CN101795400A - Method for actively tracking and monitoring infants and realization system thereof - Google Patents
Method for actively tracking and monitoring infants and realization system thereof Download PDFInfo
- Publication number
- CN101795400A CN101795400A CN 201010125536 CN201010125536A CN101795400A CN 101795400 A CN101795400 A CN 101795400A CN 201010125536 CN201010125536 CN 201010125536 CN 201010125536 A CN201010125536 A CN 201010125536A CN 101795400 A CN101795400 A CN 101795400A
- Authority
- CN
- China
- Prior art keywords
- infant
- face
- people
- module
- video
- 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
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method for actively tracking and monitoring infants, which firstly utilizes a three-frame difference method to detect movement regions, and then utilizes a skin color detection method to extract smaller candidate regions including faces, hands and other parts, based on which the AdaBoost face detection algorithm is adopted to carry out face detection. The detected face is matched by using an active shape model, and the contour of the face to be detected is extracted; and the distance d between the middle point of the central connecting line of eyes of the contour and the geometric center of the chin and the area s of the surrounding region of the nose contour are calculated, and a linear resolution analysis method is utilized to find out the best projection direction w capable of distinguishing between infant faces and non-infant faces. The distance d and the area s are projected in the direction w, the distances between the projected points and two similar central connecting lines are used for judging whether the target is an infant or not, if so, the infant is used as a tracking object. In addition, the invention provides an embedded realization system for actively tracking infants according to the method.
Description
Technical field:
The present invention relates to the Image Information Processing field, particularly a kind of family expenses infant initiatively follows the tracks of the method and the embedded hardware thereof of monitoring and realizes system.
Background technology:
The intelligent video treatment technology is constantly expanded in the application in monitoring field at present, existing video monitor system at infant in the domestic environment, generally all be to use camera to be connected to network, perhaps be directly connected to network in the mode of web camera independently by PC.These equipment generally all have measuring abilities such as sound detection.Though what have has a The Cloud Terrace, can be by preset point or user rotate the infant of client (as browser) hand-guided The Cloud Terrace in moving in advance.But existing this type of family expenses monitor system has following shortcoming in practice: the user can not supervise the infant in client for a long time, and camera can only be guarded the infant in the place of fixed viewpoint, therefore can not be automatically the infant in the moving in the family be carried out real-time tracking and take, can't reach intelligence and nurse and the warning purpose.
Existing other supervisory control system of technical grade, for example open source literature number mainly utilizes a main camera to obtain global scene information for the Chinese invention patent of CN1658570A (theming as " the intelligent-tracking supervisory control system with multiple-camera "), carry out target detection and track algorithm, utilize the information-driven that obtains target to rotate tracking and the details demonstration of realization then target from video camera.But under the domestic. applications background, if use principal and subordinate's video camera to come the monitoring to the infant, two video cameras and a The Cloud Terrace need be installed at home, this difficulty that can cause monitoring to deploy troops on garrison duty strengthens, while also increased cost greatly, and uneconomical also inapplicable.
And existing supervisory control system towards domestic. applications, number whether the someone enters for the utility model patent of CN201262766Y (theming as " multifunction intelligent household supervisory control system ") utilizes the perception of human-body infrared sensing device as open source literature, when the people entered, intelligent early-warning final drive camera was recorded a video.But its camera does not load The Cloud Terrace causes it can not multi-angle to rotate, nor possesses automatically in scene identification and choose the infant and to its function of following the tracks of.
Summary of the invention:
In view of above-mentioned technical problem; for making the user by not possessing the infant's of sense of self-protection safe condition in the supervisory control system real-time tracking family; accomplish timely strick precaution and monitoring to it, primary and foremost purpose of the present invention provides a kind of monitoring method that utilizes supervisory control system to realize initiatively following the tracks of the infant.This inventive method can be applicable in the home environment active that does not possess sense of self-protection infant target be followed the tracks of and monitoring; improve the specific aim and the automatic capability of existing home videos monitor system, help father and mother to accomplish the timely strick precaution and the safe condition of infant's fortuitous event are guarded.
The technical scheme of foregoing invention method is: after analog video signal collection and A/D conversion, this vision signal is carried out algorithm process: extract the moving region by continuous three frame frame difference methods, adopt the Face Detection model filter to fall non-face zone in the moving region again, and in the candidate region that dwindles, use the AdaBoost algorithm to carry out people's face and detect; Use active shape model to mate to detected people's face, extract the profile of people's face to be detected, and obtain the area s apart from d and nose profile enclosing region of this profile the inside eye center line intermediate point, and the best projection direction w that utilizes linear resolved analysis method to find out to distinguish infant's face and non-infant's face to chin profile geometric center.D and s are carried out projection on the w direction, and according to o'clock judging whether infant of target to be detected after the projection to the distance of two class lines of centres, if target is the infant, then with it as tracing object; Above-mentioned algorithm process is utilized pid algorithm and PELCO D agreement after finishing again, and the control The Cloud Terrace rotates up and down, thereby realizes the active real-time tracking monitoring to the infant automatically, strengthens the initiative and the practicality of existing household video monitor system greatly.
In addition, according to said method, the invention allows for a kind of embedded hardware of initiatively following the tracks of the infant of above method that uses and realize system.This system has realized automatic detection, location and tracking to infant in the home environment by modules such as video acquisition, AD conversion, FVID interface and cradle head control.The concrete structure of this system is as follows, comprising:
One is used for video input signals is comprised that three frame difference methods, skin color segmentation, the detection of people's face, statistical shape model coupling, infant's target identification and PID control the master control processing unit of these algorithm process, thereby the result of this processing unit after according to algorithm process sends action command to The Cloud Terrace and adjusts the monitoring visual angle, and the vision signal that indicates monitored object mark having after will handling is simultaneously exported by display unit demonstration in real time or by output interface;
One is connected with the AD modular converter is used for taking and generate the CMOS photographing module of the analog video signal of family's scene;
One analog video signal with the CMOS photographing module is converted into the AD modular converter that digital signal sends the master control processing unit simultaneously to;
One when connecting the operation of master control processing unit storage system code and the random access memory module of interim video data;
One by connecting the read-only memory module of master control processing unit saved system start-up code and system-program code;
One is mounted with the CMOS photographing module, and the action command that sends according to the master control processing unit is regulated real-time tracking is carried out at shooting view and visual angle to target object The Cloud Terrace.
In the such scheme, described master control processing unit (DM6437) comprises video interface module (CCDC controller), EMIF interface module, algorithm processing module, Online Video display module (OSD), video encoding module (VENC) and at least more than one DAC interface; Described video interface module is connected with the AD modular converter, receives by the digital video signal after the AD module processing; Described algorithm processing module is according to the digital video signal that receives, extract the moving region by continuous three frame frame difference methods, adopt the Face Detection model filter to fall non-face zone in the moving region again, and in the candidate region that dwindles, use the AdaBoost algorithm to carry out people's face and detect; Use active shape model to mate to detected people's face, extract the profile of people's face to be detected, and obtain the area s apart from d and nose profile enclosing region of this profile the inside eye center line intermediate point, and the best projection direction w that utilizes linear resolved analysis method to find out to distinguish infant's face and non-infant's face to chin profile geometric center.D and s are carried out projection on the w direction, and according to o'clock judging whether infant of target to be detected after the projection to the distance of two class lines of centres, if target is the infant, then with it as tracing object; Algorithm processing module is according to judging that structure sends action command by the UART interface module to The Cloud Terrace and regulates camera angle then, thereby realizes under family's monitoring scene automatic detection and location to infant's target; And the digital video signal that will handle simultaneously sends the real-time demonstration that the Online Video display module carries out video image to, or via behind the video compression coding of video encoding module with this digital video signal output.
In the such scheme, described system also is provided with a multimedia coding module that is used for realizing audio/video coding, this multimedia coding module will be sent to the strange land by the Internet by the digital video signal that the AD modular converter transmits by connecting I2C bus and ethernet transceiver; Perhaps the digital video signal that algorithm processing module was handled is sent to the strange land by the Internet; Described multimedia coding module can adopt multimedia association to handle the C627 chip.
In the such scheme, described CMOS photographing module specifically adopts the CMOS130 camera.Described random access memory module is connected with the EMIF interface of master control processing unit, the 128M SDRAM of employing.Described read-only memory module is connected with the EMIF interface of master control processing unit, adopts the 64M Flash chip of NOR type.
In the such scheme, described system also is provided with a light-emitting diode display, and this display is connected with the DAC interface of master control processing unit, adopts JV-M50D model display.Described system adopts the commutator transformer power supply of a 5V, and the voltage device of this 5V produces three kinds of voltages and gives different power devices, and wherein: 1.2V gives the DSP kernel, and 3.3V gives the I/O port of DSP, and 1.8V gives internal memory.
In the such scheme, described The Cloud Terrace adopts ST-SP8206PS model The Cloud Terrace, is connected with the UART interface module of algorithm processing module by the RS485 interface.Described cradle head control instruction sends to The Cloud Terrace according to PELCO D agreement, initiatively adjusts the direction and the position of camera by the rotation of control The Cloud Terrace.
In the such scheme, described UART interface module adopts the SC16C550 chip, and receiver that it is inner and transmitter respectively have the FIFO of 16B, can handle the serial signal up to 3Mbps speed.
In the such scheme, described system obtains video data by the preview engine from the CMOS photographing module, and is translated into the YUV422 form; Provide the DAC interface of 4 54M by the VENC encoder, NTSC, PAL, S-Video and the output of YPbPr video are provided, also provide 24 digital video to output to rgb interface simultaneously.
The main running of foregoing invention system is: at first obtain analogue video signal, utilize the AD conversion chip to be translated into YUV signal again, vision signal is input to 6437 chips carries out the intelligent algorithm processing, to choose infant's target of tracking automatically, utilize pid algorithm and PELCO D agreement again, the control The Cloud Terrace rotates and makes camera aim at the infant's target in moving all the time, thereby realize active real-time tracking monitoring, strengthen the initiative and the practicality of existing household video monitor system greatly the infant.
Comprehensive, the inventive method of above-mentioned proposition is compared with existing correlation technique with the realization system, has following conspicuous outstanding character characteristics and remarkable advantage:
1, overcome the shortcoming that existing monitor system towards family can not automatic distinguishing infant target.
2, existing supervisory control system at family expenses also all is simple web camera, also have following shortcoming even loaded the web camera of The Cloud Terrace: the user must long-rangely manually control The Cloud Terrace by the web mode and rotate, and does not have intelligent target detection and following function.
3, cmos sensor places separately on the The Cloud Terrace, and video frequency processing chip and peripheral hardware are independent of The Cloud Terrace and cmos sensor, to reduce the The Cloud Terrace load, improves The Cloud Terrace reaction speed and useful life.
4, The Cloud Terrace equipment is connected with the UART interface of video frequency processing chip by the RS485 interface, can realize the moving of 8 directions of cmos sensor (on, upper right, the right side, the bottom right is descended, the lower-left is a left side, upper left) by the PELCO-D agreement, and realization is to infant's track up.
5, be convenient control The Cloud Terrace rotation direction, the The Cloud Terrace position is set as 9 zones that are positioned at even size, and algorithm only needs to judge that current goal drops on that zone, carries out move (zone line does not move) of corresponding 8 directions then according to sequence number.
The present invention has overcome traditional domestic monitoring system needs two cameras to carry out the shortcomings of initiatively following the tracks of to target, uses a camera to realize detection and active tracking and system to infant's target in conjunction with The Cloud Terrace and three frame difference methods etc.
Description of drawings:
Further specify the present invention below in conjunction with the drawings and specific embodiments.
Fig. 1 initiatively follows the tracks of the monitoring method schematic diagram for the described infant of the inventive method.
Fig. 2 is facial contour index point and feature s and d schematic diagram.
Fig. 3 (a)-Fig. 3 (d) is for being used to train the facial contour sample schematic diagram of statistical shape model.
Fig. 4 (a) is the schematic diagram to 8 unjustified people's face shapes.
Fig. 4 (b) is the schematic diagram of the later people's face shape of alignment.
Fig. 5 is a system configuration schematic diagram of the present invention.
Fig. 6 is three frame difference method flow charts.
Fig. 7 is a video subregion schematic diagram.
Fig. 8 is a The Cloud Terrace PID control flow chart.
Embodiment:
For technological means, creation characteristic that the present invention is realized, reach purpose and effect is easy to understand, below in conjunction with concrete diagram, further set forth the present invention.
(1) the family expenses infant initiatively follows the tracks of the method for monitoring;
Referring to Fig. 1, below be concrete steps:
(1) video acquisition process: system is from the analog video signal of camera collection PAL or TSC-system formula.
(2) A/D transfer process: be converted to the digital video letter of YUV422 form by the TVP5146 chip, vision signal is stored among the SDRAM, by the FVID interface data that collect is conducted interviews, and it is sent into carry out concrete algorithm process in the DM6437 chip.The primary signal resolution of system acquisition is 640 * 480, for improving efficiency of algorithm, is translated into 320 * 240 resolution.
(3) moving target detects: initiatively follow the tracks of the influence that The Cloud Terrace rotates for adapting to, use three frame frame difference methods to detect the moving region.Respectively adjacent three two field pictures are carried out Gauss's smoothing denoising, do continuous frame difference method again, 2 frame difference image results that obtain are carried out threshold process, obtain bianry image.Respectively two frame difference bianry images are carried out elder generation's corrosion and expansion process, again the two frame bianry images that obtained by top morphology algorithm are carried out and computing.The image that obtains with computing is carried out profile extract, profile institute area surrounded is the result that moving target detects.
(4), on the basis of three frame frame difference method results, extract the littler object candidate area at positions such as comprising people's face and hand according to complexion model for dwindling processing region.After original image being carried out YUV → RGB → YCrCb complexion model transformation, can obtain a image based on the YCrCb color model, choose appropriate C r and Cb threshold value, being divided into area of skin color and non-area of skin color on the image.
(5) adopt AdaBoost people's face detection algorithm in area of skin color, to carry out people's face and detect, thus the region of search that reduces people's face detection algorithm, and the speed and the accuracy of raising algorithm more help the real-time processing of embedded hardware.When Adaboost is used for the detection of people's face, from people's face, extracting a large amount of one dimension simple feature. these simple feature all have certain people's face and non-face differentiation. and final system uses thousands of one dimension simple classification devices, combining reaches good classifying quality, and speed also can satisfy the requirement of real-time.
(6) according to people's face testing result, by active shape model people's face is mated, extract corresponding profile information, corresponding 58 index points of each profile are formed one 116 vector of tieing up.
For setting up active shape model, need carry out index point to the facial image in the training sample and demarcate, as shown in Figure 2, index point is 58.Totally 7 sections of the profiles of determining by index point comprise: facial contour (13 index points), left eye profile (8 index points), right eye profile (8 index points), left eyebrow profile (5 index points), right eyebrow profile (5 index points), face outline (8 index points) and nose profile (11 index points).The index point choosing method is: at first, consider the key point that with the naked eye can directly offer an explanation out, as the canthus and the corners of the mouth; Secondly, other the index point of evenly distributing of between these key points, will trying one's best; At last, the distribution density of index point will be considered follow-up embedded hardware enforcement demand, and density is crossed conference increases workload and the reduction arithmetic speed that index point is demarcated, and the too small meeting of density can't reach desirable effect.
The facial image X that demarcates for a width of cloth
i, can come it is carried out representing of shape with the x and the y coordinate position of 58 index points:
X
i＝[x
i1，y
i1，x
i2，y
i2，…，x
i58，y
i58]
T
Wherein, facial contour partly is [x
I1, y
I1..., x
I13, y
I13]
TThe left eye profile is [x
I14, y
I14..., x
I21, y
I21]
TThe right eye profile is [x
I22, y
I22..., x
I29, y
I29]
TLeft side eyebrow profile is [x
I30, y
I30..., x
I34, y
I34]
TRight eyebrow profile is [x
I35, y
I35..., x
I39, y
I39]
TThe face profile is [x
I40, y
I40..., x
I47, y
I47]
TThe nose profile is [x
I48, y
I48..., x
I58, y
I58]
TThe shape of the N width of cloth uncalibrated image that is used to train just can be used training set { X like this
i: i=1 ... N} represents.Fig. 3 (a)-Fig. 3 (d) is people's face shape training sample of part.
For demarcating good people's face shape, owing to there are 3 differences: 1) the absolute position difference of its facial image of living in; 2) difference of facial image size; 3) difference of direction.Therefore, directly training shapes is set up the rule that statistical shape model can not truly reflect the conversion of people's face shape, training shapes need be carried out shape alignment (shape alignment) on approximate meaning.Everyone face shape is selected suitable translation, convergent-divergent and rotation, allow training shapes be arranged in same comparable cartesian coordinate system, shape and the difference minimum between the average shape (under the least square meaning) after the feasible alignment, wherein average shape can obtain from whole training sample centralized calculation.
The process of shape alignment is: at first, choose a comparatively initial sample of desired shapes conduct, make all alignment with it of other all shapes; Then, the average shape after the alignment that calculates is carried out normalization handles, and with the average shape after the normalization as initial sample; At last, other are alignd before this good shape is alignd with new initial sample.Repeat this alignment procedure, up to adjacent twice average shape difference less than a threshold value.
In the alignment procedure of people's face shape,, need all that in the process of iteration each time average shape is done normalization and handle for the situation that shape increases gradually or reduces appears in the average shape that prevents each circulation in iterative process.Concrete grammar is: keeping the distance between certain 2 in the uncalibrated image is a certain constant, all makes average shape be positioned at a certain fixing angle at every turn, and the translation average shape is to a certain fixing position.
The result of Fig. 4 (a) for 8 unjustified people's face shape Fig. 4 (b) are alignd.As seen from the figure, through form fit, the distribution of the index point that each is corresponding is concentrated more and is reasonable.After shape sample in training set all aligns, just can set up the statistical shape model of people's face on this basis.
After people's face shape alignment that training sample is concentrated, can obtain the shape of N alignment
Everyone face shape is provided by the x and the y coordinate of 58 index points, and the average shape of these training samples is
Its covariance matrix C is:
People's face shape of alignment is 116 dimensions (2 * 58).If these shapes are plotted in the space of 116 dimensions, their variations on some direction will be greater than other directions, and not necessarily the reference axis with original is consistent for these directions certainly.These directions be on earth which and between them relative significance level can decompose by quadrature and obtain, be i.e. solving equation covariance matrix:
CP
i＝λ
iP
i
Can obtain characteristic value (λ by finding the solution
1, λ
2..., λ
116) with and characteristic of correspondence vector (P
1, P
2... P
116).Owing to, comprised more shape change information with the direction of the characteristic vector of being correlated with corresponding to the big variance of training data than big characteristic value.Therefore can use corresponding to the characteristic vector of bigger characteristic value and come approximate representation people's face shape vector arbitrarily.K maximum characteristic value satisfies before choosing:
Therefore k maximum characteristic value characteristic of correspondence vector P=(P before can obtaining
1, P
2..., P
k), any one people's face shape X can be by approximate being expressed as:
Wherein, b is the form parameter of people's face shape vector.It can be obtained by top formula:
On behalf of the difference of people's face shape, the difference of b change.Therefore people's face shape data set of a two dimension just can be similar to the statistical shape model that only has only a parameter b, and the value that changes b within the specific limits can generate rational new person's face shape sample.
In active shape model, near the image local feature each index point of local gray level model representation.Generally adopt the differential of the gradation of image value of perpendicular shape line to represent, and carry out normalization, can remove the influence of illumination variation so to a certain extent according to integral value.
In detected each human face region of AdaBoost algorithm, the original shape model that trains is placed on this regional center, the yardstick of original shape, direction and displacement parameter can be estimated according to the result that AdaBoost people's face detects.Behind the initial value of people's face shape model, people's face active shape model can utilize the gray feature of profile to carry out iteration when given.In each step iteration, change the position and the shape of current model by adjusting relevant parameter, and then produce new model instance, finally finish the coupling of model and test pattern profile.Iterative algorithm is: at first, the point that mates most with this gray-scale statistical model is found in the zone of each index point in the search facial image in the zone, obtains new people's face shape.Secondly, according to the displacement of index point, the attitude parameter in the calculating people face statistical shape model and the variation of form parameter.At last, upgrade attitude parameter and form parameter, get back to algorithm and begin the place, up to the upper limit threshold of restraining or reach iterations.
(7) utilize the area s apart from d and nose profile enclosing region of the eye center line mid point of the profile of obtaining to the facial contour geometric center, with of the input of these two features, find out the best projection direction w that can distinguish infant's face and non-infant's face as linear resolved analysis.Two feature d of expression target to be detected and the point of s are carried out projection on the w direction, and judge whether infant of target to be detected according to the distance of this distance-like line of centres.
Generally, the organ of infant face, mainly be that eyes and eyebrow position are relatively low, open wiring is near face's central region, and nose is grown up little usually and weak point, according to above analysis, this aspect utilizes statistical shape model to obtain the area s apart from d and nose profile institute enclosing region that two eyes are wired to the geometric center point of chin profile, and these two features are analyzed.As shown in Figure 2.
For feature d, d is more little, and expression eyes line is short more apart from the distance of chin, and just open wiring is low more, and people's face approaches infant's face more.Character pair s, nose are short more more little, and the area s that will cause the nose profile to surround is more little, and people's face is more near infant's face.Therefore these two features that can utilize people's face realize detection to infant's target in conjunction with linear resolved analysis.
The main thought of linear resolved analysis is: through a Linear Mapping sample data is mapped to feature space, makes with nearer between the quasi-mode sample apart from each other between the different mode sample apart.In feature space, similar sample is intensive as much as possible, and inhomogeneous sample separates as much as possible, promptly wishes: dispersion is more little in total class, and dispersion is the bigger the better between class.
(8) target following process: mainly comprise according to PELCO D agreement and PID control method The Cloud Terrace is carried out the rotation of different directions so that infant's target all is positioned at the middle section of video pictures all the time.
(2) the family expenses infant initiatively follows the tracks of the embedded hardware realization system of monitoring;
As shown in Figure 5, mainly comprise DM6437 chip, analog cmos camera, AD conversion chip, the 128M size of TI SDRAM, 64M size Flash ROM, UART chip, multimedia coding chip C627, light-emitting diode display and The Cloud Terrace.
The DM6437 chip, function in system is to realize the main algorithm of video input signals is handled, comprise: functions such as three frame difference methods, skin color segmentation, the detection of people's face, statistical shape model coupling, infant's target identification, PID control are the core of native system.
CMOS camera, the function in system are the analog video signals that generates family's scene, and it is connected with the AD conversion chip.Specifically can adopt CMOS130 camera (ov9650 chip).
The AD conversion chip, the function in system is that analog video signal is converted into digital signal, it is connected with the Video IN interface of the CCDC controller of DM6437 development board.Specifically can adopt the TVP5146M2 chip of TI company, it supports Composite or S Video, sampling precision can reach 10 bits, output format is supported CCIR-656 and BT656, analog signal can be converted into that 8/16 bit strip is capable, the YCbCr 4:2:2 digital video frequency flow of field sync signal, be input among the DSP again and handle.
Code when random-access memory (ram), the function in system are the storage system operation and interim video data are connected with the EMIF interface of DM6437 development board.Specifically can adopt the 128M SDRAM of MICRON (Micron Technology).
Read-only memory (ROM), the function in system are saved system start-up code and system-program code, are connected with the EMIF interface of DM6437 development board.Specifically can adopt the 64M Flash chip of the NOR type of Spansion company.
Multimedia coding chip, the function in system are to realize the audio/video coding function, are connected with the I2C bus.Specifically can adopt the multimedia association of the many Microtronic A/S of intelligence to handle the C627 chip.
Light-emitting diode display, the function in system are the on-the-spot results that the infant is followed the tracks of that shows, it is connected with the DAC D interface of DM6437 development board.Specifically can adopt the JV-M50D model display of the good company of kumquat.
The Cloud Terrace, the function in system are the results according to infant's target detection, rotate the CMOS camera target is carried out real-time tracking, and it is connected with the UART interface of video frequency processing chip by the RS485 interface.Specifically can adopt the prompt company in Ruian ST-SP8206PS model The Cloud Terrace.
Except with upper-part, the present invention also comprises some general peripheral hardwares: native system adopts the commutator transformer power supply of a 5V.Voltage device by this 5V produces three kinds of voltages to different power devices, wherein: 1.2V gives the DSP kernel, and 3.3V gives the I/O port of DSP, and 1.8V gives internal memory.The UART chip is selected SC16C550 for use, and receiver that it is inner and transmitter respectively have the FIFO of 16B, can handle the serial signal up to 3Mbps speed.The cradle head control instruction sends to The Cloud Terrace according to PELCO D agreement, initiatively adjusts the direction and the position of camera by the rotation of control The Cloud Terrace.Plagiarize for anti-software, can adopt the SAM chip as ciphering unit.Can add mixed-media network modules mixed-media in addition, mainly finish Network Transmission task, video compression, the initialization of protocol stack is provided with DHCP, functions such as Web page loading.
The video front VPFE of system obtains video data by the preview engine from cmos sensor, and is translated into the YUV422 form.Provide the DACs of 4 54M by the VENC encoder of video rear end VPBE, NTSC, PAL, S-Video and the output of YPbPr video are provided, also provide 24 digital video to output to rgb interface simultaneously.
The video output of system has 4 output DACs and different outputting standard cooperations, supports multiple outputting standard.DACs can support composite video, component (aberration) video or RGB by programming.Can obtain S-video output by connector P1, this connector is driven by DACs B and C.
After the system start-up, load boot, boot copies the intelligent video handling procedure the SDRAM to from FLASH and carries out, program based on the FVID interface exploitation realizes the relevant Intelligent treatment algorithms such as detection and tracking of DM6437 to the infant's target under the domestic environment, and in real time target is followed the tracks of and taken according to its positional information driving The Cloud Terrace.The major technique feature of intelligent algorithm module is:
1, moving target detects:
(frame0, frame1 frame2) carry out Gauss's smoothing denoising, again the image of continuous adjacent 3 frames are continuous frame difference method (fram1-frame0 to the image of adjacent three frames respectively; Frame2-frame1), 2 frame difference image results that obtain are carried out threshold process, obtain bianry image.To two frame difference bianry images corrosion earlier 1 to 2 time, expand again 2 to 10 times respectively, to eliminate noise and fill area inside holes.The two frame bianry images that obtained by top morphology are carried out and computing.Extracting with the basic enterprising road wheel of computing is wide, profile institute area surrounded is the result that moving target detects.Idiographic flow as shown in Figure 6.
2, complexion model dwindles processing region:
On the basis of three frame frame difference method results, extract the littler object candidate area at positions such as comprising people's face and hand according to complexion model.
3, according to people's face testing result, people's face is mated, extract corresponding profile information by active shape model.
At first will set up people's face statistical shape model, because will obtain feature s and d, only need obtain the information of human eye, nose and facial contour, therefore, this realization system chooses four sections profiles and sets up statistical shape model, and they are: 1) facial contour part ([x
I1, y
I1..., x
I13, y
I13]
T); 2) left eye profile ([x
I14, y
I14..., x
I21, y
I21]
T); 3) right eye profile ([x
I22, y
I22..., x
I29, y
I29]
T); 4) nose profile ([x
I48, y
I48..., x
I58, y
I58]
T).In the training of statistics shape, every face is demarcated (comprising 13 points of facial contour, 8 points of left eye profile, 8 points of right eye profile and 11 points of nose profile) with 40 index points, the length of local gray level model is made as 7 (3 of every sides), and the search width is made as 2 pixels.In matching process, adopt the signature search of multiresolution, divide 4 layers of realization.In search procedure, at first under low resolution, search for, rise to step by step under the higher leveled resolution then and search for.Under low resolution, mainly consider the Global Information of image, the scope of search is bigger, obtains comparatively changing higher level again over to behind the ideal matching result, and the scope that reduces to search for is mated.
By active shape model, obtain people's face left eye profile central point L coordinate (xlefteye, ylefteye), the coordinate of right eye profile central point R (xrighteye, yrighteye) and the coordinate C of nose profile central point (xnose ynose) is respectively:
The coordinate at eyes that utilization obtains and nose profile center, can carry out the work of people's face normalization: definition standard faces size is 64 * 64, and facial image is the center with nose profile mid point.Two profile centers and distance L R normalization be 28, the distance of nose profile center C to two a profile line of centres LR is 8.Can realize the normalization of people's face by L, R and C point.After obtaining people's face of normalization, again according to obtaining facial contour [x
I1, y
I1..., x
I13, y
I13]
TCarry out the removal of people's face background.
4, infant's target identification:
It is one two classification problem that the inspection of infant's target can be regarded as: one is adult's face, and one is infant's face.Face to be identified can be regarded 1 x in the two-dimensional space as, both direction represent respectively d and s value (x=[d s]
T), therefore, S
bAnd S
wCan be expressed as:
S
b＝(m
baby-m
nobaby)(m
baby-m
nobaby)
T
Can obtain the w of the direction of maximum J (w):
w＝S
w -1(m
baby-m
nobaby)
By w, can be converted into the feature space of one dimension by d and the two-dimensional space that the s direction is opened:
babyvalue＝w
Tx
The babyvalue value is the degree that expression people's face to be identified approaches infant's feature, and as Self-class, 50 non-infant's faces are as Imposter Group 50 infant infant's faces.Obtain projecting direction w by the LDA method, the threshold value of distinguishing non-infant's face and infant's face is taken as the mid point of the individual line of centres of two classes, obtains threshold value k, if babyvalue is greater than k, can judge that target is the infant, and the sizes values of definite degree and babyvalue is directly proportional.In this embodiment, because the babyvalue value that obtains can judge that greater than threshold value k people's face to be identified is non-infant's target.
If detected infant's target in the scene, system are that tracking target driving The Cloud Terrace is followed the tracks of with this zone promptly, in the process of following the tracks of, no longer call above-mentioned infant's recognizer, and keep tracking this moving region.Do not detect infant's target in scene, system can choose the zone of area maximum in the zone that three frame difference methods extract and follow the tracks of, and can reach the real-time monitoring to other targets of family this moment, also can be used for general family's scene video monitoring and use.
5, video subregion and The Cloud Terrace PID control:
Referring to Fig. 7, in order to simplify the following calculation complexity, video is divided into 9 districts, if target is positioned at 9 districts, then move the The Cloud Terrace bottom right, 8 districts: move down; 7 districts: move the lower-left; 6 districts: move right; 5 districts: remain unchanged; 4 districts: be moved to the left; 3 districts: upper right moving; 2 districts: move up; 1 district: upper left moving.
Simultaneously in order to guarantee that The Cloud Terrace can move fast and accurately, is numbered 5 zone thereby make target drop on.This need eliminate the shake and the overshoot phenomenon of The Cloud Terrace.System uses pid algorithm that The Cloud Terrace is accurately controlled.Wherein P (Proportional: ratio) the rapid response error of control, thereby reduce steady-state error, but it can not eliminate steady-state error, causes system stability when P chooses conference.(Intergration: integration) Kong Zhi effect is: when the error existence of system, integral controller is accumulation constantly, and the output controlled quentity controlled variable is eliminated this error, makes that systematic error is 0, but integral action can cause system's appearance concussion too by force in I control.Therefore (Differentiation: differential) control can reduce overshoot, overcomes reforming phenomena to use D.
Integral separation PID controling algorithm need be set integration and separate threshold epsilon, utilizing The Cloud Terrace to rotate in the tracking infant object procedure, when | e (k) | during＞ε, be that difference e (k) between the position of The Cloud Terrace current location and infant's target is when big, adopt PD control, to reduce overshoot, system is had than fast-response; As | e (k) | during≤ε, promptly difference hour then adopts PID control, to guarantee that The Cloud Terrace can accurately put in place.(e is: the difference of The Cloud Terrace current location and target location).The controlled quentity controlled variable of system is e (k), and output variable is U (k), and control flow as shown in Figure 8.
More than show and described basic principle of the present invention and principal character and advantage of the present invention.The technical staff of the industry should understand; the present invention is not restricted to the described embodiments; that describes in the foregoing description and the specification just illustrates principle of the present invention; without departing from the spirit and scope of the present invention; the present invention also has various changes and modifications, and these changes and improvements all fall in the claimed scope of the invention.The claimed scope of the present invention is defined by appending claims and equivalent thereof.
Claims (10)
1. method that the infant initiatively follows the tracks of monitoring, it is characterized in that, after analog video signal collection and A/D conversion, this vision signal is carried out algorithm process: extract the moving region by continuous three frame frame difference methods, adopt the Face Detection model filter to fall non-face zone in the moving region again, and in the candidate region that dwindles, use the AdaBoost algorithm to carry out people's face and detect; Use active shape model to mate to detected people's face, extract the profile of people's face to be detected, and obtain the area s apart from d and nose profile enclosing region of this profile the inside eye center line intermediate point, and the best projection direction w that utilizes linear resolved analysis method to find out to distinguish infant's face and non-infant's face to chin profile geometric center; D and s are carried out projection on the w direction, and according to o'clock judging whether infant of target to be detected after the projection to the distance of two class lines of centres, if target is the infant, then with it as tracing object; Above-mentioned algorithm process is utilized pid algorithm and PELCO D agreement after finishing again, and the control The Cloud Terrace rotates up and down, thereby realizes the active real-time tracking monitoring to the infant automatically.
2. initiatively follow the tracks of the method for monitoring according to the infant of claim 1, it is characterized in that, described method is specifically further comprising the steps of:
(1) video acquisition process: system is from the analog video signal of camera collection PAL or TSC-system formula;
(2) A/D transfer process: be converted to the digital video letter of YUV422 form by the TVP5146 chip, vision signal is stored among the SDRAM, by the FVID interface data that collect is conducted interviews, and it is sent into carry out concrete algorithm process in the DM6437 chip; The primary signal resolution of system acquisition is 640 * 480, for improving efficiency of algorithm, is translated into 320 * 240 resolution;
(3) moving target detects: initiatively follow the tracks of the influence that The Cloud Terrace rotates for adapting to, use three frame frame difference methods to detect the moving region; Respectively adjacent three two field pictures are carried out Gauss's smoothing denoising, do continuous frame difference method again, 2 frame difference image results that obtain are carried out threshold process, obtain bianry image; Respectively two frame difference bianry images are carried out elder generation's corrosion and expansion process, again the two frame bianry images that obtained by top morphology algorithm are carried out and computing; The image that obtains with computing is carried out profile extract, profile institute area surrounded is the result that moving target detects;
(4), on the basis of three frame frame difference method results, extract the littler object candidate area at positions such as comprising people's face and hand according to complexion model for dwindling processing region; After original image being carried out YUV → RGB → YCrCb complexion model transformation, can obtain a image based on the YCrCb color model, choose appropriate C r and Cb threshold value, being divided into area of skin color and non-area of skin color on the image;
(5) adopt AdaBoost people's face detection algorithm in area of skin color, to carry out people's face and detect, thus the region of search that reduces people's face detection algorithm, and the speed and the accuracy of raising algorithm more help the real-time processing of embedded hardware; When Adaboost is used for the detection of people's face, from people's face, extract a large amount of one dimension simple feature; These simple feature all have certain people's face and non-face differentiation, and final system uses thousands of one dimension simple classification devices, and combining reaches good classifying quality, and speed also can satisfy the requirement of real-time;
(6) according to people's face testing result, by active shape model people's face is mated, extract corresponding profile information, corresponding 58 index points of each profile are formed one 116 vector of tieing up;
(7) utilize the area s apart from d and nose profile enclosing region of the eye center line mid point of the profile of obtaining to the facial contour geometric center, with of the input of these two features, find out the best projection direction w that can distinguish infant's face and non-infant's face as linear resolved analysis; Two feature d of expression target to be detected and the point of s are carried out projection on the w direction, and judge whether infant of target to be detected according to the distance of this distance-like line of centres;
(8) target following process: mainly comprise according to PELCO D agreement and PID control method The Cloud Terrace is carried out the rotation of different directions so that infant's target all is positioned at the middle section of video pictures all the time.
3. initiatively follow the tracks of the realization system of monitoring according to a kind of infant of claim 1, it is characterized in that described system comprises:
One is used for video input signals is comprised that three frame difference methods, skin color segmentation, the detection of people's face, statistical shape model coupling, infant's target identification and PID control the master control processing unit of these algorithm process, thereby the result of this processing unit after according to algorithm process sends action command to The Cloud Terrace and adjusts the monitoring visual angle, and the vision signal that indicates monitored object mark having after will handling is simultaneously exported by display unit demonstration in real time or by output interface;
One is connected with the AD modular converter is used for taking and generate the CMOS photographing module of the analog video signal of family's scene;
One analog video signal with the CMOS photographing module is converted into the AD modular converter that digital signal sends the master control processing unit simultaneously to;
One when connecting the operation of master control processing unit storage system code and the random access memory module of interim video data;
One by connecting the read-only memory module of master control processing unit saved system start-up code and system-program code;
One is mounted with the CMOS photographing module, and the action command that sends according to the master control processing unit is regulated real-time tracking is carried out at shooting view and visual angle to target object The Cloud Terrace.
4. initiatively follow the tracks of the realization system of monitoring according to the infant of claim 3, it is characterized in that described master control processing unit comprises video interface module, EMIF interface module, algorithm processing module, Online Video display module, video encoding module and at least more than one DAC interface; Described video interface module is connected with the AD modular converter, receives by the digital video signal after the AD module processing; Described algorithm processing module is according to the digital video signal that receives, extract the moving region by continuous three frame frame difference methods, adopt the Face Detection model filter to fall non-face zone in the moving region again, and in the candidate region that dwindles, use the AdaBoost algorithm to carry out people's face and detect; Use active shape model to mate to detected people's face, extract the profile of people's face to be detected, and obtain the area s apart from d and nose profile enclosing region of this profile the inside eye center line intermediate point, and the best projection direction w that utilizes linear resolved analysis method to find out to distinguish infant's face and non-infant's face to chin profile geometric center; D and s are carried out projection on the w direction, and according to o'clock judging whether infant of target to be detected after the projection to the distance of two class lines of centres, if target is the infant, then with it as tracing object; Algorithm processing module is according to judging that structure sends action command by the UART interface module to The Cloud Terrace and regulates camera angle then, thereby realizes under family's monitoring scene automatic detection and location to infant's target; And the digital video signal that will handle simultaneously sends the real-time demonstration that the Online Video display module carries out video image to, or via behind the video compression coding of video encoding module with this digital video signal output.
5. initiatively follow the tracks of the realization system of monitoring according to the infant of claim 3, it is characterized in that, described system also is provided with a multimedia coding module that is used for realizing audio/video coding, this multimedia coding module will be sent to the strange land by the Internet by the digital video signal that the AD modular converter transmits by connecting I2C bus and ethernet transceiver; Perhaps the digital video signal that algorithm processing module was handled is sent to the strange land by the Internet; Described multimedia coding module can adopt multimedia association to handle the C627 chip.
6. initiatively follow the tracks of the realization system of monitoring according to the infant of claim 3, it is characterized in that, described CMOS photographing module specifically adopts the CMOS130 camera; Described random access memory module is connected with the EMIF interface of master control processing unit, the 128M SDRAM of employing; Described read-only memory module is connected with the EMIF interface of master control processing unit, adopts the 64M Flash chip of NOR type.
7. initiatively follow the tracks of the realization system of monitoring according to the infant of claim 3, it is characterized in that described system also is provided with a light-emitting diode display, this display is connected with the DAC interface of master control processing unit, adopts JV-M50D model display; Described system adopts the commutator transformer power supply of a 5V, and the voltage device of this 5V produces three kinds of voltages and gives different power devices, and wherein, 1.2V gives the DSP kernel, and 3.3V gives the I/O port of DSP, and 1.8V gives internal memory.
8. initiatively follow the tracks of the realization system of monitoring according to the infant of claim 3, it is characterized in that, described The Cloud Terrace adopts ST-SP8206PS model The Cloud Terrace, is connected with the UART interface module of algorithm processing module by the RS485 interface; The control command of described The Cloud Terrace sends to The Cloud Terrace according to PELCO D agreement, initiatively adjusts the direction and the position of camera by the rotation of control The Cloud Terrace.
9. initiatively follow the tracks of the realization system of monitoring according to the infant of claim 4, it is characterized in that, described UART interface module adopts the SC16C550 chip, and receiver that it is inner and transmitter respectively have the FIFO of 16B, can handle the serial signal up to 3Mbps speed.
10. initiatively follow the tracks of the realization system of monitoring according to the infant of claim 3, it is characterized in that described system obtains video data by the preview engine from the CMOS photographing module, and be translated into the YUV422 form; Provide the DAC interface of 4 54M by the VENC encoder, NTSC, PAL, S-Video and the output of YPbPr video are provided, also provide 24 digital video to output to rgb interface simultaneously.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010125536 CN101795400B (en) | 2010-03-16 | 2010-03-16 | Method for actively tracking and monitoring infants and realization system thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010125536 CN101795400B (en) | 2010-03-16 | 2010-03-16 | Method for actively tracking and monitoring infants and realization system thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101795400A true CN101795400A (en) | 2010-08-04 |
| CN101795400B CN101795400B (en) | 2013-03-27 |
Family
ID=42587791
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201010125536 Active CN101795400B (en) | 2010-03-16 | 2010-03-16 | Method for actively tracking and monitoring infants and realization system thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101795400B (en) |
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102148921A (en) * | 2011-05-04 | 2011-08-10 | 中国科学院自动化研究所 | Multi-target tracking method based on dynamic group division |
| CN102630033A (en) * | 2012-03-31 | 2012-08-08 | 彩虹集团公司 | Method for converting 2D (Two Dimension) into 3D (Three Dimension) based on dynamic object detection |
| CN102833465A (en) * | 2012-07-24 | 2012-12-19 | 武汉大千信息技术有限公司 | Criminal investigation video pretreatment method based on movement detection |
| CN102831408A (en) * | 2012-08-29 | 2012-12-19 | 华南理工大学 | Human face recognition method |
| CN102932635A (en) * | 2012-11-22 | 2013-02-13 | 南京明德软件有限公司 | Real-time image processing system and method |
| CN103108124A (en) * | 2012-12-28 | 2013-05-15 | 上海鼎为软件技术有限公司 | Image acquisition method, device and mobile terminal |
| CN103368135A (en) * | 2012-03-26 | 2013-10-23 | 鸿富锦精密工业（深圳）有限公司 | Children auxiliary system and method thereof |
| CN103391426A (en) * | 2013-07-10 | 2013-11-13 | 青岛歌尔声学科技有限公司 | Camera and security protecting method thereof |
| CN104217529A (en) * | 2014-08-27 | 2014-12-17 | 东北大学 | Intelligent infant monitoring system and method based on temperature, humidity and images |
| CN104361772A (en) * | 2014-11-17 | 2015-02-18 | 一统安易（北京）科技有限公司 | Self-service type life management and record toy and teaching aid based on infant |
| CN104732413A (en) * | 2013-12-20 | 2015-06-24 | 中国科学院声学研究所 | Intelligent individuation video advertisement pushing method and system |
| CN104967776A (en) * | 2015-06-11 | 2015-10-07 | 广东欧珀移动通信有限公司 | Photographing setting method and user terminal |
| CN105303150A (en) * | 2014-06-26 | 2016-02-03 | 腾讯科技（深圳）有限公司 | Method and system for implementing image processing |
| WO2016044996A1 (en) * | 2014-09-23 | 2016-03-31 | 王铭 | Household monitoring system |
| CN105809096A (en) * | 2014-12-31 | 2016-07-27 | 中兴通讯股份有限公司 | Figure labeling method and terminal |
| CN105960801A (en) * | 2014-02-03 | 2016-09-21 | 谷歌公司 | Enhancing video conferences |
| CN106227059A (en) * | 2016-10-08 | 2016-12-14 | 三星电子（中国）研发中心 | Intelligent home furnishing control method based on indoor threedimensional model and equipment |
| CN106507045A (en) * | 2016-11-07 | 2017-03-15 | 立德高科(北京)数码科技有限责任公司 | Wandered away the method and system of crowd, terminal, server platform with identification code positioning |
| CN107424258A (en) * | 2017-06-23 | 2017-12-01 | 深圳市盛路物联通讯技术有限公司 | A kind of control method for electronic lock and system based on Internet of Things |
| CN107480607A (en) * | 2017-07-28 | 2017-12-15 | 青岛大学 | A kind of method that standing Face datection positions in intelligent recording and broadcasting system |
| WO2018027973A1 (en) * | 2016-08-12 | 2018-02-15 | 魏羽婕 | Facial recognition verification based automatic-following assistant |
| CN108090908A (en) * | 2017-12-07 | 2018-05-29 | 深圳云天励飞技术有限公司 | Image partition method, device, terminal and storage medium |
| CN108366265A (en) * | 2018-03-08 | 2018-08-03 | 南京邮电大学 | A kind of distributed video side information generation method based on temporal correlation |
| CN109089077A (en) * | 2014-10-09 | 2018-12-25 | 中控智慧科技股份有限公司 | A kind of method remotely monitored and monitoring client |
| CN111369578A (en) * | 2020-02-25 | 2020-07-03 | 四川新视创伟超高清科技有限公司 | Intelligent tracking method and system for holder transaction |
| CN111640141A (en) * | 2020-05-20 | 2020-09-08 | 山东神戎电子股份有限公司 | Low-slow target positioning and tracking method based on thermal imaging |
| CN112232219A (en) * | 2020-10-19 | 2021-01-15 | 武汉理工大学 | Face recognition check-in system based on LBP (local binary pattern) feature algorithm |
| CN112784676A (en) * | 2020-12-04 | 2021-05-11 | 中国科学院深圳先进技术研究院 | Image processing method, robot, and computer-readable storage medium |
| CN113065454A (en) * | 2021-03-30 | 2021-07-02 | 青岛海信智慧生活科技股份有限公司 | High-altitude parabolic target identification and comparison method and device |
| US11129743B2 (en) * | 2016-02-09 | 2021-09-28 | MCE 5 Development | Set of nose-modelling instruments, production method and a method for modelling the nose |
| CN114554161A (en) * | 2022-03-08 | 2022-05-27 | 佛山市顺富元亨电子有限公司 | Baby monitor and method for alarming in dangerous area |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006006133A1 (en) * | 2004-07-07 | 2006-01-19 | Koninklijke Philips Electronics N.V. | Improvements in or relating to time-of-flight ranging systems |
| CN101232606A (en) * | 2007-01-23 | 2008-07-30 | 全成电子企业有限公司 | Wireless monitoring system and method |
| CN101572803A (en) * | 2009-06-18 | 2009-11-04 | 中国科学技术大学 | Customizable automatic tracking system based on video monitoring |
-
2010
- 2010-03-16 CN CN 201010125536 patent/CN101795400B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006006133A1 (en) * | 2004-07-07 | 2006-01-19 | Koninklijke Philips Electronics N.V. | Improvements in or relating to time-of-flight ranging systems |
| CN101232606A (en) * | 2007-01-23 | 2008-07-30 | 全成电子企业有限公司 | Wireless monitoring system and method |
| CN101572803A (en) * | 2009-06-18 | 2009-11-04 | 中国科学技术大学 | Customizable automatic tracking system based on video monitoring |
Cited By (41)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102148921A (en) * | 2011-05-04 | 2011-08-10 | 中国科学院自动化研究所 | Multi-target tracking method based on dynamic group division |
| CN102148921B (en) * | 2011-05-04 | 2012-12-12 | 中国科学院自动化研究所 | Multi-target tracking method based on dynamic group division |
| CN103368135A (en) * | 2012-03-26 | 2013-10-23 | 鸿富锦精密工业（深圳）有限公司 | Children auxiliary system and method thereof |
| CN102630033A (en) * | 2012-03-31 | 2012-08-08 | 彩虹集团公司 | Method for converting 2D (Two Dimension) into 3D (Three Dimension) based on dynamic object detection |
| CN102833465A (en) * | 2012-07-24 | 2012-12-19 | 武汉大千信息技术有限公司 | Criminal investigation video pretreatment method based on movement detection |
| CN102831408A (en) * | 2012-08-29 | 2012-12-19 | 华南理工大学 | Human face recognition method |
| CN102932635A (en) * | 2012-11-22 | 2013-02-13 | 南京明德软件有限公司 | Real-time image processing system and method |
| CN103108124A (en) * | 2012-12-28 | 2013-05-15 | 上海鼎为软件技术有限公司 | Image acquisition method, device and mobile terminal |
| CN103391426A (en) * | 2013-07-10 | 2013-11-13 | 青岛歌尔声学科技有限公司 | Camera and security protecting method thereof |
| CN103391426B (en) * | 2013-07-10 | 2016-09-21 | 青岛歌尔声学科技有限公司 | A kind of video camera and safety protection method thereof |
| CN104732413B (en) * | 2013-12-20 | 2017-11-21 | 中国科学院声学研究所 | A kind of intelligent personalized video ads method for pushing and system |
| CN104732413A (en) * | 2013-12-20 | 2015-06-24 | 中国科学院声学研究所 | Intelligent individuation video advertisement pushing method and system |
| CN105960801B (en) * | 2014-02-03 | 2020-02-07 | 谷歌有限责任公司 | Enhancing video conferencing |
| CN105960801A (en) * | 2014-02-03 | 2016-09-21 | 谷歌公司 | Enhancing video conferences |
| CN105303150A (en) * | 2014-06-26 | 2016-02-03 | 腾讯科技（深圳）有限公司 | Method and system for implementing image processing |
| CN104217529A (en) * | 2014-08-27 | 2014-12-17 | 东北大学 | Intelligent infant monitoring system and method based on temperature, humidity and images |
| WO2016044996A1 (en) * | 2014-09-23 | 2016-03-31 | 王铭 | Household monitoring system |
| CN109089077A (en) * | 2014-10-09 | 2018-12-25 | 中控智慧科技股份有限公司 | A kind of method remotely monitored and monitoring client |
| CN104361772A (en) * | 2014-11-17 | 2015-02-18 | 一统安易（北京）科技有限公司 | Self-service type life management and record toy and teaching aid based on infant |
| CN105809096A (en) * | 2014-12-31 | 2016-07-27 | 中兴通讯股份有限公司 | Figure labeling method and terminal |
| CN104967776A (en) * | 2015-06-11 | 2015-10-07 | 广东欧珀移动通信有限公司 | Photographing setting method and user terminal |
| US11129743B2 (en) * | 2016-02-09 | 2021-09-28 | MCE 5 Development | Set of nose-modelling instruments, production method and a method for modelling the nose |
| WO2018027973A1 (en) * | 2016-08-12 | 2018-02-15 | 魏羽婕 | Facial recognition verification based automatic-following assistant |
| CN106227059A (en) * | 2016-10-08 | 2016-12-14 | 三星电子（中国）研发中心 | Intelligent home furnishing control method based on indoor threedimensional model and equipment |
| CN106507045A (en) * | 2016-11-07 | 2017-03-15 | 立德高科(北京)数码科技有限责任公司 | Wandered away the method and system of crowd, terminal, server platform with identification code positioning |
| CN106507045B (en) * | 2016-11-07 | 2018-09-18 | 立德高科(北京)数码科技有限责任公司 | The method and system, terminal, server platform of the crowd that wanders away are positioned with identification code |
| CN107424258A (en) * | 2017-06-23 | 2017-12-01 | 深圳市盛路物联通讯技术有限公司 | A kind of control method for electronic lock and system based on Internet of Things |
| CN107424258B (en) * | 2017-06-23 | 2019-08-23 | 深圳市盛路物联通讯技术有限公司 | A kind of control method for electronic lock and system based on Internet of Things |
| CN107480607B (en) * | 2017-07-28 | 2020-04-07 | 青岛大学 | Method for detecting and positioning standing face in intelligent recording and broadcasting system |
| CN107480607A (en) * | 2017-07-28 | 2017-12-15 | 青岛大学 | A kind of method that standing Face datection positions in intelligent recording and broadcasting system |
| CN108090908A (en) * | 2017-12-07 | 2018-05-29 | 深圳云天励飞技术有限公司 | Image partition method, device, terminal and storage medium |
| CN108366265A (en) * | 2018-03-08 | 2018-08-03 | 南京邮电大学 | A kind of distributed video side information generation method based on temporal correlation |
| CN108366265B (en) * | 2018-03-08 | 2021-12-31 | 南京邮电大学 | Distributed video side information generation method based on space-time correlation |
| CN111369578A (en) * | 2020-02-25 | 2020-07-03 | 四川新视创伟超高清科技有限公司 | Intelligent tracking method and system for holder transaction |
| CN111369578B (en) * | 2020-02-25 | 2023-06-30 | 四川新视创伟超高清科技有限公司 | Intelligent tracking method and system for cradle head transaction |
| CN111640141A (en) * | 2020-05-20 | 2020-09-08 | 山东神戎电子股份有限公司 | Low-slow target positioning and tracking method based on thermal imaging |
| CN111640141B (en) * | 2020-05-20 | 2023-10-24 | 山东神戎电子股份有限公司 | Low-speed target positioning and tracking method based on thermal imaging |
| CN112232219A (en) * | 2020-10-19 | 2021-01-15 | 武汉理工大学 | Face recognition check-in system based on LBP (local binary pattern) feature algorithm |
| CN112784676A (en) * | 2020-12-04 | 2021-05-11 | 中国科学院深圳先进技术研究院 | Image processing method, robot, and computer-readable storage medium |
| CN113065454A (en) * | 2021-03-30 | 2021-07-02 | 青岛海信智慧生活科技股份有限公司 | High-altitude parabolic target identification and comparison method and device |
| CN114554161A (en) * | 2022-03-08 | 2022-05-27 | 佛山市顺富元亨电子有限公司 | Baby monitor and method for alarming in dangerous area |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101795400B (en) | 2013-03-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101795400B (en) | Method for actively tracking and monitoring infants and realization system thereof | |
| CN106598226B (en) | A kind of unmanned plane man-machine interaction method based on binocular vision and deep learning | |
| CN108154550B (en) | RGBD camera-based real-time three-dimensional face reconstruction method | |
| US9928605B2 (en) | Real-time cascaded object recognition | |
| Kliper-Gross et al. | Motion interchange patterns for action recognition in unconstrained videos | |
| CN105843386A (en) | Virtual fitting system in shopping mall | |
| US8855369B2 (en) | Self learning face recognition using depth based tracking for database generation and update | |
| WO2019157344A1 (en) | Real-time gesture recognition method and apparatus | |
| Zhang et al. | Code4d: color-depth local spatio-temporal features for human activity recognition from rgb-d videos | |
| CN107203745B (en) | Cross-visual angle action identification method based on cross-domain learning | |
| KR102441171B1 (en) | Apparatus and Method for Monitoring User based on Multi-View Face Image | |
| CN104969240B (en) | Method and system for image procossing | |
| CN110135277B (en) | Human behavior recognition method based on convolutional neural network | |
| CN103020655B (en) | A kind of remote identity authentication method based on single training image per person | |
| CN103020589A (en) | Face recognition method for single training sample | |
| Faisal et al. | Depth estimation from video using computer vision and machine learning with hyperparameter optimization | |
| CN112700568B (en) | Identity authentication method, equipment and computer readable storage medium | |
| CN110674751A (en) | Device and method for detecting head posture based on monocular camera | |
| CN103246878A (en) | Facial-recognition-based trial makeup system and trial makeup method thereof | |
| JP6727894B2 (en) | Processing device, processing method, and program | |
| CN103345625A (en) | Method and system for analyzing three-dimensional image | |
| van der Pol et al. | Head pose estimation for a domestic robot | |
| CN108537764A (en) | A kind of man-machine hybrid intelligent control loop | |
| CN108427922A (en) | A kind of efficient indoor environment regulating system | |
| Wang et al. | Adaptive visual tracking based on discriminative feature selection for mobile robot |
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 |