CN109924994A - A kind of x photo-beat take the photograph during detection position automatic calibrating method and system - Google Patents
A kind of x photo-beat take the photograph during detection position automatic calibrating method and system Download PDFInfo
- Publication number
- CN109924994A CN109924994A CN201910263390.4A CN201910263390A CN109924994A CN 109924994 A CN109924994 A CN 109924994A CN 201910263390 A CN201910263390 A CN 201910263390A CN 109924994 A CN109924994 A CN 109924994A
- Authority
- CN
- China
- Prior art keywords
- artis
- coordinate
- human body
- flat panel
- panel detector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 43
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000012360 testing method Methods 0.000 claims abstract description 14
- 238000006073 displacement reaction Methods 0.000 claims description 29
- 230000000875 corresponding effect Effects 0.000 claims description 16
- 238000013136 deep learning model Methods 0.000 claims description 9
- MDJRZSNPHZEMJH-MTMZYOSNSA-N artisone acetate Chemical compound C1C=C2C[C@@H](O)CC[C@]2(C)[C@@H]2[C@@H]1[C@@H]1CC[C@H](C(=O)COC(=O)C)[C@@]1(C)CC2 MDJRZSNPHZEMJH-MTMZYOSNSA-N 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000009885 systemic effect Effects 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims 1
- 238000005259 measurement Methods 0.000 claims 1
- 230000005055 memory storage Effects 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 7
- 208000033748 Device issues Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 238000013329 compounding Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000008450 motivation Effects 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Landscapes
- Apparatus For Radiation Diagnosis (AREA)
- Image Processing (AREA)
Abstract
The invention belongs to medical device technical fields, a kind of detection position automatic calibrating method during being taken the photograph more particularly to x photo-beat, including signal source, memory, processor and executing agency, automatic calibrating method is the following steps are included: S1, signal source obtain the RGBD image of human body on check bit；S2, memory storage RGBD image information and several default calibration data；S3, processor handle RGBD image information；S4, processor issue control instruction to executing agency；A kind of x photo-beat provided by the invention take the photograph during detection position automatic calibrating method and system, the height of flat panel detector can be automatically adjusted, it can judge whether position of human body standard and treats testing staff by corresponding suggestion device and carry out that corresponding position is mobile to be instructed simultaneously, it is guided without staff, also it can guarantee that the posture of personnel to be detected meets testing requirements, to guarantee the correctness and validity of image.
Description
Technical field
The invention belongs to medical device technical fields, and in particular to a kind of x photo-beat take the photograph during the automatic school in detection position
Quasi- method and system.
Background technique
In original Image chain, X source, light concentrator center, flat panel detector center are located along the same line, position phase
To fixed and only one freedom degree (moving up and down), the pyramid type light beam that X source issues passes through light concentrator, forms specific shape
The light beam of size, the light beam pass through object to be shot and are located at the flat panel detector reception at object rear.
Since all there is particular size and shape by the light beam and flat panel detector of light concentrator, just for object to be shot
Some position be imaged, when the position to be shot of object to be shot is apparently higher than or when lower than preset position, will be led
Image is caused to lack the part position to be shot.
In addition, the information being usually finally presented on image is related with the angle that X-Ray passes through object to be shot, and due to
Lack guidance, object to be shot plane there may be apparent angle between flat panel detector plane, this will lead to X-Ray and passes through
Certain tilt angle, in this case, the certain portions of object to be shot are had compared to best angle when object to be shot
Point may by other parts cover block and the certain parts of object to be shot can by abnormal amplification (such as histoorgan),
Final image information is insufficient or information is wrong.In order to avoid above situation, operator usually requires repeated multiple times disengaging X and penetrates
Line room to adjust the position of patient, and lack the patient of medical expertise do not know usually how compounding practice personnel's ability
Faster, preferably reach optimum position so that imaging time is longer, efficiency is lower, not only wasted time, but also waste of manpower.
Summary of the invention
For it is above-mentioned in the prior art, the present invention provides the detection position automatic calibrating method during a kind of x photo-beat is taken the photograph
And system, the height of flat panel detector can be automatically adjusted, at the same can judge position of human body whether standard and by corresponding
Suggestion device treats testing staff and carries out the mobile guidance in corresponding position, guides without staff, also can guarantee people to be detected
The posture of member meets testing requirements, to guarantee the correctness and validity of image.
The technical scheme adopted by the invention is as follows:
A kind of x photo-beat take the photograph during detection position automatic calibrating method and system, including signal source, memory, processing
Device and executing agency, the detection position automatic calibrating method the following steps are included:
S1, signal source obtain the RGBD image of human body on check bit；
S2, the memory store the RGBD image information and several default calibration data that the signal source acquires, institute
Stating default calibration data includes default both shoulders artis three-dimensional coordinate and preset level position；
S3, the processor RGBD image information collected to the signal source are handled, by being closed based on human body
The deep learning Models computed of nodal test obtains human joint points three-dimensional coordinate, is calculated according to human joint points three-dimensional coordinate
Coordinate difference one, actually detected angle value and horizontal displacement；
S4, the executing agency include flat panel detector driving mechanism and suggestion device, and the processor is according to the seat
It marks difference one and issues the mobile control instruction one of driving flat panel detector to the flat panel detector driving mechanism, according to the reality
Border detection angles value and horizontal displacement issue the control instruction two of guidance human body adjustment corresponding actions to the suggestion device.
Further, in S3: the calculating of the coordinate difference one includes the following steps:
S3.1, the scene on check bit is shot by the signal source, acquisition is located at human body on check bit
RGBD image information is carried out three-dimensional processing by the processor, obtains RGB image and depth image by RGBD image information,
The depth image includes image information and depth of view information；
S3.2, the deep learning model based on human joint points detection calculate several human joint points in the depth image
In position, and determine the artis image coordinate of several human joint points in the depth image；
S3.3, according to the artis image coordinate, the depth of view information and pre-set signal source calibrating parameters meter
Artis three-dimensional coordinate corresponding with artis image coordinate is calculated, the artis three-dimensional coordinate exists for characterizing human joint points
D coordinates value in the scene；The artis three-dimensional coordinate includes human body both shoulders artis three-dimensional coordinate and several human bodies
Other artis three-dimensional coordinates；
S3.4, it the longitudinal coordinate value of the longitudinal coordinate value of human body both shoulders artis and default both shoulders artis is subtracted each other obtains
Coordinate difference one:
Y3=Y1-Y2
Wherein, the ordinate value of human body both shoulders artis is Y1, and the ordinate value for presetting both shoulders artis is Y2, the seat
Mark difference one is Y3, and the processor issues control to the flat panel detector driving mechanism according to one Y3 of coordinate difference and refers to
Enable one.
Further, in S3: the actually detected angle value is the angle of angle between the detection plane and X-ray line of human body
Value, the angle value are α, the angle calculation formula of α are as follows:
Tan α=︱ H2/H1 ︱
Wherein: human joint points include the artis one and artis two in human testing plane, the artis
One and the three-dimensional coordinate of the artis two be respectively A (x1, y1, z1), B (x2, y2, z2), z1 is the artis one and institute
The distance between signal source is stated, the distance between artis two and the signal source described in z2, the distance of the two, which is subtracted each other, obtains seat
Difference two is marked, the coordinate difference two is H1, i.e. H1=z1-z2；X1, x2 are respectively the artis one, the artis two
Abscissa, subtracted each other by the abscissa to the two and obtain coordinate difference three, the coordinate difference three be H2, i.e. H2=x1-x2；
The horizontal displacement, which subtracts each other for the real standard position of human body with the preset level position, obtains coordinate difference
Four, the coordinate difference four is H3, and the real standard position of the human body is the abscissa value Xn in an artis of human body, institute
Stating preset level position is the abscissa value Xm in the systemic presupposition artis, then H3=Xn-Xm；
Actually detected angle value α and horizontal displacement H3 is calculated, you can get it human body needs the angle and level adjusted
The amplitude of displacement, the processor are issued according to the actually detected angle value α and horizontal displacement H3 to the suggestion device
Control instruction two prompts human body to adjust posture by the suggestion device, to reach human body and normal place as far as possible
It is overlapped, that is, realizes human testing plane and the real standard position of human body and being overlapped for normal place, during the adjustment,
It prompts human body to carry out the adjustment of posture by the suggestion device, carries out pose adjustment for human body and make guide, the prompt dress
The prompting mode set includes one of acousto-optic hint or display reminding mode or a variety of combinations.
Further, the signal source includes binocular camera, laser radar or ultrasonic radar.
Further, the execution movement of the flat panel detector is to move up and down.
Further, the moving direction of the flat panel detector is specific as follows:
The longitudinal coordinate value of human body both shoulders artis is greater than the longitudinal coordinate value of default both shoulders artis, the plate detection
Device moves up, on the contrary then move down.
Further, the longitudinal coordinate value of the default both shoulders artis is opposite with the height and position of the flat panel detector
It answers.
Further, the flat panel detector driving mechanism includes flat panel detector and for driving the flat panel detector
The driving mechanism moved up and down, the driving mechanism include driving device, the control terminal of the driving device and the processor
Output end connection.
Detection position automated calibration system during being taken the photograph the present invention also provides a kind of x photo-beat, which is characterized in that use institute
Detection position automatic calibrating method is stated, this automated calibration system includes:
Signal source, the signal source are binocular camera, and the binocular camera is located at the front of check bit, for obtaining detection
The RGBD image of human body on position；
Memory, it is described for storing the RGBD image information and several default calibration data that the signal source acquires
Default calibration data includes default both shoulders artis three-dimensional coordinate and preset level position；
Processor obtains human body by resolving for handling signal source RGBD image information collected
Artis three-dimensional coordinate calculates coordinate difference one according to human joint points three-dimensional coordinate；It is three-dimensional according to the human joint points
Coordinate calculates actually detected angle value and horizontal displacement；
Executing agency, the executing agency include flat panel detector driving mechanism and suggestion device, the processor according to
The coordinate difference one issues the mobile control instruction one of driving flat panel detector to the flat panel detector driving mechanism, according to
The actually detected angle value and horizontal displacement refer to the control that the suggestion device issues guidance human body adjustment corresponding actions
Enable two.
The invention has the benefit that
1. the technical program obtains the RGBD image of human body on check bit by signal source, by being examined based on human joint points
The deep learning Models computed of survey obtains human joint points three-dimensional coordinate, and artis three-dimensional coordinate includes human body both shoulders artis three
Other artis three-dimensional coordinates for tieing up coordinate and several human bodies close the longitudinal coordinate value of human body both shoulders artis and default both shoulders
The longitudinal coordinate value of node, which is subtracted each other, obtains coordinate difference one, and processor is sent out according to coordinate difference one to flat panel detector driving mechanism
Control instruction one out, driving flat panel detector move in vertical direction, can automatically adjust the height of flat panel detector, guarantee
Filming image effect.
2. calculating actually detected angle value and horizontal displacement according to human joint points three-dimensional coordinate, you can get it, and human body is needed
The adjustment amplitude for the angle and horizontal displacement to be rotated, and prompt human body to adjust posture by suggestion device, it is personnel to be detected
The mobile guidance in corresponding position is carried out, to reach being overlapped for human body and normal place as far as possible, prompts effect more preferable, is not necessarily to
Staff guidance, also can guarantee that the posture of personnel to be detected meets testing requirements, thus guarantee image correctness and effectively
Property.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of hardware setting in the present invention；
Fig. 2 is the structural schematic diagram of shifting state under flat panel detector in the present invention；
Fig. 3 is the structural schematic diagram of flat panel detector raised state in the present invention；
Fig. 4 is that actually detected angle value calculates schematic diagram in the present invention；
Fig. 5 is that horizontal displacement calculates schematic diagram in the present invention；
Fig. 6 is automatic calibrating method flow diagram in the present invention；
Fig. 7 is the flow diagram of step S3 in the present invention.
In figure: binocular camera 1；Memory 2；Processor 3；Suggestion device 4；X source 5；Light concentrator 6；Artis 1；It closes
Node 28；X-ray line 9；Real standard position 10；Preset level position 11；Human body both shoulders artis 12；Default both shoulders artis
13；Flat panel detector 14.
Specific embodiment
With reference to the accompanying drawing and specific embodiment the present invention is further elaborated.
Embodiment 1:
As shown in Fig. 1, Fig. 6, a kind of detection position automatic calibrating method during being taken the photograph the present embodiment provides x photo-beat, packet
Include signal source, memory, processor and executing agency, the detection position automatic calibrating method the following steps are included:
S1, signal source obtain the RGBD image of human body on check bit；
S2, the memory store the RGBD image information and several default calibration data that the signal source acquires, institute
Stating default calibration data includes default both shoulders artis three-dimensional coordinate and preset level position；
S3, the processor RGBD image information collected to the signal source are handled, by being closed based on human body
The deep learning Models computed of nodal test obtains human joint points three-dimensional coordinate, is calculated according to human joint points three-dimensional coordinate
Coordinate difference one, actually detected angle value and horizontal displacement；
S4, the executing agency include flat panel detector driving mechanism and suggestion device, and the processor is according to the seat
It marks difference one and issues the mobile control instruction one of driving flat panel detector to the flat panel detector driving mechanism, according to the reality
Border detection angles value and horizontal displacement issue the control instruction two of guidance human body adjustment corresponding actions to the suggestion device.
Embodiment 2:
The present embodiment is to optimize restriction on the basis of above-described embodiment 1.
As shown in Fig. 2, Fig. 3, Fig. 7, in S3: the calculating of coordinate difference one includes the following steps:
S3.1, the scene on check bit is shot by binocular camera, acquisition is located at the RGBD of human body on check bit
RGBD image information is carried out three-dimensional processing by processor 3, obtains RGB image and depth image, depth image by image information
Including image information and depth of view information；
S3.2, the deep learning model based on human joint points detection calculate several human joint points in depth image
Position, and determine artis image coordinate of several human joint points in depth image；
S3.3, it is calculated and is closed according to artis image coordinate, depth of view information and pre-set binocular camera calibrating parameters
The corresponding artis three-dimensional coordinate of node image coordinate, artis three-dimensional coordinate are used to characterize three of human joint points in the scene
Dimensional coordinate values；
Specifically, 1 calibrating parameters of binocular camera include camera focus, picture centre coordinate and pattern distortion coefficient, according to
Artis calibration coordinate is calculated in pattern distortion coefficient and artis image coordinate；According to the depth of view information in depth image,
Determine that artis depth value corresponding with artis image coordinate, artis depth value are that the Z axis in artis three-dimensional coordinate is sat
Scale value；It brings artis depth value, artis calibration coordinate and picture centre coordinate into three-dimensional coordinate computation model, is calculated
Artis three-dimensional coordinate；
The deep learning model and three-dimensional coordinate computation model of human joint points detection are existing computation model, are brought into pair
The calculating to human joint points three-dimensional coordinate can be realized in the parameter answered.
Artis three-dimensional coordinate includes other artis three-dimensional of 12 three-dimensional coordinate of human body both shoulders artis and several human bodies
Coordinate；
S3.4, keep relative position constant due to binocular camera and X source 5, light concentrator 6, flat panel detector 14, therefore plate
Detector 14 in place when, the longitudinal coordinate of the both shoulders of human body is all relatively-stationary in RGB image, by human body both shoulders
The longitudinal coordinate value of the longitudinal coordinate value of artis 12 and default both shoulders artis 13, which is subtracted each other, obtains coordinate difference one, coordinate difference
One becomes as the basis of flat panel detector 14 moving direction and amplitude, it is specific calculate it is as follows:
Y3=Y1-Y2
Wherein, the ordinate value of human body both shoulders artis 12 is Y1, and the ordinate value for presetting both shoulders artis 13 is Y2, is sat
Mark difference one is Y3, and processor 3 issues control instruction one to flat panel detector driving mechanism according to one Y3 of coordinate difference.
Specifically, the longitudinal coordinate value of default both shoulders artis 13 is corresponding with the height and position of flat panel detector 14, from
And by judging the difference of the longitudinal coordinate value of default both shoulders artis 13 and the longitudinal coordinate value in human body both shoulders joint, it can obtain
Flat panel detector 14 needs the respective distance for moving up or moving down out.
The moving direction of flat panel detector 14 is specific as follows:
The longitudinal coordinate value of human body both shoulders artis 12 is greater than the longitudinal coordinate value of default both shoulders artis 13, plate detection
Device 14 moves up, on the contrary then move down.
Embodiment 3:
The present embodiment is to optimize restriction on the basis of above-described embodiment 1 or 2.
As shown in Fig. 4, Fig. 5, Fig. 7, in S3: actually detected angle value is angle between the detection plane and X-ray line 9 of human body
Angle value, the angle calculation formula of angle value α, α are as follows:
Tan α=︱ H2/H1 ︱
Wherein: human joint points include the artis 1 and artis 28 in human testing plane, artis 1
Three-dimensional coordinate with artis 28 is respectively A (x1, y1, z1), B (x2, y2, z2), wherein y1=y2, z1 are artis 1
The distance between binocular camera, the distance between z2 artis 28 and binocular camera, the distance of the two, which is subtracted each other, obtains coordinate difference
Value two, coordinate difference two are H1, i.e. H1=z1-z2；X1, x2 are respectively the abscissa of artis 1, artis 28, by right
The abscissa of the two, which subtracts each other, obtains coordinate difference three, and coordinate difference three is H2, i.e. H2=x1-x2；
The angle of angle, that is, human body detection planar offset X-ray line 9 of α, if deviation angle is excessive, the detection plane of human body
The direction of illumination of non-face X-ray line 9, therefore, the effect of shooting will appear deformation, shooting effect be influenced, it then becomes necessary to guarantee
The angle of α is as far as possible close to 90 °, if deviation angle is more than preset value, needs to adjust human body by the prompt of suggestion device 4
Angle position in entire position adjustment, calculates angle in real time and enables people to be detected real-time by suggestion device 4
The effect of oneself adjustment is recognized, until angle adjustment meets the requirements.
Horizontal displacement is the deviation between the real standard position 10 and preset level position 11 of human body, when human body
When real standard position 10 and larger 11 deviation of preset level position, human body is needed to be adjusted, reach as far as possible with preset level
The requirement that position 11 is overlapped, therefore, horizontal displacement are that the real standard position 10 of human body and preset level position 11 are subtracted each other
Coordinate difference four out, coordinate difference four are H3, and the real standard position 10 of human body is the abscissa value in an artis of human body
Xn, preset level position 11 are the abscissa value Xm in the systemic presupposition artis, then H3=Xn-Xm；That is the positive and negative values of H3 are determined
Determined the direction that moves horizontally of human body, the absolute value of H3 is the distance that moves horizontally of human body, human body move horizontally direction and
It moves horizontally distance to guide by suggestion device 4, during entire horizontal position adjustment, calculates horizontal displacement in real time
Amount, while the effect for enabling people to be detected to recognize that oneself is adjusted in real time by suggestion device 4, until horizontal position adjustment
It meets the requirements.
To sum up, calculate actually detected angle value α and horizontal displacement H3, angle that you can get it human body needs to adjust and
The amplitude of horizontal displacement, processor 3 issue control to suggestion device 4 according to actually detected angle value α and horizontal displacement H3 and refer to
Two are enabled, prompts human body to adjust posture by suggestion device 4, to reach being overlapped for human body and normal place as far as possible, i.e., in fact
Show human testing plane and the real standard position 10 of human body and passes through prompt during the adjustment with being overlapped for normal place
Device 4 prompts human body to carry out the adjustment of posture, carries out pose adjustment for human body and makes guide, in order to meet different use need
It asks, the prompting mode of suggestion device 4 includes one of acousto-optic hint or display reminding mode or a variety of combinations.
In the present embodiment, suggestion device 4 is voice prompt device, and voice prompt device receives control instruction two, to be detected
People station can adjust the position of oneself by voice prompting later on check bit, and voice prompting automation control is not necessarily to work
Personnel's operation, is just able to achieve in the case where no staff, personnel to be detected can also carry out correct appearance by voice prompting
Gesture adjustment.
Suggestion device can also be display screen, and display screen carries out pose adjustment by prompting image to treat testing staff accordingly
Guidance.
Embodiment 4:
The present embodiment is to optimize restriction on the basis of above-described embodiment 3.
Flat panel detector driving mechanism includes flat panel detector 14 and the drive for driving flat panel detector 14 to move up and down
Motivation structure, driving mechanism include driving device, and the control terminal of driving device is connect with the output end of processor 3.
Specifically, driving device is driving motor, realized on driving motor by the modes such as gear drive or chain rotation
To the descending operation of flat panel detector 14, the drive mechanism of 14 lifting action of flat panel detector is all made of existing drive mechanism,
It is not limited to motor drive.
Embodiment 5:
As shown in Figure 1, the detection position automated calibration system during being taken the photograph the present invention also provides a kind of x photo-beat, feature
It is, using the detection position automatic calibrating method any in embodiment 1- embodiment 4, this automated calibration system includes:
Binocular camera 1, binocular camera 1 are located at the front of check bit, for obtaining the RGBD image of human body on check bit；
Memory 2 presets school for storing the RGBD image information and several default calibration data that binocular camera acquires
Quasi- data include default 13 three-dimensional coordinate of both shoulders artis and preset level position 11；
Processor 3 obtains human body pass by resolving for handling binocular camera RGBD image information collected
Nodes Three-dimensional coordinate calculates coordinate difference one according to human joint points three-dimensional coordinate；According to human joint points three-dimensional coordinate meter
Calculate actually detected angle value and horizontal displacement；
Executing agency, executing agency include flat panel detector driving mechanism and suggestion device 4, and processor 3 is according to coordinate difference
Value one issues the mobile control instruction one of driving flat panel detector 14 to flat panel detector driving mechanism, according to actually detected angle
Value and horizontal displacement issue the control instruction two of guidance human body adjustment corresponding actions to suggestion device 4.
The technical program obtains the RGBD image of human body on check bit by binocular camera 1, by being based on human joint points
The deep learning Models computed of detection obtains human joint points three-dimensional coordinate, and artis three-dimensional coordinate includes human body both shoulders artis
Other artis three-dimensional coordinates of 12 three-dimensional coordinates and several human bodies by the longitudinal coordinate value of human body both shoulders artis 12 and are preset
The longitudinal coordinate value of both shoulders artis 13, which is subtracted each other, obtains coordinate difference one, and processor 3 is according to coordinate difference one to flat panel detector
Driving mechanism issues control instruction one, and driving flat panel detector 14 moves in vertical direction, can automatically adjust plate detection
The height of device guarantees filming image effect.
Actually detected angle value and horizontal displacement are calculated according to human joint points three-dimensional coordinate, you can get it, and human body needs
The angle of rotation and the adjustment amplitude of horizontal displacement, and by suggestion device 4 prompt human body adjust posture, be personnel to be detected into
The mobile guidance in the corresponding position of row guides to reach being overlapped for human body and normal place as far as possible without staff,
It can guarantee that the posture of personnel to be detected meets testing requirements, to guarantee the correctness and validity of image.
The present invention is not limited to above-mentioned optional embodiment, anyone can show that other are various under the inspiration of the present invention
The product of form, however, make any variation in its shape or structure, it is all to fall into the claims in the present invention confining spectrum
Technical solution, be within the scope of the present invention.
Claims (10)
1. a kind of x photo-beat take the photograph during detection position automatic calibrating method, which is characterized in that including signal source, memory, place
Manage device and executing agency, the detection position automatic calibrating method the following steps are included:
S1, the signal source obtain the RGBD image of human body on check bit；
S2, the memory store the RGBD image information and several default calibration data that the signal source acquires, described pre-
If calibration data includes default both shoulders artis three-dimensional coordinate and preset level position；
S3, the processor RGBD image information collected to the signal source are handled, by being based on human joint points
The deep learning Models computed of detection obtains human joint points three-dimensional coordinate, calculates coordinate according to human joint points three-dimensional coordinate
Difference one, actually detected angle value and horizontal displacement；
S4, the executing agency include flat panel detector driving mechanism and suggestion device, and the processor is according to the coordinate difference
Value one issues the mobile control instruction one of driving flat panel detector to the flat panel detector driving mechanism, according to the practical inspection
Angle measurement angle value and horizontal displacement issue the control instruction two of guidance human body adjustment corresponding actions to the suggestion device.
The detection position automatic calibrating method during 2. a kind of x photo-beat is taken the photograph according to claim 1, which is characterized in that S3
In:
The calculating of the coordinate difference one includes the following steps:
S3.1, the scene on check bit is shot by the signal source, acquisition is located at the RGBD figure of human body on check bit
As information, RGBD image information is carried out by three-dimensional processing by the processor, obtains RGB image and depth image, the depth
Spending image includes image information and depth of view information；
S3.2, the deep learning model based on human joint points detection calculate several human joint points in the depth image
Position, and determine artis image coordinate of several human joint points in the depth image；
S3.3, according to the artis image coordinate, the depth of view information and pre-set signal source calibrating parameters calculate with
The corresponding artis three-dimensional coordinate of artis image coordinate, the artis three-dimensional coordinate is for characterizing human joint points described
D coordinates value in scene；The artis three-dimensional coordinate includes its of human body both shoulders artis three-dimensional coordinate and several human bodies
His artis three-dimensional coordinate；
S3.4, it the longitudinal coordinate value of the longitudinal coordinate value of human body both shoulders artis and default both shoulders artis is subtracted each other obtains coordinate
Difference one:
Y3=Y1-Y2
Wherein, the ordinate value of human body both shoulders artis is Y1, and the ordinate value for presetting both shoulders artis is Y2, the coordinate difference
Value one is Y3, and the processor issues control instruction to the flat panel detector driving mechanism according to one Y3 of coordinate difference
One.
3. a kind of x photo-beat according to claim 1 or 2 take the photograph during detection position automatic calibrating method, feature exists
In,
In S3: the actually detected angle value is the angle value of angle between the detection plane and X-ray line of human body, the angle value
For α, the angle calculation formula of α are as follows:
Tan α=︱ H2/H1 ︱
Wherein: human joint points include the artis one and artis two in human testing plane, one He of artis
The three-dimensional coordinate of the artis two is respectively A (x1, y1, z1), B (x2, y2, z2), and z1 is the artis one and the letter
The distance between number source, the distance between artis two and the signal source described in z2, the distance of the two, which is subtracted each other, obtains coordinate difference
Value two, the coordinate difference two are H1, i.e. H1=z1-z2；X1, x2 are respectively the cross of the artis one, the artis two
Coordinate is subtracted each other by the abscissa to the two and obtains coordinate difference three, and the coordinate difference three is H2, i.e. H2=x1-x2；
The horizontal displacement, which subtracts each other for the real standard position of human body with the preset level position, obtains coordinate difference four, institute
Stating coordinate difference four is H3, and the real standard position of the human body is the abscissa value Xn in an artis of human body, described pre-
If horizontal position is the abscissa value Xm in the systemic presupposition artis, then H3=Xn-Xm；
The processor issues control instruction to the suggestion device according to the actually detected angle value α and horizontal displacement H3
Two, prompt human body to adjust posture by the suggestion device；The prompting mode of the suggestion device includes acousto-optic hint or display
One of prompting mode or a variety of combinations.
4. a kind of x photo-beat according to claim 3 take the photograph during detection position automatic calibrating method, it is characterised in that:
The signal source includes binocular camera, laser radar or ultrasonic radar.
5. a kind of x photo-beat according to claim 4 take the photograph during detection position automatic calibrating method, it is characterised in that:
The execution movement of the flat panel detector is to move up and down.
6. a kind of x photo-beat according to claim 5 take the photograph during detection position automatic calibrating method, it is characterised in that:
The moving direction of the flat panel detector is specific as follows:
The longitudinal coordinate value of human body both shoulders artis is greater than the longitudinal coordinate value of default both shoulders artis, on the flat panel detector
It moves, it is on the contrary then move down.
7. a kind of x photo-beat according to claim 6 take the photograph during detection position automatic calibrating method, it is characterised in that:
The longitudinal coordinate value of the default both shoulders artis is corresponding with the height and position of the flat panel detector.
8. a kind of x photo-beat according to claim 7 take the photograph during detection position automatic calibrating method, it is characterised in that:
The flat panel detector driving mechanism includes flat panel detector and the driving machine for driving the flat panel detector to move up and down
Structure, the driving mechanism include driving device, and the control terminal of the driving device is connect with the output end of the processor.
9. a kind of x photo-beat take the photograph during detection position automated calibration system, which is characterized in that using according in claim 8
The detection position automatic calibrating method.
10. a kind of x photo-beat according to claim 9 take the photograph during detection position automated calibration system, which is characterized in that
Including
Signal source, the signal source are binocular camera, and the binocular camera is located at the front of check bit, for obtaining check bit
The RGBD image of human body；
Memory, it is described default for storing the RGBD image information and several default calibration data that the signal source acquires
Calibration data includes default both shoulders artis three-dimensional coordinate and preset level position；
Processor obtains human synovial by resolving for handling signal source RGBD image information collected
Point three-dimensional coordinate, calculates coordinate difference one according to human joint points three-dimensional coordinate；According to the human joint points three-dimensional coordinate
Calculate actually detected angle value and horizontal displacement；
Executing agency, the executing agency include flat panel detector driving mechanism and suggestion device, and the processor is according to
Coordinate difference one issues the mobile control instruction one of driving flat panel detector to the flat panel detector driving mechanism, according to described
Actually detected angle value and horizontal displacement issue the control instruction two of guidance human body adjustment corresponding actions to the suggestion device.
Priority Applications (5)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910263390.4A CN109924994B (en) | 2019-04-02 | 2019-04-02 | Method and system for automatically calibrating detection position in x-ray shooting process |
| CA3135998A CA3135998A1 (en) | 2019-04-02 | 2020-04-02 | An integrated x-ray precision imaging device |
| PCT/CA2020/050438 WO2020198870A1 (en) | 2019-04-02 | 2020-04-02 | An integrated x-ray precision imaging device |
| AU2020255687A AU2020255687A1 (en) | 2019-04-02 | 2020-04-02 | An integrated X-ray precision imaging device |
| US17/268,949 US20220079544A1 (en) | 2019-04-02 | 2020-04-02 | An integrated x-ray precision imaging device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910263390.4A CN109924994B (en) | 2019-04-02 | 2019-04-02 | Method and system for automatically calibrating detection position in x-ray shooting process |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109924994A true CN109924994A (en) | 2019-06-25 |
| CN109924994B CN109924994B (en) | 2023-03-14 |
Family
ID=66989120
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910263390.4A Active CN109924994B (en) | 2019-04-02 | 2019-04-02 | Method and system for automatically calibrating detection position in x-ray shooting process |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109924994B (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110301925A (en) * | 2019-08-14 | 2019-10-08 | 晓智科技（成都）有限公司 | It is a kind of to facilitate the X-ray imaging system for adjusting each component physical alignment |
| CN110338831A (en) * | 2019-08-14 | 2019-10-18 | 晓智科技（成都）有限公司 | A method of each component physical alignment in adjustment X-ray imaging system |
| CN110897651A (en) * | 2019-12-13 | 2020-03-24 | 四川大学华西医院 | Intelligent physical examination X-ray chest radiography body position tutoring method and system |
| CN111906043A (en) * | 2020-07-16 | 2020-11-10 | 深圳市格灵人工智能与机器人研究院有限公司 | Pose detection method and system |
| CN112456334A (en) * | 2020-12-02 | 2021-03-09 | 太原重工股份有限公司 | Multi-winch lifting sling leveling method and system |
| CN113069132A (en) * | 2021-03-25 | 2021-07-06 | 上海奕瑞光电子科技股份有限公司 | Automatic positioning system and method in radiographic process |
| CN113180709A (en) * | 2021-07-01 | 2021-07-30 | 晓智未来(成都)科技有限公司 | Human body to-be-detected part posture recognition method based on photogrammetry |
| CN113223517A (en) * | 2021-04-13 | 2021-08-06 | 广州爱尔眼科医院有限公司 | Device position adjusting method and device based on doctor voice instruction |
| WO2022110132A1 (en) * | 2020-11-30 | 2022-06-02 | 江苏康众数字医疗科技股份有限公司 | Unmanned smart radiographing system and radiographing method |
| CN115171150A (en) * | 2022-06-15 | 2022-10-11 | 深圳市华屹医疗科技有限公司 | Human body imaging device interaction method, human body imaging device interaction device and computer-readable storage medium |
| CN115174803A (en) * | 2022-06-20 | 2022-10-11 | 平安银行股份有限公司 | Automatic photographing method and related equipment |
| CN116473576A (en) * | 2023-04-14 | 2023-07-25 | 深圳市宝润科技有限公司 | A mobile X-ray imaging detection method and system based on three-dimensional space positioning |
| US20240252054A1 (en) * | 2021-06-18 | 2024-08-01 | Neko Health Ab | Pulse wave velocity |
| CN119183006A (en) * | 2024-08-21 | 2024-12-24 | 河南翔宇医疗设备股份有限公司 | Camera adjusting method and system based on weight-loss gait rehabilitation platform |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002127052A (en) * | 2000-10-25 | 2002-05-08 | Matsushita Electric Ind Co Ltd | Robot control position correction method and robot control position correction system |
| EP1313065A1 (en) * | 2001-11-16 | 2003-05-21 | Philips Intellectual Property & Standards GmbH | Method and apparatus for calibration of and imaging by means of an X ray imaging apparatus sensitive to the field of gravity |
| US20050203386A1 (en) * | 2004-03-11 | 2005-09-15 | Siemens Aktiengesellschaft | Method of calibrating an X-ray imaging device |
| JP2006098065A (en) * | 2004-09-28 | 2006-04-13 | Sanyo Electric Co Ltd | Calibration device and method, and three-dimensional modelling device and system capable of using the same |
| JP2010193965A (en) * | 2009-02-23 | 2010-09-09 | Shimadzu Corp | Method for acquiring positional information for correction, method for correcting positional deviation, image processor, radiation imaging apparatus, and measurement phantom |
| CN103635139A (en) * | 2011-07-06 | 2014-03-12 | 富士胶片株式会社 | X-ray imaging device and calibration method therefor |
| CN103784160A (en) * | 2014-03-06 | 2014-05-14 | 北京锐视康科技发展有限公司 | Correction device and correction method for geometric position of cone beam CT system |
| CN104414662A (en) * | 2013-09-04 | 2015-03-18 | 江苏瑞尔医疗科技有限公司 | Position calibration and error compensation device of imaging equipment and compensation method of position calibration and error compensation device |
| CN106060514A (en) * | 2016-07-06 | 2016-10-26 | 中国计量大学 | Device and method for controlling pose of television in multi-vision mode |
| JP2017007026A (en) * | 2015-06-22 | 2017-01-12 | 日産自動車株式会社 | Position correcting system |
| CN106859675A (en) * | 2015-10-13 | 2017-06-20 | 西门子保健有限责任公司 | The method and system of the scanner automation for X-ray tube carried out with 3D cameras |
| JP2018121007A (en) * | 2017-01-27 | 2018-08-02 | 株式会社Ｓｃｒｅｅｎホールディングス | Substrate transport device, detection position calibration method and substrate processing apparatus |
| CN108629946A (en) * | 2018-06-14 | 2018-10-09 | 清华大学深圳研究生院 | A kind of tumble detection method for human body based on RGBD sensors |
| CN109276266A (en) * | 2018-09-19 | 2019-01-29 | 晓智科技（成都）有限公司 | It is a kind of that X-ray imaging dosing method is carried out according to thickness value |
-
2019
- 2019-04-02 CN CN201910263390.4A patent/CN109924994B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002127052A (en) * | 2000-10-25 | 2002-05-08 | Matsushita Electric Ind Co Ltd | Robot control position correction method and robot control position correction system |
| EP1313065A1 (en) * | 2001-11-16 | 2003-05-21 | Philips Intellectual Property & Standards GmbH | Method and apparatus for calibration of and imaging by means of an X ray imaging apparatus sensitive to the field of gravity |
| US20050203386A1 (en) * | 2004-03-11 | 2005-09-15 | Siemens Aktiengesellschaft | Method of calibrating an X-ray imaging device |
| JP2006098065A (en) * | 2004-09-28 | 2006-04-13 | Sanyo Electric Co Ltd | Calibration device and method, and three-dimensional modelling device and system capable of using the same |
| JP2010193965A (en) * | 2009-02-23 | 2010-09-09 | Shimadzu Corp | Method for acquiring positional information for correction, method for correcting positional deviation, image processor, radiation imaging apparatus, and measurement phantom |
| CN103635139A (en) * | 2011-07-06 | 2014-03-12 | 富士胶片株式会社 | X-ray imaging device and calibration method therefor |
| CN104414662A (en) * | 2013-09-04 | 2015-03-18 | 江苏瑞尔医疗科技有限公司 | Position calibration and error compensation device of imaging equipment and compensation method of position calibration and error compensation device |
| CN103784160A (en) * | 2014-03-06 | 2014-05-14 | 北京锐视康科技发展有限公司 | Correction device and correction method for geometric position of cone beam CT system |
| JP2017007026A (en) * | 2015-06-22 | 2017-01-12 | 日産自動車株式会社 | Position correcting system |
| CN106859675A (en) * | 2015-10-13 | 2017-06-20 | 西门子保健有限责任公司 | The method and system of the scanner automation for X-ray tube carried out with 3D cameras |
| CN106060514A (en) * | 2016-07-06 | 2016-10-26 | 中国计量大学 | Device and method for controlling pose of television in multi-vision mode |
| JP2018121007A (en) * | 2017-01-27 | 2018-08-02 | 株式会社Ｓｃｒｅｅｎホールディングス | Substrate transport device, detection position calibration method and substrate processing apparatus |
| CN108629946A (en) * | 2018-06-14 | 2018-10-09 | 清华大学深圳研究生院 | A kind of tumble detection method for human body based on RGBD sensors |
| CN109276266A (en) * | 2018-09-19 | 2019-01-29 | 晓智科技（成都）有限公司 | It is a kind of that X-ray imaging dosing method is carried out according to thickness value |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110338831B (en) * | 2019-08-14 | 2022-12-02 | 晓智未来(成都)科技有限公司 | Method for adjusting physical alignment of each component in X-ray imaging system |
| CN110338831A (en) * | 2019-08-14 | 2019-10-18 | 晓智科技（成都）有限公司 | A method of each component physical alignment in adjustment X-ray imaging system |
| CN110301925B (en) * | 2019-08-14 | 2023-07-04 | 晓智未来(成都)科技有限公司 | An X-ray Imaging System that Conveniently Adjusts the Physical Alignment of Components |
| CN110301925A (en) * | 2019-08-14 | 2019-10-08 | 晓智科技（成都）有限公司 | It is a kind of to facilitate the X-ray imaging system for adjusting each component physical alignment |
| CN110897651A (en) * | 2019-12-13 | 2020-03-24 | 四川大学华西医院 | Intelligent physical examination X-ray chest radiography body position tutoring method and system |
| CN110897651B (en) * | 2019-12-13 | 2020-11-24 | 四川大学华西医院 | Intelligent physical examination X-ray chest photography position counseling method and system |
| CN111906043A (en) * | 2020-07-16 | 2020-11-10 | 深圳市格灵人工智能与机器人研究院有限公司 | Pose detection method and system |
| CN111906043B (en) * | 2020-07-16 | 2022-03-29 | 深圳市格灵人工智能与机器人研究院有限公司 | Pose detection method and system |
| WO2022110132A1 (en) * | 2020-11-30 | 2022-06-02 | 江苏康众数字医疗科技股份有限公司 | Unmanned smart radiographing system and radiographing method |
| CN112456334A (en) * | 2020-12-02 | 2021-03-09 | 太原重工股份有限公司 | Multi-winch lifting sling leveling method and system |
| CN113069132A (en) * | 2021-03-25 | 2021-07-06 | 上海奕瑞光电子科技股份有限公司 | Automatic positioning system and method in radiographic process |
| CN113223517A (en) * | 2021-04-13 | 2021-08-06 | 广州爱尔眼科医院有限公司 | Device position adjusting method and device based on doctor voice instruction |
| CN113223517B (en) * | 2021-04-13 | 2023-08-08 | 广州爱尔眼科医院有限公司 | A device position adjustment method and device based on doctor's voice command |
| US20240252054A1 (en) * | 2021-06-18 | 2024-08-01 | Neko Health Ab | Pulse wave velocity |
| CN113180709A (en) * | 2021-07-01 | 2021-07-30 | 晓智未来(成都)科技有限公司 | Human body to-be-detected part posture recognition method based on photogrammetry |
| CN115171150A (en) * | 2022-06-15 | 2022-10-11 | 深圳市华屹医疗科技有限公司 | Human body imaging device interaction method, human body imaging device interaction device and computer-readable storage medium |
| CN115174803A (en) * | 2022-06-20 | 2022-10-11 | 平安银行股份有限公司 | Automatic photographing method and related equipment |
| CN116473576A (en) * | 2023-04-14 | 2023-07-25 | 深圳市宝润科技有限公司 | A mobile X-ray imaging detection method and system based on three-dimensional space positioning |
| CN119183006A (en) * | 2024-08-21 | 2024-12-24 | 河南翔宇医疗设备股份有限公司 | Camera adjusting method and system based on weight-loss gait rehabilitation platform |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109924994B (en) | 2023-03-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109924994A (en) | A kind of x photo-beat take the photograph during detection position automatic calibrating method and system | |
| CN109452947B (en) | Methods, X-ray imaging systems for generating positioning images and imaging patients | |
| CN102961154B (en) | Regulate the method for exposure field of x-ray system and device and x-ray system | |
| JP5922572B2 (en) | Practical 3D vision system apparatus and method | |
| CN111161358B (en) | Camera calibration method and device for structured light depth measurement | |
| EP3406196A1 (en) | X-ray system and method for standing subject | |
| RU2478980C2 (en) | System and method for automatic calibration of tracked ultrasound | |
| CN105388478B (en) | For detect acoustics and optical information method and apparatus and corresponding computer readable storage medium | |
| CN110443879B (en) | Perspective error compensation method based on neural network | |
| CN108717715A (en) | A kind of line-structured light vision system automatic calibration method for arc welding robot | |
| JP2016526673A (en) | System and method for tracking the location of a movable target object | |
| CN111076712A (en) | Automatic building method, system and device of space three-dimensional model | |
| CN108108021A (en) | The outer parameter correction gauge of tracing of human eye system and bearing calibration | |
| CN113533498A (en) | Weld detection and positioning method and positioning device of automatic eddy current flaw detection system | |
| CN110953993A (en) | A detection device and method for transmission line sag and distance limit | |
| KR102004642B1 (en) | Control Method for Radiography Training System | |
| CN109363701A (en) | A kind of mthods, systems and devices of adjustment scanning gantry interference | |
| JP5489037B2 (en) | Radiation beam irradiation target positioning apparatus and positioning method therefor | |
| CN112568922A (en) | Method for automatically arranging an X-ray source, X-ray system, program product and medium | |
| CN113781581B (en) | Depth of field distortion model calibration method based on target loose attitude constraint | |
| CN109893160A (en) | A kind of x light measurand pose adjustment householder method | |
| CN114485390B (en) | A device and method for measuring the size of underwater defects in hydraulic structures | |
| CN111210479A (en) | Laser auxiliary calibration device and method for measuring sizes of parts with different heights | |
| JP2004309318A (en) | Position detection method, device and program thereof, and calibration information generation method | |
| CN104739508B (en) | Moving position setting method and system of moving component and medical device |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 610000 north of Chengdu modern industrial port, PI Du District, Chengdu, Sichuan, No. 269 North Road, Hong Kong. Applicant after: Xiaozhi future (Chengdu) Technology Co.,Ltd. Address before: 610000 north of Chengdu modern industrial port, PI Du District, Chengdu, Sichuan, No. 269 North Road, Hong Kong. Applicant before: XIAOZHI TECHNOLOGY (CHENGDU) Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |