CN102340628A - Camera and control method thereof - Google Patents
Camera and control method thereof Download PDFInfo
- Publication number
- CN102340628A CN102340628A CN2010102329119A CN201010232911A CN102340628A CN 102340628 A CN102340628 A CN 102340628A CN 2010102329119 A CN2010102329119 A CN 2010102329119A CN 201010232911 A CN201010232911 A CN 201010232911A CN 102340628 A CN102340628 A CN 102340628A
- Authority
- CN
- China
- Prior art keywords
- dimensional
- image
- human
- pixel value
- zone
- 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.)
- Pending
Links
Images
Landscapes
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention provides a camera and a control method thereof. The camera comprises a memory, a driving unit, a lens and a control unit, wherein the memory is used for storing a three-dimensional human-like image and a three-dimensional human face image; the control unit is used for comparing and analyzing a scene image shot by the camera with the stored three-dimensional human-like image as well as detecting a three-dimensional human-like region in the scene image; the driving unit is used for driving the lens to correspondingly move, adjust the focal length and shoot to obtain a distinct three-dimensional human-like image; the control unit can be also used for comparing and analyzing the distinct three-dimensional human-like image with the stored three-dimensional human face image as well as detecting a three-dimensional human face region; and the driving unit is used for driving the lens to correspondingly move, adjust the focal length and shoot to obtain a distinct three-dimensional human face image.
Description
Technical field
The present invention relates to a kind of supervisory control system, especially about a kind of video camera and control method thereof.
Background technology
Traditional PTZ video camera that possesses pan-tilt/convergent-divergent (Pan/Tilt/Zoom) function must rely on the keep one's eyes open image of guarded region of peace control personnel, can control the situation of guarded region.When peace control personnel discover when the suspicious figure occurring in the image, controller that also can only manual adjustment PTZ video camera carries out camera lens visual angle and focal length adjustment operation, just can obtain personage's image comparatively clearly.Yet, because the guarded region most time is a safe condition, be prone to make the peace control personnel of long-term observation image to reduce alertness, the image that is difficult to continue guarded region keeps hig diligence power.If peace control personnel have ignored the suspicious figure who occurs in the image, or the speed of the controller of peace control personnel adjustment PTZ video camera can't catch up with suspicious figure's translational speed, then causes the suspicious figure's that the PTZ video camera takes image sharpness not enough easily.
Summary of the invention
In view of above content, be necessary to propose a kind of video camera, can initiatively detect guarded region and the suspicious figure whether occur, when detecting guarded region when the suspicious figure occurring, can follow the trail of the suspicious figure, obtain three-dimensional clearly personage's image and image of face.
In addition; Also be necessary to propose a kind of control method of video camera, can initiatively detect guarded region and the suspicious figure whether occur, when detecting guarded region when the suspicious figure occurring; Can follow the trail of the suspicious figure, obtain three-dimensional clearly personage's image and image of face.
A kind of video camera, this video camera comprises memory, driver element, camera lens and control unit.Wherein, memory storage video camera three-dimensional human-like image and the three-dimensional face image taken in advance.Control unit is according to the range information of each point in the image to camera lens; The scene image of the guarded region that video camera is taken and the three-dimensional human-like image of memory storage compare, analyze; To detect the three-dimensional human-like zone in this scene image; And drive camera lens according to positional information and the proportion control drive unit of the three-dimensional human-like zone that detects in scene image and carry out corresponding mobile and focal length adjustment operation, obtain three-dimensional clearly human-like image with shooting.Control unit; Also according to the range information of each point in the image to camera lens; The three-dimensional face image of this three-dimensional clearly human-like image and memory storage is compared, analyzes; Three-dimensional face to detect in this three-dimensional clearly human-like image is regional, and carries out corresponding mobile and focal length adjustment operation according to the positional information and the proportion control drive unit driving camera lens of the three-dimensional face zone that detects in the human-like image of three-dimensional, obtains three-dimensional face image clearly with shooting.
A kind of control method of video camera; This method comprises: (A) three-dimensional human-like area detecting step: according to the range information of each point in the image to camera lens the scene image of the guarded region of video camera shooting and the three-dimensional human-like image of memory storage are compared, analyze, to detect the three-dimensional human-like zone in this scene image; (B) first controlled step: carry out corresponding mobile and focal length adjustment operation according to the positional information and the proportion control drive unit driving camera lens of the three-dimensional human-like zone that detects in scene image, obtain three-dimensional clearly human-like image with shooting; (C) three-dimensional face area detecting step: compare, analyze to the range information of camera lens three-dimensional face image according to each point in the image, to detect the three-dimensional face zone in this three-dimensional clearly human-like image with this three-dimensional clearly human-like image and memory storage; And (D) second controlled step: drive camera lens according to positional information and the proportion control drive unit of the three-dimensional face zone that detects in the human-like image of three-dimensional and carry out corresponding mobile and focal length adjustment operation, obtain three-dimensional face image clearly with shooting.
Compared to prior art; Video camera provided by the present invention and control method thereof can initiatively be detected guarded region and the suspicious figure whether occur, when detecting guarded region when the suspicious figure occurring; Can follow the trail of the suspicious figure, obtain three-dimensional clearly personage's image and image of face.
Description of drawings
Fig. 1 is the hardware structure figure of video camera preferred embodiment of the present invention.
Fig. 2 is the functional block diagram of control unit and memory among Fig. 1.
Fig. 3 and Fig. 4 are the flow charts of camera control method preferred embodiment of the present invention.
Fig. 5 and Fig. 6 are respectively the sketch mapes of the three-dimensional human-like image of positive three-dimensional human-like image and side.
Fig. 7 is the sketch map of a three-dimensional face.
Fig. 8 is the sketch map of a scene image.
Fig. 9 is a sketch map of three-dimensional human-like image clearly.
Figure 10 is a sketch map of three-dimensional face image clearly.
The main element symbol description
|
100 |
The |
10 |
The |
11 |
The |
12 |
The |
13 |
|
20 |
|
21 |
|
22 |
|
30 |
Three-dimensional template is set up |
31 |
The image |
32 |
Three-dimensional human- |
33 |
The three-dimensional |
34 |
|
35 |
|
40 |
|
50 |
Three-dimensional human- |
51 |
The three- |
52 |
Embodiment
As shown in Figure 1, be the hardware structure figure of video camera 100 preferred embodiments of the present invention.This video camera 100 comprises pan-tilt/convergent-divergent (pan/tilt/zoom is called for short PTZ) driver element 10, image capture unit 20, control unit 30, processor 40 and memory 50.Wherein, image capture unit 20 comprises the camera lens 21 and image sensor 22 of the continuous image that is used to take monitoring scene, and image sensor 22 focuses on through 21 pairs of monitoring scenes of camera lens.This image sensor 22 can for charge coupled device (charged coupled device, CCD) or complementary metal oxide semiconductors (CMOS) (complementary metal oxide semiconductor, CMOS).
In the present embodiment, this video camera 100 is that (Time of Flight, TOF) video camera are used to absorb the scene image in the monitoring scene scope, and the depth of view information of obtaining subject in the scene image in a kind of time flight.The depth of view information of said subject is meant the range information of the camera lens 21 of subject each point and video camera 100.Because the TOF video camera is when the photographic subjects thing; Signal with the certain wavelength of emission;, signal can reflex to the camera lens of TOF video camera when running into object; Can calculate the range information between the each point and TOF camera lens on the object according to signal emission and time difference between receiving, so said video camera 100 can obtain the range information between the camera lens 21 of subject each point and video camera 100 in the scene image.
As shown in Figure 2 is the functional block diagram of control unit 30 and memory 50 among Fig. 1.
Three-dimensional human-like data 51 comprises a large amount of three-dimensional human-like image of taking before the video camera 100 of collection; In the present embodiment, these three-dimensional human-like images mainly are divided three classes according to posture: the three-dimensional human-like image (not shown) in positive human-like image (as shown in Figure 5), the human-like image in side (as shown in Figure 6) and the back side.The three-dimensional human-like image that can comprise more kinds of postures in other embodiments.
Three-dimensional face data 52 comprises a large amount of three-dimensional face image (a three-dimensional face image being shown like Fig. 7) of taking before the video camera 100 of collection.
In the present embodiment, this control unit 30 comprises that three-dimensional template sets up module 31, image information processing module 32, three-dimensional human-like detecting module 33, three-dimensional face identification module 34 and control module 35.
Three-dimensional template is set up module 31 and is set up three-dimensional human-like template according to the range information between each point and the camera lens 21 in the three-dimensional human-like image of memory 50 storages, is used to store the permissible range of the pixel value of each characteristic point on the 3 D human body profile, the concrete introduction as follows:
Three-dimensional template is set up the every Zhang San who stores in the module 31 analyzing stored devices 50 and is tieed up human-like image; Obtain in the human-like image of this three-dimensional the range data of each characteristic point (for example nose, place between the eyebrows etc.) to camera lens 21 on the human body contour outline, and convert this range data into eigenmatrix that pixel value (span is 0~255) is stored to the human-like image of this three-dimensional.After three-dimensional template is set up module 31 and also the eigenmatrix of all three-dimensional human-like images of same type (for example positive) is alignd according to a characteristic point (for example human body central point) of setting; Pixel value to the characteristic point on this type human body carries out the pointwise statistics, obtains the three-dimensional human-like template of the permissible range composition of the pixel value of each characteristic point on this type human body contour outline.For example; The permissible range of the pixel value of each characteristic point is formed positive human-like three-dimensional template on the front face human body profile; The permissible range of the pixel value of each characteristic point is formed the human-like three-dimensional template of side on the human body contour outline of side, and the permissible range of the pixel value of each characteristic point is formed the human-like three-dimensional template at the back side on the human body contour outline of the back side.
For example; Three-dimensional template is set up module 31 and is analyzed 200 characteristic points on the human body contour outline among the three-dimensional human-like image N1 (as shown in Figure 5) in a front; Obtain the range data of each characteristic point to camera lens 21 and convert pixel value into; For example the distance of the Z direction of nose to camera lens 21 is 61 centimetres and is converted into pixel value 255, and the distance of the Z direction of place between the eyebrows to camera lens 21 is 59 centimetres and is converted into pixel value 253, or the like.Three-dimensional template is set up the eigenmatrix that module 31 is stored to the pixel value of these 200 characteristic points the human-like image N1 of this three-dimensional.Suppose that positive three-dimensional human-like image one has 10; Three-dimensional template is set up the eigenmatrix that module 31 Using such method calculate the three-dimensional human-like image in other 9 fronts; After 10 eigenmatrixes that obtain are alignd according to the pixel value of human body central point; Pixel value to same characteristic features point in these 10 eigenmatrixes is added up, and obtains the permissible range of the pixel value of each characteristic point.For example, the pixel value scope of nose is [251,255] in these 10 eigenmatrixes, and the pixel value scope of place between the eyebrows is [250,254].
Three-dimensional template is set up module 31 and is also set up the three-dimensional face template according to the range information between each point and the camera lens 21 in the three-dimensional face image of memory 50 storages, is used to store the permissible range of the pixel value of each characteristic point on the three-dimensional face, the concrete introduction as follows:
Every the three-dimensional face image that stores in the analyzing stored device 50; Obtain in this three-dimensional face image the range data of each characteristic point (for example eyes, nose, place between the eyebrows, lip, eyebrow etc.) to camera lens 21 on the face contour, and convert this range data into eigenmatrix that pixel value (span is 0~255) is stored to this three-dimensional face image.After three-dimensional template is set up module 31 and also is used for the eigenmatrix of all three-dimensional face images alignd according to one or more characteristic points (for example eyes) of setting; Pixel value to same characteristic features point in all eigenmatrixes carries out the pointwise statistics, obtains face's three-dimensional template of the permissible range composition of the pixel value of each characteristic point on the three-dimensional face.The alignment schemes of the eigenmatrix of three-dimensional face is similar to the alignment schemes of the eigenmatrix of the human-like image of aforementioned three-dimensional; The statistics of the permissible range of the characteristic point pixel value of three-dimensional face is similar to the statistics of permissible range of the characteristic point pixel value of the human-like image of aforementioned three-dimensional, gives unnecessary details no longer for example at this.
Image information processing module 32 is obtained the scene image (scene image A as shown in Figure 8) of the guarded region that video camera 100 takes, and this converts each point in the scene image into eigenmatrix that pixel value is stored to this scene image to the distance of camera lens 21.
Three-dimensional human-like detecting module 33 compares the permissible range of the pixel value of individual features point in the pixel value of each point in the eigenmatrix of this scene image and the three-dimensional human-like template of all kinds (for example front, side and the back side); Judge whether this scene image exists a certain zone, this zone to have to satisfy the pixel value of the characteristic point of first preset number to fall into the permissible range of pixel value of the three-dimensional human-like template individual features point of certain type (for example positive, side or the back side), to detect whether three-dimensional human-like zone is arranged in this scene image.For example; The eigenmatrix of supposing scene image is a 800*600 matrix; And the eigenmatrix of positive three-dimensional human-like template is a 100*100 matrix; That is three-dimensional human-like template that should the front stored the permissible range of the pixel value of 100*100 characteristic point, and first preset number is 80% of the number of the characteristic point that stores in the three-dimensional human-like template more than or equal to the front.Then three-dimensional human-like detecting module 33 reads 100*100 characteristic point at every turn in the eigenmatrix of scene image; And with the pixel value of this 100*100 characteristic point respectively with the three-dimensional human-like template in front in the permissible range of pixel value of individual features point compare; If have at least in this 100*100 characteristic point the pixel value of 80% characteristic point to fall into the permissible range of the pixel value of positive three-dimensional human-like template individual features point, then three-dimensional human-like detecting module 33 judges that the zone of this 100*100 characteristic point correspondence is three-dimensional human-like zone (like the rectangular region of alphabetical a sign among Fig. 8).The comparative approach of the three-dimensional human-like template of this scene image and other types (for example side, the back side) is similar to the comparative approach of above-mentioned with positive three-dimensional human-like template, repeats no more at this.
For example; As shown in Figure 8; If three-dimensional human-like regional a is in the lower right of scene image A, then control module 35 is assigned first control command and " is moved " controls lens 21 to the right down and move right that the back is downward-sloping to be overlapped with the center of this scene image A up to the human-like regional a of three-dimensional, and camera lens 21 stops to move.Afterwards, control module 35 is assigned second control command " amplify (zoom in) " controls lens 21 and is transferred big focal length to reach 45% up to the human-like regional a of three-dimensional proportion in this scene image A.
Image information processing module 32 should convert each point in the human-like image of three-dimensional into eigenmatrix that pixel value is stored to the human-like image of this three-dimensional to the distance of camera lens 21.
Three-dimensional face identification module 34 should the eigenmatrix of the human-like image of three-dimensional in pixel value and the three-dimensional face template of each point the permissible range of the pixel value of individual features point compare; Judge that the pixel value that whether exists a certain zone, this zone that the characteristic point that satisfies second preset number is arranged in the human-like image of this three-dimensional falls into the permissible range of the pixel value of three-dimensional face template individual features point, to detect whether the three-dimensional face zone is arranged in this scene image.For example; If have a certain regional b among the three-dimensional human-like image B of Fig. 9; This zone b has the pixel value of the characteristic point (for example this zone has 85% characteristic point at least) that satisfies second preset number to fall into the permissible range of the pixel value of three-dimensional face template individual features point, and then three-dimensional face identification module 34 judges zone b to be the three-dimensional face zone.The comparative approach of three-dimensional face identification module 34 is similar to the comparative approach of the human-like detecting module 33 of aforesaid three-dimensional, repeats no more at this.
Like Fig. 3 and shown in Figure 4, be the flow chart of camera control method preferred embodiment of the present invention.
Step S301,100 pairs of guarded regions of video camera are taken, and obtain the scene image (scene image A as shown in Figure 8) of guarded region.
Step S303, image information processing module 32 converts each point in this scene image into eigenmatrix that pixel value is stored to this scene image to the distance of camera lens 21.
Step S305; Three-dimensional human-like detecting module 33 with the pixel value of each point in the eigenmatrix of this scene image respectively with the three-dimensional human-like template of all kinds (for example positive, side and the back side) in the permissible range of pixel value of individual features point compare, to detect whether three-dimensional human-like zone is arranged in this scene image.Comparative approach sees also the aforementioned example that provides.
Step S307, three-dimensional human-like detecting module 33 judge whether this scene image exists a certain zone, this zone to have to satisfy the pixel value of the characteristic point of first preset number to fall into the permissible range of pixel value of the three-dimensional human-like template individual features point of certain type (for example positive, side or the back side).If the pixel value that has no in this scene image the zone to have the characteristic point that satisfies first preset number falls into the permissible range of pixel value of certain type three-dimensional human-like template individual features point; Then three-dimensional human-like detecting module 33 judges that appearance is three-dimensional human-like in the current scene image; Flow process is returned step S301, and video camera 100 continues guarded region is taken.If (for example the pixel value that a) has a characteristic point (having 80% characteristic point at least) that satisfies first preset number of the rectangular region shown in Fig. 8 falls into the permissible range of the pixel value of positive three-dimensional human-like template individual features point, then flow process entering step S309 to have a certain zone in this scene image.
Step S309, three-dimensional human-like detecting module 33 judges the zone to be three-dimensional human-like zone.For example, three-dimensional human-like detecting module 33 judges that the rectangular region a among the scene image A shown in Figure 8 is three-dimensional human-like zone.
Step S311, control module 35 is assigned the first control command controls lens 21 according to the positional information of the human-like zone of this three-dimensional in scene image and is made corresponding inclination, translation, overlaps with the center of this scene image up to the center in the human-like zone of three-dimensional.For example; As shown in Figure 8; If three-dimensional human-like regional a is in the lower right of scene image A, then control module 35 is assigned first control command and " is moved " controls lens 21 to the right down and move right that the back is downward-sloping to be overlapped with the center of this scene image A up to the human-like regional a of three-dimensional, and camera lens 21 stops to move.
Step S313, control module 35 is assigned the second control command controls lens focusing and is adjusted accordingly and make three-dimensional human-like zone proportion in scene image satisfy the first preset ratio requirement.For example, control module 35 is assigned second control command " amplify (zoom in) " controls lens 21 and is transferred big focal length to reach 45% up to the human-like regional a of three-dimensional proportion in scene image B as shown in Figure 9.
Step S315, video camera 100 is taken and is obtained three-dimensional clearly human-like image (as shown in Figure 9), and is stored to memory 50.
Step S317, image information processing module 32 will be somebody's turn to do and be converted each point in the human-like image of three-dimensional into eigenmatrix that pixel value is stored to the human-like image of this three-dimensional to the distance of camera lens 21.
Step S319; Three-dimensional face identification module 34 should the eigenmatrix of the human-like image of three-dimensional in pixel value and the three-dimensional face template of each point the permissible range of the pixel value of individual features point compare, should in the human-like image of three-dimensional whether the three-dimensional face zone be arranged with detecting.The comparative approach of three-dimensional face identification module 34 is similar to the comparative approach of the human-like detecting module 33 of aforesaid three-dimensional, repeats no more at this.
Step S321, three-dimensional face identification module 34 judge that the pixel value that whether exists a certain zone, this zone to have the characteristic point that satisfies second preset number in the human-like image of this three-dimensional falls into the permissible range of the pixel value of three-dimensional face template individual features point.If the pixel value that has no the zone to have the characteristic point that satisfies second preset number in the human-like image of this three-dimensional falls into the permissible range of the pixel value of three-dimensional face template individual features point; Then three-dimensional face identification module 34 is judged and should three-dimensional face do not occurred in the human-like image of three-dimensional, and flow process is returned step S315.Otherwise; If the pixel value that exists a certain zone (having regional b among the three-dimensional human-like image B for example shown in Figure 9) to have the characteristic point (for example this zone has 85% characteristic point at least) that satisfies second preset number in the human-like image of this three-dimensional falls into the permissible range of the pixel value of three-dimensional face template individual features point, then flow process gets into step S323.
Step S323, three-dimensional face identification module 34 judge the zone to be the three-dimensional face zone.For example, three-dimensional face identification module 34 judges that rectangular region b is the three-dimensional face zone among the three-dimensional human-like image B shown in Figure 9.
Step S325, control module 35 is assigned the 3rd control command controls lens 21 according to the positional information of this three-dimensional face zone in the human-like image of three-dimensional and is made corresponding translation, overlaps with the center of the human-like image of this three-dimensional up to the center in this three-dimensional face zone.For example; As shown in Figure 9; If the center of three-dimensional face zone b is directly over the center of the human-like image B of three-dimensional; Then control module 35 is assigned the 3rd control command the move up center of up to three-dimensional face zone b of controls lens 21 that " moves up " and is overlapped with the center of the human-like image B of this three-dimensional, and camera lens 21 stops to move.
Step S327, control module 35 is assigned the 4th control command controls lens focusing and is adjusted accordingly and make this three-dimensional face zone proportion in the human-like image of three-dimensional satisfy the second preset ratio requirement.For example, control module 35 is assigned the 4th control command " amplify (zoom in) " controls lens 21 and is transferred big focal length to reach 33% up to three-dimensional face zone b proportion in the human-like image of three-dimensional.
Step S329, video camera 100 is taken and is obtained three-dimensional face image (image C shown in figure 10) clearly, and is stored to memory 50.
Claims (14)
1. a video camera is characterized in that, this video camera comprises memory, driver element, camera lens and control unit, wherein:
Memory storage video camera three-dimensional human-like image and the three-dimensional face image taken in advance;
Control unit is according to the range information of each point in the image to camera lens; The scene image of the guarded region that video camera is taken and the three-dimensional human-like image of memory storage compare, analyze; To detect the three-dimensional human-like zone in this scene image; And drive camera lens according to positional information and the proportion control drive unit of the three-dimensional human-like zone that detects in scene image and carry out corresponding mobile and focal length adjustment operation, obtain three-dimensional clearly human-like image with shooting;
Control unit; Also according to the range information of each point in the image to camera lens; The three-dimensional face image of this three-dimensional clearly human-like image and memory storage is compared, analyzes; Three-dimensional face to detect in this three-dimensional clearly human-like image is regional, and carries out corresponding mobile and focal length adjustment operation according to the positional information and the proportion control drive unit driving camera lens of the three-dimensional face zone that detects in the human-like image of three-dimensional, obtains three-dimensional face image clearly with shooting.
2. video camera as claimed in claim 1 is characterized in that, said control unit comprises:
Three-dimensional template is set up module; Be used for setting up three-dimensional human-like template according to the three-dimensional human-like image each point of memory storage and the range information between the camera lens; And set up the three-dimensional face template according to the range information between each point and the camera lens in the three-dimensional face image of memory storage; The human-like template of wherein said three-dimensional has stored the permissible range of the pixel value of each characteristic point on the 3 D human body profile, and said three-dimensional face template has stored the permissible range of the pixel value of each characteristic point on the three-dimensional face;
The image information processing module is used to obtain the scene image of the guarded region that video camera takes, and this converts each point in the scene image into eigenmatrix that pixel value is stored to this scene image to the distance of camera lens;
Three-dimensional human-like detecting module; Be used for the permissible range of the pixel value of individual features point in the pixel value of the eigenmatrix each point of this scene image and the three-dimensional human-like template is compared; Judge whether this scene image exists a certain zone, this zone to have to satisfy the pixel value of the characteristic point of first preset number to fall into the permissible range of the pixel value of three-dimensional human-like template individual features point, to detect the three-dimensional human-like zone in this scene image;
Control module is used for according to the three-dimensional human-like zone that detects after the positional information of scene image and proportion controls lens are carried out corresponding mobile and focal length adjustment operation, taking and obtaining three-dimensional clearly human-like image;
The image information processing module also is used for converting the human-like image each point of this three-dimensional into eigenmatrix that pixel value is stored to the human-like image of this three-dimensional to the distance of camera lens;
The three-dimensional face identification module; The permissible range that is used for the pixel value of individual features point in pixel value and the three-dimensional face template of eigenmatrix each point that should the human-like image of three-dimensional compares; Judge that the pixel value that whether exists a certain zone, this zone that the characteristic point that satisfies second preset number is arranged in the human-like image of this three-dimensional falls into the permissible range of the pixel value of three-dimensional face template individual features point, to detect the three-dimensional face zone in this scene image;
Control module also is used for according to the three-dimensional face zone that detects after the positional information of the human-like image of three-dimensional and proportion controls lens are carried out corresponding mobile and focal length adjustment operation, taking and obtaining three-dimensional face image clearly.
3. video camera as claimed in claim 2 is characterized in that, said three-dimensional template is set up module and " set up three-dimensional human-like template according to the range information between each point and the camera lens in the three-dimensional human-like image of memory storage " and comprising:
The every Zhang San who stores in the analyzing stored device ties up human-like image, obtains in the human-like image of this three-dimensional the range data of each characteristic point to camera lens on the human body contour outline, and converts this range data into eigenmatrix that pixel value is stored to the human-like image of this three-dimensional; And
After the eigenmatrix of all three-dimensional human-like images of same type alignd according to a characteristic point of setting; Pixel value to the characteristic point on this type human body carries out the pointwise statistics, obtains the three-dimensional human-like template of the permissible range composition of the pixel value of each characteristic point on this type human body contour outline.
4. video camera as claimed in claim 3; It is characterized in that, the human-like detecting module of said three-dimensional be with the pixel value of each point in the eigenmatrix of this scene image respectively with the human-like template of various types of three-dimensionals in the permissible range of pixel value of individual features point compare.
5. video camera as claimed in claim 2 is characterized in that, said three-dimensional template is set up module and " set up the three-dimensional face template according to the range information between each point and the camera lens in the three-dimensional face image of memory storage " and comprising:
Every the three-dimensional face image that stores in the analyzing stored device obtains in this three-dimensional face image the range data of each characteristic point to camera lens on the face contour, and converts this range data into eigenmatrix that pixel value is stored to this three-dimensional face image; And
After the eigenmatrix of all three-dimensional face images alignd according to one or more characteristic points of setting; Pixel value to same characteristic features point in all eigenmatrixes carries out the pointwise statistics, obtains the three-dimensional face template of the permissible range composition of the pixel value of each characteristic point on the three-dimensional face.
6. video camera as claimed in claim 2 is characterized in that, said control module " is carried out corresponding mobile and focal length adjustment operation according to the positional information and the proportion controls lens of the human-like zone of this three-dimensional in scene image " and being comprised:
Assign the first control command controls lens according to the positional information of the human-like zone of this three-dimensional in scene image and make corresponding inclination, translation, overlap with the center of this scene image up to the center in this human-like zone of three-dimensional; And
Assigning the second control command controls lens focusing adjusts accordingly and makes three-dimensional human-like zone proportion in this scene image satisfy the first preset ratio requirement.
7. video camera as claimed in claim 2 is characterized in that, wherein, said control module " is carried out corresponding mobile and focal length adjustment operation according to the positional information and the proportion controls lens of this three-dimensional face zone in the human-like image of three-dimensional " and being comprised:
Assign the 3rd control command controls lens according to the positional information of this three-dimensional face zone in the human-like image of three-dimensional and make corresponding translation, overlap with the center of the human-like image of this three-dimensional up to the center in this three-dimensional face zone; And
Assigning the 4th control command controls lens focusing adjusts accordingly and makes this three-dimensional face zone proportion in the human-like image of this three-dimensional satisfy the second preset ratio requirement.
8. the control method of a video camera is characterized in that, this method comprises:
Three-dimensional human-like area detecting step: according to the range information of each point in the image scene image of the guarded region of video camera shooting and the three-dimensional human-like image of memory storage are compared, analyze, to detect the three-dimensional human-like zone in this scene image to camera lens;
First controlled step: carry out corresponding mobile and focal length adjustment operation according to the positional information and the proportion control drive unit driving camera lens of the three-dimensional human-like zone that detects in scene image, obtain three-dimensional clearly human-like image with shooting;
Three-dimensional face area detecting step: compare, analyze to the range information of camera lens three-dimensional face image according to each point in the image, to detect the three-dimensional face zone in this three-dimensional clearly human-like image with this three-dimensional clearly human-like image and memory storage; And
Second controlled step: carry out corresponding mobile and focal length adjustment operation according to the positional information and the proportion control drive unit driving camera lens of the three-dimensional face zone that detects in the human-like image of three-dimensional, obtain three-dimensional face image clearly with shooting.
9. the control method of video camera as claimed in claim 8 is characterized in that, the human-like area detecting step of said three-dimensional comprises:
Three-dimensional human-like template establishment step: set up three-dimensional human-like template according to the range information between each point and the camera lens in the three-dimensional human-like image of memory storage, be used to store the permissible range of the pixel value of each characteristic point on the 3 D human body profile;
The first image information treatment step: obtain the scene image of the guarded region that video camera takes, convert each point in this scene image into eigenmatrix that pixel value is stored to this scene image to the distance of camera lens; And
Detecting step: the permissible range of the pixel value of individual features point in the pixel value of each point in the eigenmatrix of this scene image and the three-dimensional human-like template is compared; Judge whether this scene image exists a certain zone, this zone to have to satisfy the pixel value of the characteristic point of first preset number to fall into the permissible range of the pixel value of three-dimensional human-like template individual features point, to detect the three-dimensional human-like zone in this scene image.
10. the control method of video camera as claimed in claim 9 is characterized in that, " sets up three-dimensional human-like template according to the range information between each point and the camera lens in the three-dimensional human-like image of memory storage " in the human-like template establishment step of said three-dimensional and comprising:
The every Zhang San who stores in the analyzing stored device ties up human-like image, obtains in the human-like image of this three-dimensional the range data of each characteristic point to camera lens on the human body contour outline, and converts this range data into eigenmatrix that pixel value is stored to the human-like image of this three-dimensional; And
After the eigenmatrix of all three-dimensional human-like images of same type alignd according to a characteristic point of setting; Pixel value to the characteristic point on this type human body carries out the pointwise statistics, obtains the three-dimensional human-like template of the permissible range composition of the pixel value of each characteristic point on this type human body contour outline.
11. the control method of video camera as claimed in claim 8; It is characterized in that, " carry out corresponding mobile and focal length adjustment operation " in said first controlled step and comprising according to the positional information and the proportion control drive unit driving camera lens of the human-like zone of this three-dimensional in scene image:
Assign the first control command controls lens according to the positional information of the human-like zone of this three-dimensional in scene image and make corresponding inclination, translation, overlap with the center of this scene image up to the center in this human-like zone of three-dimensional; And
Assigning the second control command controls lens focusing adjusts accordingly and makes three-dimensional human-like zone proportion in this scene image satisfy the first preset ratio requirement.
12. the control method of video camera as claimed in claim 8 is characterized in that, said three-dimensional face area detecting step comprises:
Three-dimensional face template establishment step: set up the three-dimensional face template according to the range information between each point and the camera lens in the three-dimensional face image of memory storage, be used to store the permissible range of the pixel value of each characteristic point on the three-dimensional face;
The second image information treatment step: each point converts the eigenmatrix that pixel value is stored to the human-like image of this three-dimensional into to the distance of camera lens in should the human-like image of three-dimensional; And
Identification step: in should the eigenmatrix of the human-like image of three-dimensional in the pixel value of each point and the three-dimensional face template permissible range of the pixel value of individual features point compare; Judge that the pixel value that whether exists a certain zone, this zone that the characteristic point that satisfies second preset number is arranged in the human-like image of this three-dimensional falls into the permissible range of the pixel value of three-dimensional face template individual features point, to detect the three-dimensional face zone in this scene image.
13. the control method of video camera as claimed in claim 12 is characterized in that, " sets up the three-dimensional face template according to the range information between each point and the camera lens in the three-dimensional face image of memory storage " in the said three-dimensional face template establishment step and comprising:
Every the three-dimensional face image that stores in the analyzing stored device obtains in this three-dimensional face image the range data of each characteristic point to camera lens on the face contour, and converts this range data into eigenmatrix that pixel value is stored to this three-dimensional face image; And
After the eigenmatrix of all three-dimensional face images alignd according to one or more characteristic points of setting; Pixel value to same characteristic features point in all eigenmatrixes carries out the pointwise statistics, obtains the three-dimensional face template of the permissible range composition of the pixel value of each characteristic point on the three-dimensional face.
14. the control method of video camera as claimed in claim 8; It is characterized in that, " carry out corresponding mobile and focal length adjustment operation " in said second controlled step and comprising according to the positional information and the proportion control drive unit driving camera lens of this three-dimensional face zone in the human-like image of three-dimensional:
Assign the 3rd control command controls lens according to the positional information of this three-dimensional face zone in the human-like image of three-dimensional and make corresponding translation, overlap with the center of the human-like image of this three-dimensional up to the center in this three-dimensional face zone; And
Assigning the 4th control command controls lens focusing adjusts accordingly and makes this three-dimensional face zone proportion in the human-like image of this three-dimensional satisfy the second preset ratio requirement.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2010102329119A CN102340628A (en) | 2010-07-21 | 2010-07-21 | Camera and control method thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2010102329119A CN102340628A (en) | 2010-07-21 | 2010-07-21 | Camera and control method thereof |
Publications (1)
Publication Number | Publication Date |
---|---|
CN102340628A true CN102340628A (en) | 2012-02-01 |
Family
ID=45516113
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2010102329119A Pending CN102340628A (en) | 2010-07-21 | 2010-07-21 | Camera and control method thereof |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102340628A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104867148A (en) * | 2015-04-22 | 2015-08-26 | 北京爱普力思健康科技有限公司 | Method and apparatus for obtaining images of predetermined type of objects and remote oral diagnosis system |
CN105234940A (en) * | 2015-10-23 | 2016-01-13 | 上海思依暄机器人科技有限公司 | Robot and control method thereof |
CN107277359A (en) * | 2017-07-13 | 2017-10-20 | 深圳市魔眼科技有限公司 | Method, device, mobile terminal and the storage medium of adaptive zoom in 3D scannings |
CN104021569B (en) * | 2013-02-28 | 2018-03-20 | 杭州海康威视数字技术股份有限公司 | Human body target locking tracking apparatus and method |
TWI761047B (en) * | 2021-01-22 | 2022-04-11 | 明泰科技股份有限公司 | Method for verifying detection range of image detector |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1529278A (en) * | 2003-10-09 | 2004-09-15 | 重庆大学 | Multi-category-based human face classifying and identifying method |
US20080130985A1 (en) * | 2006-12-02 | 2008-06-05 | Electronic And Telecommunications Research Institute | Correlation extract method for generating 3d motion data, and motion capture system and method for easy composition of humanoid character on real background image using the same |
CN101448142A (en) * | 2007-11-27 | 2009-06-03 | 鸿富锦精密工业(深圳)有限公司 | Image tracking device and image tracking method thereof |
CN101540890A (en) * | 2009-04-28 | 2009-09-23 | 南京航空航天大学 | Method for obtaining a clear face image of a moving human body in a monitored video |
-
2010
- 2010-07-21 CN CN2010102329119A patent/CN102340628A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1529278A (en) * | 2003-10-09 | 2004-09-15 | 重庆大学 | Multi-category-based human face classifying and identifying method |
US20080130985A1 (en) * | 2006-12-02 | 2008-06-05 | Electronic And Telecommunications Research Institute | Correlation extract method for generating 3d motion data, and motion capture system and method for easy composition of humanoid character on real background image using the same |
CN101448142A (en) * | 2007-11-27 | 2009-06-03 | 鸿富锦精密工业(深圳)有限公司 | Image tracking device and image tracking method thereof |
CN101540890A (en) * | 2009-04-28 | 2009-09-23 | 南京航空航天大学 | Method for obtaining a clear face image of a moving human body in a monitored video |
Non-Patent Citations (1)
Title |
---|
吴翔等: "基于双目视觉的立体人脸识别***", 《计算机***应用》, no. 5, 31 December 2009 (2009-12-31), pages 61 - 64 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104021569B (en) * | 2013-02-28 | 2018-03-20 | 杭州海康威视数字技术股份有限公司 | Human body target locking tracking apparatus and method |
CN104867148A (en) * | 2015-04-22 | 2015-08-26 | 北京爱普力思健康科技有限公司 | Method and apparatus for obtaining images of predetermined type of objects and remote oral diagnosis system |
CN105234940A (en) * | 2015-10-23 | 2016-01-13 | 上海思依暄机器人科技有限公司 | Robot and control method thereof |
CN107277359A (en) * | 2017-07-13 | 2017-10-20 | 深圳市魔眼科技有限公司 | Method, device, mobile terminal and the storage medium of adaptive zoom in 3D scannings |
TWI761047B (en) * | 2021-01-22 | 2022-04-11 | 明泰科技股份有限公司 | Method for verifying detection range of image detector |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109691079B (en) | Imaging device and electronic apparatus | |
EP4105101A1 (en) | Monitoring system, monitoring method, and monitoring device for railway train | |
US8754945B2 (en) | Image capturing device and motion tracking method | |
CN102348102B (en) | Roof safety monitoring system and method thereof | |
CN101814130A (en) | Iris identification device by using camera array and multimodal biometrics identification method | |
CN103168467B (en) | The security monitoring video camera using heat picture coordinate is followed the trail of and monitoring system and method | |
US20130076943A1 (en) | Apparatus and method for image recognition of facial areas in photographic images from a digital camera | |
WO2020029921A1 (en) | Monitoring method and device | |
US20140176764A1 (en) | Information processing device and recording medium | |
JP2007504562A (en) | Method and apparatus for performing iris authentication from a single image | |
US8520130B2 (en) | Image capturing apparatus and auto-focus control method thereof | |
KR100691348B1 (en) | Method for tracking moving target with using stereo camera based on pan/tilt contol and system implementing thereof | |
CN101106700A (en) | Intelligent target detail capturing device and method in video monitoring system | |
CN109451233B (en) | Device for collecting high-definition face image | |
WO2013126567A1 (en) | Method and apparatus for robustly collecting facial, ocular, and iris images using a single sensor | |
CN111246172A (en) | Linkage image acquisition device and image acquisition method | |
DE19955714A1 (en) | Method and apparatus for photographing / recognizing a face | |
CN102340628A (en) | Camera and control method thereof | |
CN102855471A (en) | Remote iris intelligent imaging device and method | |
CN103905727A (en) | Object area tracking apparatus, control method, and program of the same | |
US8406468B2 (en) | Image capturing device and method for adjusting a position of a lens of the image capturing device | |
DE102018101034A1 (en) | Image processing apparatus and image processing method | |
US20110074973A1 (en) | Camera and recording method therefor | |
CN115546738A (en) | Rail foreign matter detection method | |
JP6265602B2 (en) | Surveillance camera system, imaging apparatus, and imaging method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120201 |