 2016-11-30 | 2017-03-22 | 深圳市泰沃德自动化技术有限公司 | Vision alignment control device and calibration method |
CN106863355A (en) * | 2016-12-27 | 2017-06-20 | 北京光年无限科技有限公司 | A kind of object identification method and robot for robot |
WO2017132901A1 (en) * | 2016-02-03 | 2017-08-10 | 华为技术有限公司 | Image acquisition method, device and system |
CN107093171A (en) * | 2016-02-18 | 2017-08-25 | 腾讯科技(深圳)有限公司 | A kind of image processing method and device, system |
CN107197200A (en) * | 2017-05-22 | 2017-09-22 | 北斗羲和城市空间科技(北京)有限公司 | It is a kind of to realize the method and device that monitor video is shown |
CN108759834A (en) * | 2018-04-28 | 2018-11-06 | 温州大学激光与光电智能制造研究院 | A kind of localization method based on overall Vision |
CN108875507A (en) * | 2017-11-22 | 2018-11-23 | 北京旷视科技有限公司 | Pedestrian tracting method, equipment, system and computer readable storage medium |
CN109190534A (en) * | 2018-08-23 | 2019-01-11 | 广州凯风科技有限公司 | A kind of synchronous method in All-in-One channel |
CN109558781A (en) * | 2018-08-02 | 2019-04-02 | 北京市商汤科技开发有限公司 | A kind of multi-angle video recognition methods and device, equipment and storage medium |
CN109670451A (en) * | 2018-12-20 | 2019-04-23 | 天津天地伟业信息系统集成有限公司 | Automatic face recognition tracking |
CN109784199A (en) * | 2018-12-21 | 2019-05-21 | 深圳云天励飞技术有限公司 | Analysis method of going together and Related product |
CN110008811A (en) * | 2019-01-21 | 2019-07-12 | 北京工业职业技术学院 | Face identification system and method |
CN110032922A (en) * | 2017-12-13 | 2019-07-19 | 虚拟现实软件 | The method and system of augmented reality is provided for mining |
CN110445966A (en) * | 2019-08-09 | 2019-11-12 | 北京润博互联数字科技有限公司 | A kind of panorama camera video capture method, apparatus, electronic equipment and storage medium |
WO2020057355A1 (en) * | 2018-09-21 | 2020-03-26 | 深圳市九洲电器有限公司 | Three-dimensional modeling method and device |
CN111372051A (en) * | 2020-03-17 | 2020-07-03 | 三一重工股份有限公司 | Multi-camera linkage blind area detection method and device and electronic equipment |
RU2731519C1 (en) * | 2019-09-19 | 2020-09-03 | Акционерное общество "Федеральный научно-производственный центр "Производственное объединение "Старт" им. М.В. Проценко" (АО "ФНПЦ ПО "Старт" им. М.В. Проценко") | Combined access control and monitoring system for system for physical protection of critical objects |
CN111629144A (en) * | 2020-04-28 | 2020-09-04 | 深圳供电局有限公司 | Image acquisition equipment control method and device, computer equipment and storage medium |
CN111666444A (en) * | 2020-06-02 | 2020-09-15 | 中国科学院计算技术研究所 | Audio push method and system based on artificial intelligence, and related method and equipment |
CN111860063A (en) * | 2019-04-30 | 2020-10-30 | 杭州海康威视数字技术股份有限公司 | Gait data construction system, method and device |
CN111915842A (en) * | 2020-07-02 | 2020-11-10 | 广东技术师范大学 | Abnormity monitoring method and device, computer equipment and storage medium |
CN111931564A (en) * | 2020-06-29 | 2020-11-13 | 北京大学 | Target tracking method and device based on face recognition |
CN111967311A (en) * | 2020-07-06 | 2020-11-20 | 广东技术师范大学 | Emotion recognition method and device, computer equipment and storage medium |
CN113095116A (en) * | 2019-12-23 | 2021-07-09 | 深圳云天励飞技术有限公司 | Identity recognition method and related product |
CN113542587A (en) * | 2021-05-25 | 2021-10-22 | 浙江大华技术股份有限公司 | Image capturing method and device, electronic equipment and computer readable storage medium |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
RU2694140C1 (en) * | 2019-04-04 | 2019-07-09 | Общество с ограниченной ответственностью "Скайтрэк" (ООО "Скайтрэк") | Method of human identification in a mode of simultaneous operation of a group of video cameras |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101174298A (en) * | 2006-11-02 | 2008-05-07 | 上海银晨智能识别科技有限公司 | Scattered-point high-volume face recognition system and recognizing method thereof |
CN101236599A (en) * | 2007-12-29 | 2008-08-06 | 浙江工业大学 | Human face recognition detection device based on multi- video camera information integration |
CN101894428A (en) * | 2010-05-12 | 2010-11-24 | 北京海鑫智圣技术有限公司 | ATM (Automated Teller Machine) intelligent monitoring system |
-
2011
- 2011-08-23 CN CN2011102424264A patent/CN102254169B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101174298A (en) * | 2006-11-02 | 2008-05-07 | 上海银晨智能识别科技有限公司 | Scattered-point high-volume face recognition system and recognizing method thereof |
CN101236599A (en) * | 2007-12-29 | 2008-08-06 | 浙江工业大学 | Human face recognition detection device based on multi- video camera information integration |
CN101894428A (en) * | 2010-05-12 | 2010-11-24 | 北京海鑫智圣技术有限公司 | ATM (Automated Teller Machine) intelligent monitoring system |
Cited By (50)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103310219A (en) * | 2012-03-08 | 2013-09-18 | 佳能株式会社 | Method and equipment for precision evaluation of object shape registration, and method and equipment for registration |
CN103310219B (en) * | 2012-03-08 | 2016-11-30 | 佳能株式会社 | The precision assessment method of registration with objects shape and equipment, the method and apparatus of registration |
CN103824045A (en) * | 2012-11-16 | 2014-05-28 | 中兴通讯股份有限公司 | Face recognition and tracking method and face recognition and tracking system |
CN105474232A (en) * | 2013-06-17 | 2016-04-06 | 匡特莫格公司 | System and method for biometric identification |
CN103856762A (en) * | 2014-01-26 | 2014-06-11 | 孙培梁 | Multi-camera intelligent selection and video priority judgment system and selection method |
CN103986917B (en) * | 2014-06-03 | 2017-04-26 | 中科融通物联科技无锡有限公司 | Multi-angle thermal image monitoring system |
CN103986917A (en) * | 2014-06-03 | 2014-08-13 | 中科融通物联科技无锡有限公司 | Multi-angle thermal image monitoring system |
CN105512602A (en) * | 2014-10-16 | 2016-04-20 | 南京索酷信息科技有限公司 | Method of applying face recognition based on global and local features to smart community |
CN104615997A (en) * | 2015-02-15 | 2015-05-13 | 四川川大智胜软件股份有限公司 | Human face anti-fake method based on multiple cameras |
CN104615996A (en) * | 2015-02-15 | 2015-05-13 | 四川川大智胜软件股份有限公司 | Multi-view two-dimension facial feature point automatic positioning method |
CN104615996B (en) * | 2015-02-15 | 2017-11-21 | 四川川大智胜软件股份有限公司 | A kind of various visual angles two-dimension human face automatic positioning method for characteristic point |
CN105857252A (en) * | 2015-12-07 | 2016-08-17 | 乐视网信息技术(北京)股份有限公司 | Automobile anti-theft tracing method and system |
WO2017132901A1 (en) * | 2016-02-03 | 2017-08-10 | 华为技术有限公司 | Image acquisition method, device and system |
CN107431751A (en) * | 2016-02-03 | 2017-12-01 | 华为技术有限公司 | Gather the methods, devices and systems of image |
CN107431751B (en) * | 2016-02-03 | 2020-04-21 | 华为技术有限公司 | Method, device and system for acquiring image |
CN107093171B (en) * | 2016-02-18 | 2021-04-30 | 腾讯科技(深圳)有限公司 | Image processing method, device and system |
CN107093171A (en) * | 2016-02-18 | 2017-08-25 | 腾讯科技(深圳)有限公司 | A kind of image processing method and device, system |
CN106127911A (en) * | 2016-08-25 | 2016-11-16 | 济南新吉纳远程测控股份有限公司 | Gate inhibition's tracking identification method and device, recognition and tracking system, gate control system |
CN106530357B (en) * | 2016-11-30 | 2023-10-03 | 深圳市泰沃德技术有限公司 | Visual alignment control device and calibration method |
CN106530357A (en) * | 2016-11-30 | 2017-03-22 | 深圳市泰沃德自动化技术有限公司 | Vision alignment control device and calibration method |
CN106863355A (en) * | 2016-12-27 | 2017-06-20 | 北京光年无限科技有限公司 | A kind of object identification method and robot for robot |
CN107197200A (en) * | 2017-05-22 | 2017-09-22 | 北斗羲和城市空间科技(北京)有限公司 | It is a kind of to realize the method and device that monitor video is shown |
CN108875507A (en) * | 2017-11-22 | 2018-11-23 | 北京旷视科技有限公司 | Pedestrian tracting method, equipment, system and computer readable storage medium |
CN110032922A (en) * | 2017-12-13 | 2019-07-19 | 虚拟现实软件 | The method and system of augmented reality is provided for mining |
CN108759834A (en) * | 2018-04-28 | 2018-11-06 | 温州大学激光与光电智能制造研究院 | A kind of localization method based on overall Vision |
CN108759834B (en) * | 2018-04-28 | 2023-03-21 | 温州大学激光与光电智能制造研究院 | Positioning method based on global vision |
CN109558781A (en) * | 2018-08-02 | 2019-04-02 | 北京市商汤科技开发有限公司 | A kind of multi-angle video recognition methods and device, equipment and storage medium |
CN109190534A (en) * | 2018-08-23 | 2019-01-11 | 广州凯风科技有限公司 | A kind of synchronous method in All-in-One channel |
WO2020057355A1 (en) * | 2018-09-21 | 2020-03-26 | 深圳市九洲电器有限公司 | Three-dimensional modeling method and device |
CN109670451A (en) * | 2018-12-20 | 2019-04-23 | 天津天地伟业信息系统集成有限公司 | Automatic face recognition tracking |
CN109784199A (en) * | 2018-12-21 | 2019-05-21 | 深圳云天励飞技术有限公司 | Analysis method of going together and Related product |
CN109784199B (en) * | 2018-12-21 | 2020-11-24 | 深圳云天励飞技术有限公司 | Peer-to-peer analysis method and related product |
CN110008811A (en) * | 2019-01-21 | 2019-07-12 | 北京工业职业技术学院 | Face identification system and method |
CN111860063B (en) * | 2019-04-30 | 2023-08-11 | 杭州海康威视数字技术股份有限公司 | Gait data construction system, method and device |
CN111860063A (en) * | 2019-04-30 | 2020-10-30 | 杭州海康威视数字技术股份有限公司 | Gait data construction system, method and device |
CN110445966A (en) * | 2019-08-09 | 2019-11-12 | 北京润博互联数字科技有限公司 | A kind of panorama camera video capture method, apparatus, electronic equipment and storage medium |
RU2731519C1 (en) * | 2019-09-19 | 2020-09-03 | Акционерное общество "Федеральный научно-производственный центр "Производственное объединение "Старт" им. М.В. Проценко" (АО "ФНПЦ ПО "Старт" им. М.В. Проценко") | Combined access control and monitoring system for system for physical protection of critical objects |
CN113095116B (en) * | 2019-12-23 | 2024-03-22 | 深圳云天励飞技术有限公司 | Identity recognition method and related product |
CN113095116A (en) * | 2019-12-23 | 2021-07-09 | 深圳云天励飞技术有限公司 | Identity recognition method and related product |
CN111372051A (en) * | 2020-03-17 | 2020-07-03 | 三一重工股份有限公司 | Multi-camera linkage blind area detection method and device and electronic equipment |
CN111629144B (en) * | 2020-04-28 | 2022-11-18 | 深圳供电局有限公司 | Image acquisition equipment control method and device, computer equipment and storage medium |
CN111629144A (en) * | 2020-04-28 | 2020-09-04 | 深圳供电局有限公司 | Image acquisition equipment control method and device, computer equipment and storage medium |
CN111666444A (en) * | 2020-06-02 | 2020-09-15 | 中国科学院计算技术研究所 | Audio push method and system based on artificial intelligence, and related method and equipment |
CN111931564A (en) * | 2020-06-29 | 2020-11-13 | 北京大学 | Target tracking method and device based on face recognition |
CN111915842A (en) * | 2020-07-02 | 2020-11-10 | 广东技术师范大学 | Abnormity monitoring method and device, computer equipment and storage medium |
CN111915842B (en) * | 2020-07-02 | 2021-11-05 | 广东技术师范大学 | Abnormity monitoring method and device, computer equipment and storage medium |
CN111967311B (en) * | 2020-07-06 | 2021-09-10 | 广东技术师范大学 | Emotion recognition method and device, computer equipment and storage medium |
CN111967311A (en) * | 2020-07-06 | 2020-11-20 | 广东技术师范大学 | Emotion recognition method and device, computer equipment and storage medium |
CN113542587A (en) * | 2021-05-25 | 2021-10-22 | 浙江大华技术股份有限公司 | Image capturing method and device, electronic equipment and computer readable storage medium |
RU2792788C1 (en) * | 2022-03-31 | 2023-03-24 | Акционерное общество "Научно-исследовательский и конструкторский институт радиоэлектронной техники" (АО "НИКИРЭТ") | Combined set of checkpoints for the physical protection system (pps) of critical facilities with automation of access control and management processes |
Also Published As
Publication number | Publication date |
---|---|
CN102254169B (en) | 2012-08-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102254169B (en) | Multi-camera-based face recognition method and multi-camera-based face recognition system | |
CN103049734A (en) | Method and system for finding person in public place | |
CN202306582U (en) | System for searching for persons in public place | |
CN101465033B (en) | Automatic tracking recognition system and method | |
CN101635834A (en) | Automatic tracing identification system for artificial neural control | |
CN105979210A (en) | Pedestrian identification system based on multi-ball multi-gun camera array | |
KR101541272B1 (en) | Apparatus and method for detecting terrorism using irregular motion of peoples | |
JP5894807B2 (en) | Surveillance device, surveillance camera system, and video transmission method | |
CN107529221A (en) | A kind of follow-up analysis system and method for combination video monitoring and Wi Fi positioning | |
CN110321780B (en) | Abnormal falling behavior detection method based on space-time motion characteristics | |
JP5674307B2 (en) | Subject detection system and subject detection method | |
JP5406074B2 (en) | Worker position recognition system | |
CN102009879A (en) | Elevator automatic keying control system and method, face model training system and method | |
CN108540780A (en) | Intelligent mobile household monitoring system based on sweeping robot equipment | |
CN108710827B (en) | A kind of micro- police service inspection in community and information automatic analysis system and method | |
CN105894700A (en) | Image-based in-vehicle moving object remote observing and warning device and method | |
CN103440475A (en) | Automatic teller machine user face visibility judging system and method | |
CN106339657B (en) | Crop straw burning monitoring method based on monitor video, device | |
KR20120119144A (en) | Apparatus and method of camera-based intelligent management | |
CN108540752A (en) | The methods, devices and systems that target object in video monitoring is identified | |
CN105872481A (en) | Tracking and positioning device, method and system combining visual signals and wireless signals | |
KR101326707B1 (en) | Camera system for vehicle number recognition and security | |
CN110378292A (en) | Three dimension location system and method | |
CN108259846A (en) | Industrial patrol robot system | |
CN100496122C (en) | Method for tracking principal and subordinate videos by using single video camera |
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: 20120822 Termination date: 20210823 |
|
CF01 | Termination of patent right due to non-payment of annual fee |