CN107038418A - A kind of intelligent air condition dual camera follows the trail of the method for obtaining clear human body image - Google Patents
A kind of intelligent air condition dual camera follows the trail of the method for obtaining clear human body image Download PDFInfo
- Publication number
- CN107038418A CN107038418A CN201710182512.8A CN201710182512A CN107038418A CN 107038418 A CN107038418 A CN 107038418A CN 201710182512 A CN201710182512 A CN 201710182512A CN 107038418 A CN107038418 A CN 107038418A
- Authority
- CN
- China
- Prior art keywords
- camera
- image
- face
- angle
- common camera
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/61—Control of cameras or camera modules based on recognised objects
- H04N23/611—Control of cameras or camera modules based on recognised objects where the recognised objects include parts of the human body
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/62—Control of parameters via user interfaces
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/70—Circuitry for compensating brightness variation in the scene
- H04N23/73—Circuitry for compensating brightness variation in the scene by influencing the exposure time
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The method for obtaining clear human body image is followed the trail of the present invention relates to a kind of intelligent air condition dual camera, it on air-conditioning by adding wide-angle camera, common camera and can drive the stepper motor that common camera is rotated come the spatial dimension for solving to obtain image is small and the problem of not clear enough detail pictures.Wherein, wide-angle imaging obtains the global image of wide viewing angle and obtains the big image of spatial dimension to realize, stepper motor drives common camera to carry out back rotation in the angular field of view of wide viewing angle camera, catches and follow the trail of human body, to obtain clearly detail pictures.
Description
Technical field
The method for obtaining clear human body image is followed the trail of the present invention relates to a kind of intelligent air condition dual camera.
Background technology
Existing air conditioner intelligent is a big inevitable trend.It is very that speech recognition and image recognition are added on air-conditioning
Feasible mode.Image acquisition is related to, is typically all that a fixed camera is installed on air-conditioning to pass through the shooting
Head obtains image.
Due to only one of which camera, influenceed to have the disadvantage that by hardware factor:
1st, the spatial dimension according to the high camera of definition then acquired image is smaller;
If the 2, the definition for the image detail that the image range that selection wide-angle camera is obtained increases but obtained is not
It is high.
Therefore this technical scheme tends not to meet intelligent air condition acquisition wide viewing angle well and retains many images
The demand of details.Other existing technical scheme is usually the indiscriminate acquisition image for there be not emphasis in the part of acquisition image,
This point is not intelligent enough, can waste calculation resources.
The content of the invention
The technical problems to be solved by the invention are to provide a kind of intelligent air condition dual camera and follow the trail of the clear human figure of acquisition
The method of picture, it on air-conditioning by installing wide-angle camera, common camera and can drive the step of common camera rotation
Stepper motor solves to obtain the problem of spatial dimension is small and detail pictures are not clear enough of image.
The technical solution adopted by the present invention is,
A kind of intelligent air condition dual camera follows the trail of the method for obtaining clear human body image, and wide-angle is installed on intelligent air condition and is taken the photograph
As head, common camera and can drive common camera be scanned pattern rotate or tracking pattern rotate stepping electricity
Machine;The advanced row Face datection of image of wide-angle camera is read by pattern process module, only when wide-angle camera is detected
Face datection further just is carried out to the image that common camera under scan pattern is gathered during face, if now common camera does not have
There is detection face, then the image of output module reading wide-angle camera is exported, if now common camera detects face,
Then the image of output module reading common camera is exported, meanwhile, step motor control common camera enters tracking mould
Formula is tracked to the image face of subsequent acquisition, calculate image in face location camera center offset direction and partially
Angle is moved, camera direction is adjusted according to the offset direction and deviation angle, until face tracking is lost, then step motor control
Common camera enters scan pattern, and the advanced row Face datection of image of wide-angle camera, weight are read by pattern process module
Multiple above-mentioned work.
Further, methods described specifically includes following steps,
Wide-angle camera starts to obtain image after step 1, start, and step motor control common camera enters scanning mould
Formula, control the common camera in the angular field of view of wide-angle camera from left to right, particles from right to left;
Step 2, pattern process module read the advanced row Face datection of image of wide-angle camera, only work as wide-angle camera
Face datection further just is carried out to the image that common camera under scan pattern is gathered when detecting face, if now commonly taking the photograph
As head does not detect face, then output module reads the image of wide-angle camera and exported, return to step 2;If now common
Camera detects face, then the image of output module reading common camera is exported, into step 3;
Step 3, step motor control common camera are tracked into tracing mode to the image face of subsequent acquisition,
Offset direction and deviation angle of the face location in image at camera center are calculated, according to the offset direction and deviation angle
Adjust camera direction so that face location is in camera center, and when face tracking is lost, step motor control is commonly taken the photograph
As head enters scan pattern, while going to step 2.
Further, increase by a step before the image that pattern process module reads wide-angle camera carries out Face datection,
I.e. pattern process module reads the image of wide-angle camera and common camera collection simultaneously, and brightness judgement and brightness are carried out to it
Adjustment, if the brightness value of image is not in default qualitative picture brightness section, adjusts the exposure of two cameras respectively
Time and ISO sensitivity simultaneously abandon the image, until the brightness value of the image of two cameras is bright in default qualitative picture
In degree is interval.
Further, when carrying out Face datection, image is changed into gray-scale map, and gray-scale map is regarded as a m*n
Picture element matrix, then set up coordinate system on picture element matrix, top left co-ordinate is origin (0,0), x-axis represent width from a left side to
Right increase, y-axis represents that height increases from top to bottom, then with a default minimum face frame (Wmin,Hmin), face frame
Initial coordinate is (0,0), and the slip since the image upper left corner from left to right from top to bottom has been handled until whole image;
Then the gradually size of increase face frame and the slip in whole sub-picture again, until face frame increases to the maximum of default face frame
It is worth (Wmax, Hmax);The pixel composition piece image just taken out in face frame a position is slided into every time carries out Face datection,
The coordinate (x, y) and size (width, height) in the upper left corner of human face frame are recorded if face is detected, wherein, Wmin
≤width≤Wmax, Hmin≤height≤Hmax。
Further, when common camera is in tracing mode, pattern process module reads the image of the common camera
And carry out Face datection, when detecting face, record human face frame the upper left corner coordinate (x, y) and size (width,
Height), judge whether face location is in the central area of the central area i.e. image of common camera of common camera
Position, if face location is in the central area of common camera, i.e. x<m/2<During x+width, then camera is not rotated, figure
The image that processing module continues to read common camera carries out Face datection;If face location is not in common camera
Heart district domain, i.e. m/2<X or m/2>During x+width, then angle [alpha]=((x+width/ that common camera needs to rotate is calculated
2)-m/2)/m*30 °,
If α=0, common camera need not be rotated, i.e., the angle that need to be rotated is zero;
If α>0, common camera needs the angle of turn to be α, that is, needs to obtain the right side of image to common camera
Side is rotated;
If α<0, common camera needs counter-rotational angle to be α, that is, needs to obtain image to common camera
The left side is rotated;
Common camera rotation direction of the controllor for step-by-step motor according to obtained by calculating and the tooth ratio between angle and gear,
Conversion calculates the rotation direction and rotational angle of stepper motor;Then further according to the rotation direction and rotational angle of stepper motor
Generation pulse control signal is sent to stepper motor driver, and stepper motor driver is according to received pulse control signal
Powered to stepper motor, stepper motor is rotated due angle towards the direction calculated, common shooting is driven by gear
Head turns to the position that face in image is currently obtained towards common camera, and then output module reads the common camera and obtained
The image taken is exported, and common camera proceeds face tracking.
Further, the visual angle of the wide-angle camera is 90 °, and the visual angle of common camera is 30 °.
The present invention on air-conditioning by adding wide-angle camera, common camera and can drive what common camera was rotated
Stepper motor solves to obtain the problem of spatial dimension is small and detail pictures are not clear enough of image.Wherein, wide-angle imaging is obtained
The global image of wide viewing angle obtains the big image of spatial dimension to realize, first carries out face inspection to the image that wide-angle camera is gathered
Survey, when only detecting face, Face datection, stepping electricity are just further carried out to the high-definition image that common camera is gathered
Machine control common camera enters tracing mode, and adjusts the angle of camera, accurately catches and follows the trail of face, it is meant that works as place
In under scan pattern, Face datection only is carried out to the image that wide-angle camera is gathered, the visual angle of image is not only obtained than larger, and
Amount of calculation is relatively low, once into tracing mode, then the image higher to definition progress Face datection, catches simultaneously so as to accurate
Face is followed the trail of, the present invention can reach more accurate, real-time image detection effect using limited computing resource.
Brief description of the drawings
Fig. 1 is flow chart of the present invention;
Fig. 2 is the common camera tracing mode flow chart of the present invention.
Embodiment
The method for obtaining clear human body image is followed the trail of the invention discloses a kind of intelligent air condition dual camera, it passes through in sky
Wide-angle camera, common camera are added on tune and the stepper motor that common camera is rotated can be driven to solve to obtain image
The problem of spatial dimension is small and detail pictures are not clear enough.Wherein, wide-angle imaging obtains the global image of wide viewing angle to realize
The big image of spatial dimension is obtained, step motor control common camera is turned round in the angular field of view of wide viewing angle camera
It is dynamic, catch and follow the trail of human body, to obtain clearly detail pictures.Shown in reference picture 1, the following institute of specific method of the invention
Show.
Wide-angle camera and common camera start to obtain image, step motor control common camera after step 1, start
Open scan pattern, the mode of rotation be in the angular field of view of wide viewing angle camera from left to right, from right to left carry out flyback
Retouch.
In the present embodiment, the visual angle of wide-angle camera is that the visual angle of 90 ° of common cameras is 30 °.It is general under equal pixel
Image and the appropriate section contrast gone out from the image interception of wide-angle camera acquisition acquired in logical camera, pixel is wide-angle
3 times of camera, picture detail is also 3 times of wide-angle camera.So, commonly imaged when wide-angle camera detects face
Head is not when detecting face, and the image that output module reads wide-angle camera is simultaneously exported, with meet people to image compared with
Demand in terms of big spatial dimension.And when detecting face in common camera, output module reads common camera
Image is exported, to meet people to the demand in terms of the details of image.
Step 2, pattern process module read wide-angle camera and the image of common camera respectively, to wide-angle camera and
The image of common camera carries out brightness judgement and brightness adjustment.
When carrying out brightness judgement, see respectively the image that wide-angle camera and common camera are obtained brightness value whether
In default qualitative picture brightness section, if the brightness value of image is distinguished not in default qualitative picture brightness section
The time for exposure of two cameras of adjustment and ISO sensitivity simultaneously abandon the image, until the brightness value of the image of two cameras
In default qualitative picture brightness section, subsequently into step 3.
The adjustable strategies of time for exposure and ISO sensitivity are:If brightness value is too low prolonging exposure time, heighten ISO
Sensitivity;The time for exposure is cut short if brightness value is too high, ISO sensitivity is turned down.
Step 3, pattern process module carry out Face datection to the image that wide-angle camera is obtained first, when the wide-angle imaging
When not detecting face in the image of head, into step 2;When detecting face in the image of wide-angle camera, figure inspection
Survey the image progress Face datection that module is obtained to common camera.Now, if not detected in the image of common camera
Face, the image that output module reads the wide-angle camera is exported, and enters step 2;If in the image of common camera
Face is detected, the image that output module reads common camera is exported, and enters next step.
When carrying out the Face datection of image of wide-angle camera or common camera, image is changed into gray scale first
Scheme and gray-scale map regarded as m*n picture element matrix, wherein, m is the width of pixel, and n is the height of pixel.Then in picture
Coordinate system is set up in prime matrix, top left co-ordinate is origin (0,0), and x-axis represents that width from left to right increases, and y-axis represents height
Increase from top to bottom, each pixel represents 1 value.Then with a default minimum face frame (Wmin,Hmin) from image
The upper left corner starts slip from left to right from top to bottom, has been handled until whole image;Then face frame is gradually increased again
Size is simultaneously slided in whole sub-picture, until face frame increases to the maximum (W of default face framemax, Hmax)。
Assuming that the initial coordinate of face frame is (0,0) and then first slided from left to right, a pixel is slided every time, works as slip
B times, this line in image terminates, i.e. a*b=m, then face frame enters next line and continues to repeat slip above, until handle
Whole image has been slided.Then the gradually size of increase face frame and the slip in whole sub-picture again, until face frame increases to
The maximum of default face frame.The pixel composition piece image just taken out in face frame a position is slided into every time enters pedestrian
Face is detected, the coordinate (x, y) and size (width, height) in the upper left corner of human face frame are recorded if face is detected, its
In, Wmin≤width≤Wmax, Hmin≤height≤Hmax。
Step 4, step motor control common camera are tracked into tracing mode to the image face of subsequent acquisition,
Offset direction and deviation angle of the face location in image at camera center are calculated, according to the offset direction and deviation angle
Camera direction is adjusted, when tracking loss, step motor control common camera enters scan pattern, while going to step 2.
Shown in reference picture 2, common camera is to rotate, so in tracing mode, common camera is with face
Move and move.In the present embodiment, whether the method for track face location is to judge each frame facial image in camera acquisition figure
The central area of picture, camera direction is adjusted with face location relative to the position of picture centre region.Work as common camera
Center face is all thought in face range areas at common camera center, when face is in common camera center
When, common camera is not just rotated further by.
Common camera is to carry out face tracking in tracing mode, and pattern process module reads the figure of the common camera
Picture simultaneously carries out Face datection, when detecting face, and judging face location, whether the central area in common camera is i.e. general
The central area position of the image of logical camera, if face location is in the central area of common camera, camera does not turn
Dynamic, the image that pattern process module continues to read common camera carries out Face datection;If face location is not on commonly taking the photograph
As the central area of head, then offset direction of the face location relative to camera center and deviation angle, stepper motor root are calculated
The position of camera is adjusted according to the offset direction and deviation angle, then output module reads the image of common camera acquisition
Exported, common camera proceeds face tracking.When can't detect face in the image of common camera be with
Track is lost, now, into next step.
For example, the position of face is the coordinate (x, y) and size (width, height) in the upper left corner, view picture in image
The size of image is (m, n), and m is width, and n is height.
So first determine whether whether the facial image in image is in the central area of camera, that is, judge common camera
Center whether be in face range areas.Work as x<m/2<During x+width, that is, think that face obtains image in common camera
Central area, camera stops operating.Work as m/2<X or m/2>During x+width, i.e., artificial face is not in common camera
The central area of image, camera is rotated, now, and common camera needs angle [alpha]=((the x+width/2)-m/ rotated
2)/m*30°。
If α=0, common camera need not be rotated, i.e., the angle that need to be rotated is zero;
If α>0, common camera needs the angle of turn to be α, that is, needs to obtain the right side of image to common camera
Side is rotated;
If α<0, common camera needs counter-rotational angle to be α, that is, needs to obtain image to common camera
The left side is rotated.
Common camera rotation direction of the controllor for step-by-step motor according to obtained by calculating and the tooth ratio between angle and gear,
Conversion calculates the rotation direction and rotational angle of stepper motor;Then further according to the rotation direction and rotational angle of stepper motor
Generation pulse control signal is sent to stepper motor driver.Stepper motor driver is according to received pulse control signal
Powered to stepper motor, stepper motor is rotated due angle towards the direction calculated, common shooting is driven by gear
Head turns to the position that face in image is currently obtained towards common camera.
The present invention on air-conditioning by adding wide-angle camera, common camera and can drive what common camera was rotated
Stepper motor solves to obtain the problem of spatial dimension is small and detail pictures are not clear enough of image.Wherein, wide-angle imaging is obtained
The global image of wide viewing angle obtains the big image of spatial dimension to realize, first carries out face inspection to the image that wide-angle camera is gathered
Survey, when only detecting face, Face datection, stepping electricity are just further carried out to the high-definition image that common camera is gathered
Machine control common camera enters tracing mode, and adjusts the angle of camera, accurately catches and follows the trail of face, it is meant that works as place
In under scan pattern, Face datection only is carried out to the image that wide-angle camera is gathered, the visual angle of image is not only obtained than larger, and
Amount of calculation is relatively low, once into tracing mode, then the image higher to definition progress Face datection, catches simultaneously so as to accurate
Face is followed the trail of, the present invention can reach more accurate, real-time image detection effect using limited computing resource.
The above described is only a preferred embodiment of the present invention, be not intended to limit the scope of the present invention,
Therefore the change or modification that claim under this invention and specification are done in every case, it should all belong to the scope that patent of the present invention covers
Within.
Claims (6)
1. a kind of intelligent air condition dual camera follows the trail of the method for obtaining clear human body image, it is characterised in that:On intelligent air condition
Wide-angle camera, common camera are installed and common camera can be driven to be scanned pattern and is rotated or the rotation of tracking pattern
Stepper motor;The advanced row Face datection of image of wide-angle camera is read by pattern process module, only works as wide-angle imaging
Head just further carries out Face datection when detecting face to the image that common camera under scan pattern is gathered, if now common
Camera does not detect face, then the image of output module reading wide-angle camera is exported, if now common camera is examined
Face is measured, then the image of output module reading common camera is exported, meanwhile, step motor control common camera is entered
Enter tracing mode to be tracked the image face of subsequent acquisition, calculate skew of the face location in image at camera center
Direction and deviation angle, camera direction is adjusted according to the offset direction and deviation angle, until face tracking is lost, then stepping
Motor control common camera enters scan pattern, and the advanced pedestrian's face of image of wide-angle camera is read by pattern process module
Detection, repeats above-mentioned work.
2. a kind of intelligent air condition dual camera according to claim 1 follows the trail of the method for obtaining clear human body image, it is special
Levy and be:Methods described specifically includes following steps,
Wide-angle camera starts to obtain image after step 1, start, and step motor control common camera enters scan pattern, controls
Make the common camera in the angular field of view of wide-angle camera from left to right, particles from right to left;
Step 2, pattern process module read the advanced row Face datection of image of wide-angle camera, only when wide-angle camera is detected
To just further Face datection is carried out to the image that common camera under scan pattern is gathered during face, if now common camera
Face is not detected, then the image of output module reading wide-angle camera is exported, return to step 2;If now common shooting
Head detects face, then the image of output module reading common camera is exported, into step 3;
Step 3, step motor control common camera are tracked into tracing mode to the image face of subsequent acquisition, are calculated
Offset direction and deviation angle of the face location at camera center in image, is adjusted according to the offset direction and deviation angle
Camera direction so that face location is in camera center, when face tracking is lost, step motor control common camera
Into scan pattern, while going to step 2.
3. a kind of intelligent air condition dual camera according to claim 2 follows the trail of the method for obtaining clear human body image, it is special
Levy and be:Increase a step, i.e. graphics process before the image that pattern process module reads wide-angle camera carries out Face datection
Module reads the image of wide-angle camera and common camera collection simultaneously, and brightness judgement and brightness adjustment are carried out to it, if
The brightness value of image then adjusts time for exposure and the ISO of two cameras respectively not in default qualitative picture brightness section
Sensitivity simultaneously abandons the image, until the brightness value of the image of two cameras is in default qualitative picture brightness section.
4. a kind of intelligent air condition dual camera according to claim 3 follows the trail of the method for obtaining clear human body image, it is special
Levy and be:When carrying out Face datection, image is changed into gray-scale map, and gray-scale map is regarded as m*n picture element matrix,
Then coordinate system is set up on picture element matrix, top left co-ordinate is origin (0,0), and x-axis represents that width from left to right increases, y-axis
Represent that height increases from top to bottom, then with a default minimum face frame (Wmin,Hmin), the initial coordinate of face frame is
(0,0), the slip since the image upper left corner from left to right from top to bottom, has been handled until whole image;Then again gradually
Increase the size of face frame and slided in whole sub-picture, until face frame increases to the maximum (W of default face framemax,
Hmax);The pixel composition piece image just taken out in face frame a position is slided into every time and carries out Face datection, if inspection
Measure face then record human face frame the upper left corner coordinate (x, y) and size (width, height), wherein, Wmin≤width
≤Wmax, Hmin≤height≤Hmax。
5. the side for obtaining clear human body image is followed the trail of according to a kind of any described intelligent air condition dual cameras of claim 1-4
Method, it is characterised in that:When common camera is in tracing mode, the image that pattern process module reads the common camera is gone forward side by side
Row Face datection, when detecting face, record human face frame the upper left corner coordinate (x, y) and size (width,
Height), judge whether face location is in the central area of the central area i.e. image of common camera of common camera
Position, if face location is in the central area of common camera, i.e. x<m/2<During x+width, then camera is not rotated, figure
The image that processing module continues to read common camera carries out Face datection;If face location is not in common camera
Heart district domain, i.e. m/2<X or m/2>During x+width, then angle [alpha]=((x+width/ that common camera needs to rotate is calculated
2)-m/2)/m*30 °,
If α=0, common camera need not be rotated, i.e., the angle that need to be rotated is zero;
If α>0, common camera needs the angle of turn to be α, that is, needs to turn to the right that common camera obtains image
It is dynamic;
If α<0, common camera needs counter-rotational angle to be α, that is, needs to obtain the left side of image to common camera
Rotate;
Common camera rotation direction of the controllor for step-by-step motor according to obtained by calculating and the tooth ratio between angle and gear, conversion
Calculate the rotation direction and rotational angle of stepper motor;Then rotation direction further according to stepper motor and rotational angle generation
Pulse control signal is sent to stepper motor driver, and stepper motor driver is according to received pulse control signal to step
Stepper motor is powered, and stepper motor is rotated due angle towards the direction calculated, drives common camera to turn by gear
The position that face in image is currently obtained towards common camera is moved, then output module reads common camera acquisition
Image is exported, and common camera proceeds face tracking.
6. a kind of intelligent air condition dual camera according to claim 1 follows the trail of the method for obtaining clear human body image, it is special
Levy and be:The visual angle of the wide-angle camera is 90 °, and the visual angle of common camera is 30 °.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710182512.8A CN107038418A (en) | 2017-03-24 | 2017-03-24 | A kind of intelligent air condition dual camera follows the trail of the method for obtaining clear human body image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710182512.8A CN107038418A (en) | 2017-03-24 | 2017-03-24 | A kind of intelligent air condition dual camera follows the trail of the method for obtaining clear human body image |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107038418A true CN107038418A (en) | 2017-08-11 |
Family
ID=59533393
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710182512.8A Pending CN107038418A (en) | 2017-03-24 | 2017-03-24 | A kind of intelligent air condition dual camera follows the trail of the method for obtaining clear human body image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107038418A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109190557A (en) * | 2018-08-31 | 2019-01-11 | 广州洪荒智能科技有限公司 | A kind of monitoring method based on recognition of face |
WO2019179441A1 (en) * | 2018-03-21 | 2019-09-26 | 北京猎户星空科技有限公司 | Focus tracking method and device of smart apparatus, smart apparatus, and storage medium |
CN110738078A (en) * | 2018-07-19 | 2020-01-31 | 青岛海信移动通信技术股份有限公司 | face recognition method and terminal equipment |
CN111263081A (en) * | 2020-02-28 | 2020-06-09 | 浙江大华技术股份有限公司 | Method and apparatus for determining exposure statistic, storage medium, and electronic apparatus |
CN111277737A (en) * | 2020-01-19 | 2020-06-12 | 浪潮金融信息技术有限公司 | Intelligent AI camera module for self-service terminal and control method |
CN113744418A (en) * | 2021-11-01 | 2021-12-03 | 环球数科集团有限公司 | Ticket business verification system based on edge cloud integrated environment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102708361A (en) * | 2012-05-11 | 2012-10-03 | 哈尔滨工业大学 | Human face collecting method at a distance |
CN202939756U (en) * | 2012-11-29 | 2013-05-15 | 长安大学 | Dual-video-head target invasion recording system |
CN104301669A (en) * | 2014-09-12 | 2015-01-21 | 重庆大学 | Suspicious target detection tracking and recognition method based on dual-camera cooperation |
CN106161941A (en) * | 2016-07-29 | 2016-11-23 | 深圳众思科技有限公司 | Dual camera chases after burnt method, device and terminal automatically |
-
2017
- 2017-03-24 CN CN201710182512.8A patent/CN107038418A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102708361A (en) * | 2012-05-11 | 2012-10-03 | 哈尔滨工业大学 | Human face collecting method at a distance |
CN202939756U (en) * | 2012-11-29 | 2013-05-15 | 长安大学 | Dual-video-head target invasion recording system |
CN104301669A (en) * | 2014-09-12 | 2015-01-21 | 重庆大学 | Suspicious target detection tracking and recognition method based on dual-camera cooperation |
CN106161941A (en) * | 2016-07-29 | 2016-11-23 | 深圳众思科技有限公司 | Dual camera chases after burnt method, device and terminal automatically |
Non-Patent Citations (1)
Title |
---|
谢剑斌 等: "《视觉感知与智能视频监控》", 31 March 2012, 国防科技大学出版社 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019179441A1 (en) * | 2018-03-21 | 2019-09-26 | 北京猎户星空科技有限公司 | Focus tracking method and device of smart apparatus, smart apparatus, and storage medium |
CN110738078A (en) * | 2018-07-19 | 2020-01-31 | 青岛海信移动通信技术股份有限公司 | face recognition method and terminal equipment |
CN109190557A (en) * | 2018-08-31 | 2019-01-11 | 广州洪荒智能科技有限公司 | A kind of monitoring method based on recognition of face |
CN111277737A (en) * | 2020-01-19 | 2020-06-12 | 浪潮金融信息技术有限公司 | Intelligent AI camera module for self-service terminal and control method |
CN111263081A (en) * | 2020-02-28 | 2020-06-09 | 浙江大华技术股份有限公司 | Method and apparatus for determining exposure statistic, storage medium, and electronic apparatus |
CN113744418A (en) * | 2021-11-01 | 2021-12-03 | 环球数科集团有限公司 | Ticket business verification system based on edge cloud integrated environment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107038418A (en) | A kind of intelligent air condition dual camera follows the trail of the method for obtaining clear human body image | |
CN107087107B (en) | Image processing apparatus and method based on dual camera | |
CN101441076B (en) | Method and device for detecting barrier | |
US9503658B2 (en) | Method for generating and evaluating an image | |
JP5448617B2 (en) | Distance estimation device, distance estimation method, program, integrated circuit, and camera | |
US8600170B2 (en) | Edge extraction apparatus, surveying equipment, and program | |
CN101169380B (en) | Bridge cable surface damage dynamic detection method and device | |
CN1584729A (en) | Image projection method and device | |
CN109739239B (en) | Planning method for uninterrupted instrument recognition of inspection robot | |
CN102331795B (en) | Method for controlling sunlight reflecting device to automatically track sun based on facula identification | |
US20110216194A1 (en) | Camera calibration apparatus | |
CN101848327B (en) | Image capturing device and image processing method | |
CN112348775B (en) | Vehicle-mounted looking-around-based pavement pit detection system and method | |
CN107687818A (en) | Three-dimensional measurement method and three-dimensional measurement device | |
JP3637226B2 (en) | Motion detection method, motion detection device, and recording medium | |
CN105865350A (en) | 3D object point cloud imaging method | |
CN107071347A (en) | The method of adjustment and headend equipment of a kind of wireless localization apparatus | |
CN115690190B (en) | Moving target detection and positioning method based on optical flow image and pinhole imaging | |
CN111031245A (en) | Controller and control method for adjusting industrial camera lens | |
CN113141491B (en) | Projection method, system and storage medium based on monochromatic light reference line emitter | |
CN112203069B (en) | Wide-screen projection method and system based on camera and readable storage medium | |
CN105025219A (en) | Image acquisition method | |
CN108550144B (en) | Laser light bar sequence image quality evaluation method based on gray scale reliability | |
US20160057446A1 (en) | Method of generating a framed video system | |
CN107147891B (en) | The adjustable three mesh depth of optical axis obtains video camera |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170811 |