CN100576908C - Stereo intelligent camera apparatus based on omnibearing vision sensor - Google Patents
Stereo intelligent camera apparatus based on omnibearing vision sensor Download PDFInfo
- Publication number
- CN100576908C CN100576908C CN200710164441A CN200710164441A CN100576908C CN 100576908 C CN100576908 C CN 100576908C CN 200710164441 A CN200710164441 A CN 200710164441A CN 200710164441 A CN200710164441 A CN 200710164441A CN 100576908 C CN100576908 C CN 100576908C
- Authority
- CN
- China
- Prior art keywords
- clipping
- image
- vision sensor
- ball
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Studio Devices (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
A kind of stereo intelligent camera apparatus based on omnibearing vision sensor, comprise fixed support, be installed in the clipping the ball video camera on the fixed support and be used to control the video data microprocessor that the clipping the ball video camera is captured, described stereo intelligent camera apparatus also comprises the omnibearing vision sensor ODVS that is used to obtain comprehensive video data in the monitoring field, microprocessor carries out processing such as dynamic image understanding with the full-view video image that collects, obtain the object be concerned about in the monitoring field or the spatial positional information of incident, according to the visual field spatial mappings relation between ODVS and the clipping the ball video camera, the fixed position that indication clipping the ball video camera turns to hope fast carries out the part monitoring then.The invention provides a kind of multiple target automatic stereo intelligent camera apparatus that absorbs the monitored object target, realizes dynamic, intelligent in real time process monitoring in motion tracking, the big visual field of omni-directional visual of realizing based on omnibearing vision sensor.
Description
Technical field
The present invention relates to a kind of intelligent camera apparatus, especially the application of technology aspect stereo intelligent camera apparatus such as omnibearing vision sensor, clipping the ball video camera, computer vision, dynamic image understanding, Electromechanical Design.
Background technology
The clipping the ball video camera merges distinguishing features such as good to have fast rotational, presetting bit, tour track, coordinate function, automatic homing function, long burnt speed limit function, night vision function, good looking appearance and environment, just becoming the new lover of safe city supervision control engineering, in view of intelligent high-speed ball video camera has the incomparable advantage of other video cameras, in indoor and outdoor monitoring fields such as intelligent building monitoring, the security personnel of bank, long-distance education, market, airport, station, road, square, sub-district, campus, subway, obtained using widely.
Some intelligent high-speed ball video cameras can realize following the tracks of automatically single object at present, but can't realize but that under multiple target object situation multiple target is from motion tracking; Even when manual manipulation or fixed point patrol, still have temporal monitoring dead angle problem in addition; Especially when control clipping the ball video camera rotates, because can only control motions such as up/down/left/right, the position that the clipping the ball video camera is turned to hope that often can't be very fast is not known to control the clipping the ball video camera in other words and where is turned to, and causes the reduction of monitoring efficiency; In addition in actual applications, even video camera has been captured monitored object, in affirmation process afterwards because monitored object shared ratio on the whole video image is too small, and caused can't confirm this monitored object be who etc. problem.From the viewpoint of global monitoring and local monitoring, above-mentioned several key to the issue points are caused owing to allowing the clipping the ball video camera serve as the global monitoring role, and the most competent role of clipping the ball video camera should be the part monitoring of certain locus.
Summary of the invention
For overcome multiple target that existing intelligent camera apparatus exists from the motion tracking difficulty, exist the deficiency of the blindness of monitoring dead angle, clipping the ball video camera rotating, the invention provides a kind of multiple target that realizes and in motion tracking, the big visual field of omni-directional visual, freely absorb the stereo intelligent camera apparatus based on omnibearing vision sensor of dynamic, the intelligent in real time process monitoring of target, realization.
The technical solution adopted for the present invention to solve the technical problems is:
A kind of stereo intelligent camera apparatus based on omnibearing vision sensor, comprise fixed support, be installed in the clipping the ball video camera on the fixed support and be used to control the video data microprocessor that the clipping the ball video camera is captured, described stereo intelligent camera apparatus also comprises the omnibearing vision sensor that is used to obtain comprehensive video data in the monitoring field, described omnibearing vision sensor comprises the catadioptric minute surface in order to object in the reflection monitoring field, in order to the dark circles cone that prevents that anaclasis and light are saturated, transparent housing and image unit, described catadioptric minute surface is positioned at the top of transparent housing, the catadioptric minute surface down, the dark circles cone is fixed on the bottom center of catadioptric minute surface, described image unit is positioned at the focal position of described catadioptric minute surface, described transparent housing is installed on the fixed support, described catadioptric minute surface, the dark circles cone, transparent housing and image unit are on same central axis; Described video data microprocessor comprises: the demarcating module of ODVS is used to set up the image of monitoring space and the corresponding relation of the video image that is obtained;
Video data Fusion Module between ODVS and the clipping the ball video camera is used to control the rotation and the focusing of clipping the ball video camera, makes clipping the ball shooting function aim at institute's tracking target object and carries out feature candid photograph or shooting;
The dummy line customized module is used to customize the sensitizing range detection line in the monitoring field, and the dummy line customization is at the place, gateway in monitoring field;
The automatically-generating module of destination object ID number and file, be used for the destination object that just enters the sensitizing range detection line is named, when destination object enters in the sensitizing range of monitoring range, system can produce a destination object ID number and generate simultaneously a file with ID number name of this destination object automatically, is used to deposit the close-up image or the video image of this destination object; The multi-object target tracking module is used for the multiple target object in the tracing and monitoring field, comprising:
The adaptive background reduction unit is used for will monitoring domain background at the low-level image feature layer pixel of foreground object target part is extracted, and mixture gaussian modelling comes the feature of each pixel in the phenogram picture frame; If be used for describing total K of each Gaussian Profile of putting distribution of color, be labeled as respectively:
η(Y
t，μ
t，i，∑
t，i)，i＝1，2，3…，k
(12)
Subscript t express time in the formula (12), each Gaussian Profile has different weights and priority respectively, K background model is sorted according to priority order from high to low again, gets suitable surely background model weights and threshold value, when detecting the foreground point, according to priority order with Y
tMate one by one with each Gaussian distribution model, if coupling is judged that then this point may be the foreground point, otherwise is the foreground point; If certain Gaussian Profile and Y
tCoupling is then upgraded by certain turnover rate weights and Gauss's parameter of this Gaussian Profile;
Shade suppresses the unit, be used for realizing by morphology operations, utilize corrosion and expansion operator to remove isolated noise foreground point and the aperture of filling up the target area respectively, offset earlier except the prospect point set F after the background model and expand respectively and corrosion treatment, obtaining expansion collection Fe and shrink collecting Fc, is the result who initial prospect point set F is filled up aperture and removal isolated noise point by handling resulting expansion collection Fe and shrinking collection Fc; Therefore there is the following Fc of relation＜F＜Fe to set up, then, on expansion collection Fe, detects connected region to shrink collection Fc as starting point, then testing result is designated as { Rei, i=1,2,3 ..., n}, the connected region that will detect gained at last projects on the initial prospect point set F again, gets to the end communication with detection result { Ri=Rei ∩ F, i=1,2,3 ..., n};
The connected region identify unit is used to adopt eight connected region extraction algorithms to extract the foreground target object;
Tracking treatment unit; Be used for based target color characteristic track algorithm, utilize the color characteristic of destination object in video image, to find the position and the size at moving target object place, in the next frame video image, with moving target current position and big or small initialization search window, handle the orientation and the size of the resulting destination object in back at the connected region identify unit, then result is submitted to color of object signature tracking algorithm, with realize the multiple target object from motion tracking;
The monitor video data transmission module is used for giving Surveillance center with the tracking target object video transfer of data that includes that dynamically perceives.
As preferred a kind of scheme: described catadioptric mirror is a hyperbolic mirror, described image unit is positioned at the virtual focus position of described hyperbolic mirror, external screw thread on the described fixed support connects with the internal thread that is connected 5, and the clipping the ball video camera is connected with the internal thread of connector.
As preferred another kind of scheme: described catadioptric mirror is a hyperbolic mirror, described image unit is positioned at the real focus position of described hyperbolic mirror, an aperture is left in the centre of hyperbolic mirror, former configuration at hyperbolic mirror has a circular arc secondary mirror, the center of circular arc secondary mirror overlaps with the virtual focus of hyperbolic mirror, an aperture is left at the middle part of circular arc secondary mirror, embeds a wide-angle lens in this aperture.
Further, described transparent housing is bowl-shape, promptly is made of semi-round ball and circular cone, and the centre of sphere of semi-round ball overlaps with the focus of hyperbolic mirror, carries out transition at semi-round ball radius and tapered segment partly.
Further again, the scaling method of described video data Fusion Module is: with the panoramic picture center is the center of circle, as required panoramic picture is divided into annulus, then each annulus is divided equal portions, one width of cloth panoramic picture just has been divided into several zones regularly, and there are its specific angle, direction and size in each zone; According to the parameters such as height, clipping the ball video camera locus and camera focal length of ODVS, determine regional required level, vertical angle and the focal length that rotates that the clipping the ball video camera will detect successively apart from ground.
Further, in the automatically-generating module of described destination object ID number and file, the naming method of image file determines that according to the number of times of capturing video image file also is to name according to camera head and shooting sequencing; Because omnibearing vision sensor is being taken, so its video file called after ODVS.avi always; The shooting that the clipping the ball video camera is taken for the first time just is named as 1.avi.
The image unit of described omnibearing vision sensor and clipping the ball video camera are screwed respectively on fixed head, video line, control line and the power line that is connected with omnibearing vision sensor image unit, clipping the ball video camera laid on the top of fixed head, and what omnibearing vision sensor camera head and clipping the ball video camera were exposed in the bottom of fixed head obtains the video image part.
Technical conceive of the present invention is: the omnibearing vision sensor ODVS that developed recently gets up (OmniDirectional Vision Sensors) provide a kind of new solution for the panoramic picture that obtains scene in real time.The characteristics of ODVS are looking away (360 degree), can become piece image to the Information Compression in the hemisphere visual field, and the amount of information of piece image is bigger; When obtaining a scene image, ODVS is because the riding position freedom can be designed to integrated with the clipping the ball video camera; ODVS is without run-home during monitoring environment; Algorithm is simpler during moving target in the detection and tracking monitoring range; Can obtain the realtime graphic of scene on a large scale.This ODVS video camera mainly is made up of a CCD (perhaps CMOS) video camera and a reflective mirror that faces image unit.Reflective mirror reflects the video camera imaging to CCD (perhaps CMOS) with the image in one week of horizontal direction, like this, just can obtain the environmental information of 360 ° of horizontal directions in piece image.This omnidirectional vision camera has very outstanding advantage, under the real-time processing requirements to panorama, is a kind of quick, approach of visual information collection reliably especially.But ODVS also exists and can not carry out feature to the monitored object target and capture, and can't obtain tracing object target details, and the most competent role of ODVS should be the overall video monitoring in space on a large scale.
From motion tracking and Intelligent Recognition is a key character of Intellectualized monitoring.And Intelligent Recognition be based upon realized that large-scale, overall video image obtains and multiple target on the motion tracking basis, also need in addition the behavior of the object be concerned about in the monitoring field, incident, object is effectively analyzed, thereby making to allow the clipping the ball video camera know will to rotate needs to make the monitoring of the overall situation realize perfect the combination with the monitoring of part to where going and it being carried out local feature shooting.
In addition, particular design by ODVS can make the locus of monitored object and the position of this object on imaging plane conveniently to calculate, the positioning algorithm based that realizes the monitored object target is very simple, add that present clipping the ball video camera has all possessed the coordinate function, realize ODVS and the information fusion of clipping the ball video camera on the locus easily.Allow ODVS realize the real-time monitoring of the overall situation, and allow the clipping the ball video camera realize object or the local part monitoring of being concerned about of incident appearance, each Xingqi duty, the speciality of performance oneself is worked in coordination with and is finished intelligent monitoring separately.
Beneficial effect of the present invention mainly shows: 1) all monitored object activities, behavior and the incident in the energy while continuous monitoring 360 degree panorama spaces, make the video mass data obtain rapid processing and handle, use dynamic perception recording function to reduce and preserve vedio data memory space and transmission of video images amount; 2) can shorten the tracing object goal verification time; What the incident of 3) being monitored took place did retrieval mark at that time on the video image of the incident of generation, can save and browse the time that incident takes place; 4) intelligent-tracking of the action of based target object, behavior and incident rather than environmental change has improved clipping the ball useful life; When 5) focusing on a certain subject object, still can keep continual monitoring, realize the perfect unity of global monitoring with local monitoring to panorama; 6) can realize multiobject real-time tracking automatically, control the candid photograph that realizes a plurality of subject object by turned position, realize the stereo intelligent monitoring clipping the ball by control strategy; 7) can realize making more senior identification such as the recognition of face of the person of being identified under automatism to use and carry out the basis; 8) this novel solid intelligent camera apparatus adopts the integrated design scheme, has adopted outdoor simultaneously and indoor two types, can satisfy the actual demand of various monitoring occasions, and is convenient for installation and maintenance.The release of this novel solid intelligent monitoring means will improve the dynamic perception of people to things greatly with application, distinguish and monitoring capacity, be that fields such as every field especially military affairs, national defence reduce risk, improve fail safe and make outstanding contribution.
Description of drawings
Fig. 1 is the structure chart of a kind of ODVS;
Fig. 2 is the novel solid formula camera head structure chart (outdoor pattern) of a kind of ODVS and clipping the ball integrated design;
Fig. 3 is a kind of structure chart of clipping the ball video camera;
Fig. 4 is a kind of structure chart that does not have the ODVS at dead angle;
Fig. 5 has eliminated the captured image of ODVS at dead angle for adopting;
Fig. 6 is the optical schematic diagram that cam lens and wide-angle lens make up;
Fig. 7 is the novel solid formula camera head structure chart (indoor mode) of a kind of ODVS and clipping the ball integrated design;
Fig. 8 is the image-forming principle explanation schematic diagram of ODVS;
Fig. 9 is that the visual field spatial mappings between ODVS and the clipping the ball video camera concerns schematic diagram;
Figure 10 is the process chart of multi-object target following.
Embodiment
Below in conjunction with accompanying drawing the present invention is further described.
With reference to Fig. 1～Figure 10, a kind of stereo intelligent camera apparatus based on omnibearing vision sensor, by adopting an ODVS to obtain the video image of on-the-spot 360 ° of monitoring scenes, microprocessor carries out processing such as dynamic image understanding with the full-view video image that collects, obtain the object be concerned about in the monitoring field or the spatial positional information of incident, then according to the visual field spatial mappings relation between ODVS and the clipping the ball video camera, the fixed position that indication clipping the ball video camera turns to hope fast carries out the part monitoring, can obtain simultaneously the monitoring video information and the monitored object of the overall situation, the detailed local video information of incident, solved multiple target in the present video monitoring from the motion tracking difficulty, exist the monitoring dead angle, the problems such as blindness that the clipping the ball video camera rotates, realized the perfect adaptation that 360 ° of conduct monitoring at all levels and local modulation are followed the tracks of, device can freely absorb target in 360 ° of comprehensive big visual fields; Tracking lock target simultaneously automatically realizes full automatic, omnibearing, synchronous, real-time and dynamic, intelligent process monitoring, and monitor procedure need not manual operation fully.
It at first is the manufacturing technology scheme of the opticator of ODVS camera head, ODVS comprises straight vertical catadioptric mirror, transparent housing, image unit, fixed support, last fixed cap down, described catadioptric mirror is a hyperbolic mirror, described image unit comprises collector lens and image unit, and described image unit is positioned at the virtual focus position of described hyperbolic mirror; The optical system that described hyperbolic mirror constitutes is represented by following 5 equatioies;
((X
2+Y
2)/a
2)-(Z
2/b
2)＝-1(Z＞0) (1)
β＝tan
-1(Y/X) (3)
α＝tan
-1[(b
2+c
2)sinγ-2bc]/(b
2+c
2)cosγ?(4)
In the following formula, X, Y, Z representation space coordinate, c represents the focus of hyperbolic mirror, 2c represents two distances between the focus, a, b are respectively the real axis of hyperbolic mirror and the length of the imaginary axis, and β represents angle one azimuth of incident ray on the XY plane, α represents angle one angle of depression of incident ray on the XZ plane, and f represents the distance of imaging plane to the virtual focus of hyperbolic mirror.
In order to make transparent housing can not produce the reflection interference light of inwall, as shown in Figure 1.Specific practice be transparent housing is designed to bowl-shape, be that semi-round ball and circular cone constitute, the centre of sphere of semi-round ball overlaps with the focus of hyperbolic mirror, can avoid like this at transparent housing generation reflection interference light, radius and tapered segment in the semi-round ball part carry out transition, the angle of inclination of tapered segment is 2～3 °, mainly is the draw taper of considering when die production, and the structure of ODVS as shown in Figure 1;
The mounting technology scheme of ODVS camera head: the assembly relation of ODVS as shown in Figure 2, at first with image unit 2 by screw in fixed support 4, then fixed support 4 is inserted in the circular hole of transparent housing 6, owing on the external diameter of fixed support 4, have external screw thread, therefore with nut 10 fixed support be threaded with transparent housing 6 by sealing; Then be that catadioptric mirror 11 and loam cake 8 are connected by screw, the transparent housing 6 behind the loam cake 8 that will fix catadioptric mirror 11 then and the good image unit of said fixing is connected by screw 9, forms an ODVS camera head;
The connectivity scenario of the novel solid formula camera head of a kind of ODVS of outdoor use and clipping the ball integrated design: as shown in Figure 2, connector 5 has the internal thread that the external screw thread of external screw thread on the fixed support 4 with the ODVS camera head and clipping the ball 1 matches, external screw thread on the fixed support 4 of ODVS camera head is connected with the internal thread of connector 5, the internal thread of clipping the ball 1 with connector 5 is connected, makes ODVS and high-speed dome be integral; The middle part of connector 5 has a circular hole, is used for and being connected of vertical rod 7; The inside of connector 5 and vertical rod 7 is hollow, is used to connect the video line, control line and the power line that are connected with ODVS camera head, clipping the ball; This integrated design scheme is mainly used in the three-dimensional Intellectualized monitoring of pattern out of doors;
360 ° of comprehensive principles of making a video recording: a some A (X on the space, Y, Z) 17 reflex on the lens a subpoint P (x should be arranged through catadioptric minute surface 11, y) 19, the ray cast of scioptics is carried out imaging to CCD (CMOS) image unit, microprocessor 24 reads in this ring-type image by video interface 23, and the ring-type image that reads as shown in Figure 9;
Described microprocessor also comprises: the demarcating module of ODVS is used to set up the image of monitoring space and the corresponding relation of the video image that is obtained; Video data Fusion Module between ODVS and the clipping the ball video camera 1 is used to control the rotation and the focusing of clipping the ball video camera 1, makes clipping the ball video camera 1 can aim at institute's tracking target object and carries out feature candid photograph or shooting; The dummy line customized module is used to customize the sensitizing range detection line in the monitoring field, and general dummy line customization is at the place, gateway in monitoring field; Destination object ID number and deposit the image of tracking target object or the automatically-generating module of the file of video image, be used for the destination object that just enters the sensitizing range detection line is named, when destination object enters in the sensitizing range of monitoring range, system can produce a destination object ID number and generate simultaneously a file with ID number name of this destination object automatically, is used to deposit the close-up image or the video image of this destination object; The multi-object target tracking module is used to follow the tracks of the multiple target destination object that enters in the monitoring field; The monitor video data transmission module is used for giving Surveillance center with the tracking target object video transfer of data that includes that dynamically perceives;
Video data Fusion Module between described ODVS and the clipping the ball video camera 1, be used to set up the spatial mappings relation between ODVS camera head and the clipping the ball video camera 1, be used for when ODVS detects subject object, make clipping the ball video camera 1 to locate fast according to own corresponding relation calibration scale with ODVS.Scaling method is: with the panoramic picture center is the center of circle, as required panoramic picture is divided into some annulus, as shown in Figure 9, then each annulus is divided several equal portions, can divide according to actual needs.Like this, a width of cloth panoramic picture just has been divided into several zones regularly, and there are its specific angle, direction and size in each zone.Then, according to the parameters such as height, clipping the ball video camera locus and camera focal length of ODVS, determine regional required level, vertical angle and the focal length that rotates that the clipping the ball video camera will detect successively apart from ground.Exist if ODVS has detected subject object, system just can regulate each quick ball-shaped camera automatically according to this subject object region and calibration scale, and this subject object is carried out feature candid photograph or shooting; The clipping the ball video camera of Xiao Shouing can be with 80～256 preset point in the market, apace the clipping the ball video camera is turned to detected subject object of ODVS and focusing rapidly in order to reach, the spatial positional information of realizing ODVS and clipping the ball video camera merges, visual field with ODVS in this patent is divided into 128 zones, as shown in Figure 9, each zone corresponding a preset point of clipping the ball video camera, therefore as long as on the captured video image of ODVS, divide 128 zones, then to the preset point of each regional center as the clipping the ball video camera; Detect the tracking target object when certain surveyed area occurs at ODVS, control clipping the ball video camera 1 arrived and this area relative preset point after microprocessor 24 obtained this information, the destination object of being followed the tracks of is carried out feature candid photograph or video recording, realize the dynamic perception recording function of ODVS;
Described destination object ID number, be used to produce one and can identify the major key that the destination object followed the tracks of, storage write down the time data that enters the monitoring field of this destination object, destination object ID number naming rule is: YYYYMMDDHHMMSS* names with 14 bit signs, and YYYY-represents the year of the Gregorian calendar; MM-represents the moon; DD-represents day; HH-represents hour; MM-represents branch; SS-represents second; Automatically produce by the system for computer time;
Described destination object ID number and deposit the image of tracking target object or the automatically-generating module of the file of video image, be used to preserve the close-up image and the video image of the destination object of tracking, when destination object enters monitoring range (dummy line of outermost), system automatically produce one destination object ID number, in certain deposits the file of image, create simultaneously one with destination object ID number file of the same name, be used for depositing the close-up image and the video image of this destination object, the naming method of image file is to determine according to the number of times of capturing, if the image of capturing is with regard to called after 1.jpg for the first time, video image file also be according to camera head and the shooting sequencing name, because ODVS is taking always, therefore its video file just has only one, among the present invention with regard to called after ODVS.avi; If the shooting that the clipping the ball video camera is taken for the first time just is named as 1.avi; Above-mentioned processing mainly is that the multiple target tracking problem is arranged, and the clipping the ball video camera will be captured each target respectively, and when only having a tracking target object in the monitoring range, the clipping the ball video camera is made a video recording with regard to following the tracks of this object all the time;
Described multi-object target tracking module is used to follow the tracks of the multiple target subject object that enters in the monitoring field; For the direction of motion of computational activity destination object, the detection and the candid photograph of subject object features location of variety of event provide technical foundation; When realizing the multi-object target following, at first to will monitor in the domain background foreground object target pixel will partly be extracted at the low-level image feature layer.Extract the subject object part pixel method mainly by adaptive background subdue, shade suppresses and three parts of connected region sign are formed, its precedence diagram is as shown in Figure 10;
Described adaptive background is subdued, be used for real-time cutting object target, the adaptive background elimination algorithm that is based on mixture gaussian modelling that we adopt, its basic thought is: the feature of using mixture gaussian modelling to come each pixel in the phenogram picture frame; When obtaining new picture frame, upgrade mixture gaussian modelling; On each time period, select the subclass of mixture gaussian modelling to characterize current background; If the pixel of present image and mixture gaussian modelling are complementary, judge that then this point is a background dot, otherwise judge that this point is the foreground point.
Detect at the brightness value Y component in the YCrCb color space of image.The ADAPTIVE MIXED Gauss model has adopted the hybrid representation of a plurality of Gauss models to each picture point, establishes total K of the Gaussian Profile that is used for describing each some distribution of color, is labeled as respectively
η(Y
t，μ
t，i，∑
t，i)，i＝1，2，3…，k (12)
Subscript t express time in the formula (12).Each Gaussian Profile has different weights and priority respectively, K background model is sorted according to priority order from high to low again, gets suitable surely background model weights and threshold value.When detecting the foreground point, according to priority order with Y
tMate one by one with each Gaussian distribution model.If coupling is judged that then this point may be the foreground point, otherwise is the foreground point.If certain Gaussian Profile and Y
tCoupling is then upgraded by certain turnover rate weights and Gauss's parameter of this Gaussian Profile.
Described connected region sign is used to extract the foreground target object, and the method for connected region sign is a lot, and what adopt among the present invention is eight connected region extraction algorithms.The connected region sign is subjected to the noise effect in the primary data very big in addition, generally need carry out denoising earlier, denoising can realize by morphology operations, utilize corrosion and expansion operator to remove isolated noise foreground point and the aperture of filling up the target area respectively among the present invention, specific practice is: offset earlier except the prospect point set F after the background model and expand respectively and corrosion treatment, obtain expansion collection Fe and shrink collecting Fc, by handling resulting expansion collection Fe and shrinking the result that collection Fc can think initial prospect point set F is filled up aperture and removal isolated noise point.Therefore there is the following Fc of relation＜F＜Fe to set up, then to shrink collection Fc as starting point, on expansion collection Fe, detect connected region, then testing result is designated as { Rei, i=1,2,3,, n}, the connected region that will detect gained at last projects on the initial prospect point set F again, get communication with detection result { Ri=Rei ∩ F to the end, i=1,2,3, n} can keep the integrality of target also to avoid the influence of noise foreground point simultaneously by this target partitioning algorithm, has also kept the edge details part of target.
Just can from the monitoring domain background of static state, dynamic foreground target object extraction be come out through above three steps, and can obtain some initial informations of based target object, as the size of destination object and locus or the like.Extract the foreground target object in the monitoring domain background after, the processing of following is that the destination object in the monitoring field is followed the tracks of processing; Adopt based target color characteristic track algorithm among the present invention, this algorithm is the further improvement to the MEANSHIFT algorithm, in this algorithm, utilize the color characteristic of destination object in video image, to find the position and the size at moving target object place, in the next frame video image, with moving target current position and big or small initialization search window, repeat this process and just can realize Continuous Tracking target.The initial value of search window is set to current position of moving target and size before each the search, because search window is just searched near the zone that moving target may occur, so just can save a large amount of search times, make this algorithm have good real time performance.This algorithm is to find moving target by color matching simultaneously, and in the process of destination object motion, colouring information changes little, so this algorithm has good robustness.The orientation of resulting destination object and size after the connected region identification process are submitted to result color of object signature tracking algorithm then, with realize the multiple target object from motion tracking;
Described monitor video data transmission module is used for giving Surveillance center with the tracking target object video transfer of data that includes that dynamically perceives; Taking place back two kinds with incident during the monitor video transfer of data takes place according to incident among the present invention transmits, the transmission of the video image during incident takes place and currently used technology type are together, with ODVS and clipping the ball shot by camera to video image compress through Network Transmission and give Surveillance center, Surveillance center decompresses according to the video image that transmits, and is presented at then on the control flow; The transmission of the video image after incident takes place, the image file of depositing the video file of ODVS video file in the file of the image of tracking target object or video image, clipping the ball video camera or candid photograph packed after compression give Surveillance center through Network Transmission, Surveillance center decompresses to the video file in the file respectively according to the packaging file that transmits, and is presented on the control flow then.
With reference to Fig. 1～Figure 10, the realization principle of present embodiment is identical with embodiment 1, different with the three-dimensional intelligent video monitoring apparatus of embodiment 1 outdoor pattern is: in the three-dimensional intelligent video monitoring of indoor mode, need consider from angle attractive in appearance, wish that all lead-in wires are hidden and do not have vertical rod; Need consider from people's psychological aspects, can not give a kind of sensation that is monitored of people; Therefore need be fit to the design of indoor mode to the ODVS camera head; The structure chart of a kind of ODVS that does not have a dead angle as shown in Figure 4, in the present embodiment image unit 2 is configured in the back of hyperbolic mirror 11, be placed in the real focus place of hyperbolic mirror 11, an aperture is left in the centre of hyperbolic mirror, and image unit 2 can photograph the video information of the front of hyperbolic mirror 11 by this aperture; Former configuration at hyperbolic mirror 11 has a circular arc secondary mirror 12, and the center of circular arc secondary mirror 12 overlaps with the virtual focus of hyperbolic mirror 11, and an aperture is left at the middle part of circular arc secondary mirror 12, embeds a wide-angle lens 13 in this aperture; Comprehensive video information after circular arc secondary mirror 12 carries out the secondary catadioptric, is passed through aperture imaging in image unit 2 of hyperbolic mirror in hyperbolic mirror 11 catadioptrics then; Object in hyperbolic mirror 11 dead aheads passes through wide-angle lens 13 imaging between the camera lens of wide-angle lens 13 and image unit 2 in addition, is called first imaging point, and this imaging point passes through the focus place imaging of the aperture of hyperbolic mirror at the camera lens of image unit;
Through the captured image effect figure that comes out of the ODVS of above-mentioned design as shown in Figure 5, Fig. 5 middle part circular portion is by the captured image of wide-angle lens, the outer circle loop section is to carry out the resulting video image of secondary catadioptric by hyperbolic mirror through the circular arc secondary mirror, design ODVS has eliminated the dead angle part of original ODVS like this, and add the design of hyperbolic mirror and circular arc secondary mirror by the compound mode of cam lens and wide-angle lens, the visual range of wide-angle lens can be covered the dead angle part of original ODVS;
Aperture on described hyperbolic mirror, the hyperbolic mirror, image unit, transparent housing, circular arc secondary mirror, wide-angle lens are on same central axis; The camera lens of image unit is placed on the position of real focus of hyperbolic mirror;
Come the principle of expanded field of vision scope to be by wide-angle lens: among the present invention wide-angle lens is configured in the middle part of circular arc secondary mirror, the position of design wide-angle lens and definite wide-angle lens is a task of the present invention.Fig. 6 is the location diagram of image unit camera lens and wide-angle lens.Wide-angle lens is configured in the middle part of circular arc secondary mirror, and imaging between wide-angle lens and the image unit camera lens is called first imaging point, this imaging point by the image unit camera lens in the imaging of focus place.Here with the focal length of image unit camera lens as the focal length of f1, wide-angle lens as the focal length of the camera lens of f2, image unit and image unit as S1, focal length from the camera lens of image unit to first imaging point as S2, distance from wide-angle lens to first imaging point as S3, the distance of point can obtain following relational expression as S4 according to the imaging formula of camera lens from the wide-angle lens to the material object:
d＝S2+S3 (8)
Formula (8) is set up, and is the place configuration wide-angle lens of d with the image unit distance of camera lens on the real focus of hyperbolic mirror among Fig. 4 just, just can obtain the shown wide-angle image in image middle part among Fig. 5.
For among Fig. 4 camera lens and wide-angle lens being considered that as a compound lens its focal distance f can be represented by following formula:
In addition, as D, its multiplication factor can be represented by following formula with the diameter of process shot:
For the visual field of process shot and the dead angle part of hyperbolic mirror are matched, when the design process shot, need satisfy following formula:
The mounting technology scheme of the ODVS camera head at no dead angle: the assembly relation of ODVS at first is embedded into catadioptric mirror 11 in the loam cake 8 as shown in Figure 4, then with image unit 2 by screw on loam cake 8; Then wide-angle lens 13 being embedded in circular arc secondary mirror (secondary reflection minute surface) 12 holes, then is that the circular arc secondary mirror that will assemble is embedded in the hole of transparent housing 6; The wide-angle lens 13 that the loam cake 8 that will fix catadioptric mirror 11 and image unit 2 at last and said fixing are good and the transparent housing 6 of circular arc secondary mirror 12 are connected by screw 9, form the ODVS camera head at a no dead angle;
The connectivity scenario of the novel solid formula camera head of a kind of ODVS of indoor use and clipping the ball integrated design: as shown in Figure 7, fixed head 14 is designed to and the same size of Indoor decorative plate, the ODVS camera head and the clipping the ball video camera 1 at no dead angle are screwed respectively on fixed head 14, the top of fixed head 14 is used to lay video line, control line and the power line that is connected with ODVS camera head, clipping the ball video camera, and what the bottom of fixed head 14 only exposed ODVS camera head and clipping the ball video camera 1 obtains the video image part.
Claims (7)
1, a kind of stereo intelligent camera apparatus based on omnibearing vision sensor, comprise fixed support, be installed in the clipping the ball video camera on the fixed support and be used to control the video data microprocessor that the clipping the ball video camera is captured, it is characterized in that: described stereo intelligent camera apparatus also comprises the omnibearing vision sensor that is used to obtain comprehensive video data in the monitoring field, described omnibearing vision sensor comprises the catadioptric minute surface in order to object in the reflection monitoring field, in order to the dark circles cone that prevents that anaclasis and light are saturated, transparent housing and image unit, described catadioptric minute surface is positioned at the top of transparent housing, the catadioptric minute surface down, the dark circles cone is fixed on the bottom center of catadioptric minute surface, described image unit is positioned at the focal position of described catadioptric minute surface, described transparent housing is installed on the fixed support, described catadioptric minute surface, the dark circles cone, transparent housing and image unit are on same central axis; Described catadioptric mirror is a hyperbolic mirror, and described video data microprocessor comprises:
The demarcating module of omnibearing vision sensor is used to set up the image of monitoring space and the corresponding relation of the video image that is obtained;
Video data Fusion Module between omnibearing vision sensor and the clipping the ball video camera is used to control the rotation and the focusing of clipping the ball video camera, makes clipping the ball shooting function aim at institute's tracking target object and carries out feature candid photograph or shooting;
The dummy line customized module is used to customize the sensitizing range detection line in the monitoring field, and the dummy line customization is at the place, gateway in monitoring field;
The automatically-generating module of destination object ID number and file, be used for the destination object that just enters the sensitizing range detection line is named, when destination object enters in the sensitizing range of monitoring range, system can produce a destination object ID number and generate simultaneously a file with ID number name of this destination object automatically, is used to deposit the close-up image or the video image of this destination object; The multi-object target tracking module is used for the multiple target object in the tracing and monitoring field, comprising:
The adaptive background reduction unit is used for will monitoring domain background at the low-level image feature layer pixel of foreground object target part is extracted, and mixture gaussian modelling comes the feature of each pixel in the phenogram picture frame; If be used for describing total K of each Gaussian Profile of putting distribution of color, be labeled as respectively:
η(Y
t，μ
t，i，∑
t，i)，i＝1，2，3…，k (12)
Subscript t express time in the formula (12), each Gaussian Profile has different weights and priority respectively, K background model is sorted according to priority order from high to low again, gets suitable surely background model weights and threshold value, when detecting the foreground point, according to priority order with Y
tMate one by one with each Gaussian distribution model, if coupling is judged that then this point is a background dot, otherwise is the foreground point; If certain Gaussian Profile and Y
tCoupling is then upgraded by certain turnover rate weights and Gauss's parameter of this Gaussian Profile;
Shade suppresses the unit, be used for realizing by morphology operations, utilize corrosion and expansion operator to remove isolated noise foreground point and the aperture of filling up the target area respectively, offset earlier except the prospect point set F after the background model and expand respectively and corrosion treatment, obtaining expansion collection Fe and shrink collecting Fc, is the result who initial prospect point set F is filled up aperture and removal isolated noise point by handling resulting expansion collection Fe and shrinking collection Fc; Therefore there is the following Fc of relation＜F＜Fe to set up, then, on expansion collection Fe, detects connected region to shrink collection Fc as starting point, then testing result is designated as { Rei, i=1,2,3 ..., n}, the connected region that will detect gained at last projects on the initial prospect point set F again, gets to the end communication with detection result { Ri=Rei ∩ F, i=1,2,3 ..., n};
The connected region identify unit is used to adopt eight connected region extraction algorithms to extract the foreground target object;
Tracking treatment unit; Be used for based target color characteristic track algorithm, utilize the color characteristic of destination object in video image, to find the position and the size at moving target object place, in the next frame video image, with moving target current position and big or small initialization search window, handle the orientation and the size of the resulting destination object in back at the connected region identify unit, then result is submitted to color of object signature tracking algorithm, with realize the multiple target object from motion tracking;
The monitor video data transmission module is used for giving Surveillance center with the tracking target object video transfer of data that includes that dynamically perceives.
2, the stereo intelligent camera apparatus based on omnibearing vision sensor as claimed in claim 1, it is characterized in that: described image unit is positioned at the virtual focus position of described hyperbolic mirror, external screw thread on the described fixed support is connected with the internal thread of connector (5), and the clipping the ball video camera is connected with the internal thread of connector.
3, the stereo intelligent camera apparatus based on omnibearing vision sensor as claimed in claim 1, it is characterized in that: described image unit is positioned at the real focus position of described hyperbolic mirror, an aperture is left in the centre of hyperbolic mirror, former configuration at hyperbolic mirror has a circular arc secondary mirror, the center of circular arc secondary mirror overlaps with the virtual focus of hyperbolic mirror, an aperture is left at the middle part of circular arc secondary mirror, embeds a wide-angle lens in this aperture.
4, as the described stereo intelligent camera apparatus of one of claim 1-3 based on omnibearing vision sensor, it is characterized in that: described transparent housing is bowl-shape, promptly constitute by semi-round ball and circular cone, the centre of sphere of semi-round ball overlaps with the focus of hyperbolic mirror, carries out transition at semi-round ball radius and tapered segment partly.
5, the stereo intelligent camera apparatus based on omnibearing vision sensor as claimed in claim 4, it is characterized in that: the scaling method of described video data Fusion Module is: with the panoramic picture center is the center of circle, as required panoramic picture is divided into annulus, then each annulus is divided equal portions, one width of cloth panoramic picture just has been divided into several zones regularly, and there are its specific angle, direction and size in each zone; According to the parameters such as height, clipping the ball video camera locus and camera focal length of omnibearing vision sensor, determine regional required level, vertical angle and the focal length that rotates that the clipping the ball video camera will detect successively apart from ground.
6, stereo intelligent camera apparatus based on omnibearing vision sensor as claimed in claim 5, it is characterized in that: in the automatically-generating module of described destination object ID number and file, be used to preserve the close-up image and the video image of the destination object of tracking, when destination object enters monitoring range, system automatically produce one destination object ID number, in certain deposits the file of image, create simultaneously one with destination object ID number file of the same name, the naming method of close-up image file is determined according to the number of times of capturing, omnibearing vision sensor is being taken always, with its video file called after omnibearing vision sensor .avi; The shooting that the clipping the ball video camera is taken for the first time just is named as 1.avi.
7, the stereo intelligent camera apparatus based on omnibearing vision sensor as claimed in claim 3, it is characterized in that: the image unit of described omnibearing vision sensor and clipping the ball video camera are screwed respectively on fixed head, video line, control line and the power line that is connected with omnibearing vision sensor image unit, clipping the ball video camera laid on the top of fixed head, and what omnibearing vision sensor camera head and clipping the ball video camera were exposed in the bottom of fixed head obtains the video image part.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200710164441A CN100576908C (en) | 2007-11-30 | 2007-11-30 | Stereo intelligent camera apparatus based on omnibearing vision sensor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200710164441A CN100576908C (en) | 2007-11-30 | 2007-11-30 | Stereo intelligent camera apparatus based on omnibearing vision sensor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101291427A CN101291427A (en) | 2008-10-22 |
| CN100576908C true CN100576908C (en) | 2009-12-30 |
Family
ID=40035466
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200710164441A Expired - Fee Related CN100576908C (en) | 2007-11-30 | 2007-11-30 | Stereo intelligent camera apparatus based on omnibearing vision sensor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100576908C (en) |
Families Citing this family (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101488220B (en) * | 2009-01-05 | 2011-01-05 | 浙江大学 | Real-time video and image abstraction method based on vision attention |
| CN101511004A (en) * | 2009-03-25 | 2009-08-19 | 北京中星微电子有限公司 | Method and apparatus for monitoring camera shot |
| CN101930603B (en) * | 2010-08-06 | 2012-08-22 | 华南理工大学 | Method for fusing image data of medium-high speed sensor network |
| WO2012056437A1 (en) | 2010-10-29 | 2012-05-03 | École Polytechnique Fédérale De Lausanne (Epfl) | Omnidirectional sensor array system |
| CN102158652A (en) * | 2011-04-11 | 2011-08-17 | 天津汇讯视通科技有限公司 | Method for controlling MOAT-Single Chip for intelligent ball products |
| CN102799887B (en) * | 2012-06-19 | 2015-09-02 | 上海地铁盾构设备工程有限公司 | Automatic calibration method of structural distortion detection image sensor sensitivity |
| CN102957852A (en) * | 2012-07-19 | 2013-03-06 | 中国人民解放军国防科学技术大学 | Full-focusing refraction and reflection imaging method and full-focusing refraction and reflection imaging device based on coded aperture |
| CN103297692A (en) * | 2013-05-14 | 2013-09-11 | 温州市凯能电子科技有限公司 | Quick positioning system and quick positioning method of internet protocol camera |
| CN103327238B (en) * | 2013-05-29 | 2016-05-11 | 湖南源信光电科技有限公司 | Based on the spatial registration method of panorama clipping the ball all-in-one |
| CN103327298A (en) * | 2013-05-29 | 2013-09-25 | 山西绿色光电产业科学技术研究院（有限公司） | Linkage browse method for panoramic speed dome all-in-one machine |
| CN104539896B (en) * | 2014-12-25 | 2018-07-10 | 桂林远望智能通信科技有限公司 | The intelligent monitor system and method for a kind of overall view monitoring and hot spot feature |
| CN104639916A (en) * | 2015-03-04 | 2015-05-20 | 合肥巨清信息科技有限公司 | Large-scene multi-target tracking shooting video monitoring system and monitoring method thereof |
| CN106331596B (en) * | 2015-07-06 | 2021-02-19 | 中兴通讯股份有限公司 | Household monitoring method and system |
| JP7304820B2 (en) * | 2017-02-14 | 2023-07-07 | アトラクシス エス・アー・エール・エル | Fast optical tracking with compression and/or CMOS windowing |
| CN108765457B (en) * | 2018-04-11 | 2021-09-07 | 深圳市瑞立视多媒体科技有限公司 | Motion gesture recognition method and device for catching ball |
| CN109557534B (en) * | 2018-09-29 | 2021-11-12 | 河北德冠隆电子科技有限公司 | Multi-element omnibearing tracking detection radar sensor equipment and use method thereof |
| CN110413353B (en) * | 2019-07-25 | 2023-05-12 | 浙江大华技术股份有限公司 | Data processing method, device and storage medium |
| CN110351469B (en) * | 2019-07-29 | 2020-05-12 | 江西联益光学有限公司 | Camera module and assembling process thereof |
| CN111083444B (en) * | 2019-12-26 | 2021-10-15 | 浙江大华技术股份有限公司 | Snapshot method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2705807Y (en) * | 2004-04-26 | 2005-06-22 | 上海鸣俱妥国际贸易有限公司 | Omnibearing vision sensor |
| CN1804927A (en) * | 2005-12-28 | 2006-07-19 | 浙江工业大学 | Omnibearing visual sensor based road monitoring apparatus |
| CN1812570A (en) * | 2005-12-31 | 2006-08-02 | 浙江工业大学 | Vehicle antitheft device based on omnibearing computer vision |
| CN1893671A (en) * | 2005-07-06 | 2007-01-10 | 浙江工业大学 | Digital-code pick-up apparatus with panoramic pick-up function |
| CN101004538A (en) * | 2007-01-18 | 2007-07-25 | 汤一平 | Omnibearing vision sensor with no dead angle |
-
2007
- 2007-11-30 CN CN200710164441A patent/CN100576908C/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2705807Y (en) * | 2004-04-26 | 2005-06-22 | 上海鸣俱妥国际贸易有限公司 | Omnibearing vision sensor |
| CN1893671A (en) * | 2005-07-06 | 2007-01-10 | 浙江工业大学 | Digital-code pick-up apparatus with panoramic pick-up function |
| CN1804927A (en) * | 2005-12-28 | 2006-07-19 | 浙江工业大学 | Omnibearing visual sensor based road monitoring apparatus |
| CN1812570A (en) * | 2005-12-31 | 2006-08-02 | 浙江工业大学 | Vehicle antitheft device based on omnibearing computer vision |
| CN101004538A (en) * | 2007-01-18 | 2007-07-25 | 汤一平 | Omnibearing vision sensor with no dead angle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101291427A (en) | 2008-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100576908C (en) | Stereo intelligent camera apparatus based on omnibearing vision sensor | |
| CN103795976B (en) | A kind of full-time empty 3 d visualization method | |
| US10366509B2 (en) | Setting different background model sensitivities by user defined regions and background filters | |
| CN103402052B (en) | Rotationally symmetrical pantoscope is utilized to obtain panorama and the method and apparatus of rectilinear images | |
| CN105516654B (en) | A kind of supervision of the cities video fusion method based on scene structure analysis | |
| CN100432836C (en) | Method for realizing two-dimensional panoramic true imaging | |
| CN107483889A (en) | The tunnel monitoring system of wisdom building site control platform | |
| CN101542232B (en) | Normal information generating device and normal information generating method | |
| WO2020029921A1 (en) | Monitoring method and device | |
| CN109416413A (en) | Solar energy forecast | |
| CN112449093A (en) | Three-dimensional panoramic video fusion monitoring platform | |
| CN102036054A (en) | Intelligent video monitoring system based on three-dimensional virtual scene | |
| CN105139332A (en) | Method for obtaining a composite image using rotationally symmetrical wide-angle lenses, imaging system for same, and cmos image sensor for image-processing | |
| CN206260046U (en) | A kind of thermal source based on thermal infrared imager and swarm into tracks of device | |
| CN107038754A (en) | A kind of wisdom garden management system and method based on three-dimensional live | |
| CN101021669A (en) | Whole-view field imaging and displaying method and system | |
| CN102170563A (en) | Intelligent person capture system and person monitoring management method | |
| CN103929592A (en) | All-dimensional intelligent monitoring equipment and method | |
| WO2021077607A1 (en) | Data processing method and apparatus | |
| CN110175587A (en) | A kind of video frequency tracking method based on recognition of face and Algorithm for gait recognition | |
| CN202818503U (en) | Multidirectional monitoring area early warning positioning automatic tracking and monitoring device | |
| US11703820B2 (en) | Monitoring management and control system based on panoramic big data | |
| CN111163246A (en) | High-altitude falling object monitoring system | |
| CN107018359A (en) | A kind of vehicle-carried mobile monitoring system that there is intelligent picture to recognize | |
| CN103672623B (en) | A kind of light fixture and monitoring system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20091230 Termination date: 20101130 |