CN110319899A - Volume measuring method, device and system - Google Patents
Volume measuring method, device and system Download PDFInfo
- Publication number
- CN110319899A CN110319899A CN201910741396.8A CN201910741396A CN110319899A CN 110319899 A CN110319899 A CN 110319899A CN 201910741396 A CN201910741396 A CN 201910741396A CN 110319899 A CN110319899 A CN 110319899A
- Authority
- CN
- China
- Prior art keywords
- depth
- under test
- value
- object under
- scene
- 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
- G01—MEASURING; TESTING
- G01F—MEASURING VOLUME, VOLUME FLOW, MASS FLOW OR LIQUID LEVEL; METERING BY VOLUME
- G01F17/00—Methods or apparatus for determining the capacity of containers or cavities, or the volume of solid bodies
Abstract
The invention discloses a kind of volume measuring methods, device and system, and wherein volume measuring method is the following steps are included: receive the depth map of the scene containing object under test, the depth map of scene is on the basis of preset first level face;Reference distance value is received from single-point range unit, target detection point is located at the upper surface of object under test;Target depth corresponding with target detection point is obtained according to the depth map of scene;Compensation factor is obtained, is compensated according to depth value of the compensation factor to the depth map of scene, to obtain compensation depth value;The point cloud coordinate compensated according to compensation depth value, and the volume of object under test is obtained according to the point cloud coordinate of compensation.The numerical value that the present invention passes through the accurate distance measured compensates the depth value of depth map, can be with error existing for the depth value of Corrected Depth figure, to improve the precision of the volume of obtained object under test.
Description
Technical field
The invention belongs to object volume field of measuring technique more particularly to a kind of volume measuring methods, device and system.
Background technique
Currently, being that representative object volume measures in factory, warehouse etc. with logistics goods with the popularity of the internet
Scene it is more and more.It quickly object is sorted, is assembled to realize, and precisely dispatching, it is necessary to object body
Product is measured and is recorded.
Currently, the method for having had already appeared the cubing based on depth camera.Referring to Fig.1, depth camera is from upper direction
Lower shooting object under test obtains the depth map of the scene containing object under test.Depth map, that is, depth Z axis coordinate set, depth
Figure is also referred to as range image, refers to figure of the distance (depth) by each point in from image acquisition device to scene as pixel value
Picture, it directly reflects the geometry of each object visible surface in scene.The depth map of scene includes corresponding to each pixel
Depth information.By taking A point in Fig. 1 as an example, corresponding depth is d.It is available according to the corresponding depth information of each pixel
The point cloud coordinate of scene.Then, the corresponding coordinate point set of object under test is filtered out from the point cloud coordinate of scene.Finally, root
The volume of object under test can be calculated according to the corresponding coordinate point set of object under test.
However, the precision that existing depth camera obtains depth information is not high enough, especially by environment (such as the temperature that works
Degree etc.) influence when, error is larger.For example, the corresponding depth of A point that depth camera obtains is d by taking A point in Fig. 1 as an example.But
The numerical value is usually present certain error.The point cloud coordinate that the error is introduced into subsequent scene calculates and object under test
The calculating of volume, the bulking value error that will lead to are larger.
Summary of the invention
The embodiment of the present invention technical problems to be solved are to overcome volume in the prior art based on depth camera
The not high enough defect of the method precision of measurement provides a kind of volume measuring method, device and system.
The embodiment of the present invention is to solve above-mentioned technical problem by following technical proposals:
A kind of volume measuring method of the embodiment of the present invention, comprising the following steps:
The depth map of the scene containing object under test is received, the depth map of scene is on the basis of preset first level face;
Receive reference distance value from single-point range unit, reference distance value be target detection point and preset first level face it
Between distance, target detection point is located at the upper surface of object under test;
Target depth corresponding with target detection point is obtained according to the depth map of scene;
Compensation factor is obtained, is compensated according to depth value of the compensation factor to the depth map of scene, it is deep to obtain compensation
Angle value;Compensation factor is the ratio of reference distance value and target depth;
The point cloud coordinate compensated according to compensation depth value, and the body of object under test is obtained according to the point cloud coordinate of compensation
Product.
Optionally, single-point range unit includes laser range finder, and laser range finder is set on preset first level face, is swashed
The detection light beam of optar is projeced into the upper surface of object under test vertically.
Optionally, it is obtained according to the depth map of scene with the step of target detection point corresponding target depth and includes:
The corresponding depth value of target detection point is extracted from the depth map of scene according to the location information of laser range finder to make
For target depth.
Optionally, it is obtained according to the depth map of scene with the step of target detection point corresponding target depth and includes:
Depth data corresponding with the upper surface of object under test is obtained from the depth map of scene;
Average depth value is obtained, and using average depth value as target depth, average depth value is being averaged for depth data
Value.
Optionally, object under test is cuboid, and object under test is horizontal positioned;
It is then obtained according to the depth map of scene with the step of target detection point corresponding target depth and includes:
The point cloud coordinate of the upper surface of object under test is obtained according to the depth map of scene, according to cloud coordinate using minimum two
Multiplication is fitted an objective plane;
It obtains the distance between objective plane and preset first level face and is used as target depth.
Optionally, volume measuring method is further comprising the steps of:
The image of the scene containing object under test is obtained, obtain the profile of object under test in the picture and is labeled.
The embodiment of the present invention also provides a kind of volume measurement device, including depth map receiving unit, reference distance value connect
Receive unit, target depth acquiring unit, compensating unit, volume output unit;
Depth map receiving unit is used to receive the depth map of the scene containing object under test, and the depth map of scene is to preset the
On the basis of one horizontal plane;
Reference distance value receiving unit is used to receive reference distance value from single-point range unit, and reference distance value is target inspection
The distance between measuring point and preset first level face, target detection point are located at the upper surface of object under test;
Target depth acquiring unit is used to obtain target depth corresponding with target detection point according to the depth map of scene;
Compensating unit is used to be carried out according to depth value of the compensation factor to the depth map of scene for obtaining compensation factor
Compensation, to obtain compensation depth value;Compensation factor is the ratio of reference distance value and target depth;
Volume output unit is used for the point cloud coordinate compensated according to compensation depth value, and according to the point cloud coordinate of compensation
Obtain the volume of object under test.
Optionally, single-point range unit includes laser range finder, and laser range finder is set on preset first level face, is swashed
The detection light beam of optar is projeced into the upper surface of object under test vertically;
Then target depth acquiring unit is used to extract mesh from the depth map of scene according to the location information of laser range finder
The corresponding depth value of test point is marked as target depth.
Optionally, target depth acquiring unit is also used to obtain the upper surface pair with object under test from the depth map of scene
The depth data answered;
Target depth acquiring unit is also used to obtain average depth value, and using average depth value as target depth, average
Depth value is the average value of depth data.
Optionally, object under test is cuboid, and object under test is horizontal positioned;
Target depth acquiring unit is also used to obtain the point cloud coordinate of the upper surface of object under test according to the depth map of scene,
And an objective plane is fitted using least square method according to cloud coordinate;
Target depth acquiring unit is also used to obtain the distance between objective plane and preset first level face as target
Depth.
The embodiment of the present invention also provides a kind of volume measuring system, and volume measuring system includes depth camera mould group, list
The volume measurement device of point range unit and the embodiment of the present invention;
Depth camera mould group is used to obtain the depth map of scene;
Single-point range unit is for obtaining reference distance value.
Optionally, volume measuring system further includes image acquisition unit and mark unit;
Image acquisition unit is used to obtain the image of the scene containing object under test;
Mark unit is for obtaining the profile of object under test in the picture and being labeled.
The positive effect of the embodiment of the present invention is: the embodiment of the present invention passes through the accurate distance measured
Numerical value compensates the depth value of depth map, can be with error existing for the depth value of Corrected Depth figure, thus what raising obtained
The precision of the volume of object under test.
Detailed description of the invention
Detailed description of the invention several embodiments of the invention, and together with the description to explain the principle of the present invention.Fields
It will be recognized that the specific embodiment illustrated in figure is exemplary, and the model being not intended to be limiting of the invention
It encloses.
Fig. 1 is the structural schematic diagram of the volume measurement device of the embodiment of the present invention 1.
Fig. 2 is the structural schematic diagram of the volume measuring system of the embodiment of the present invention 1.
Fig. 3 is that the volume measuring system of the embodiment of the present invention 1 measures the schematic diagram of object under test volume.
Fig. 4 is the flow chart of the volume measuring method of the embodiment of the present invention 1.
Fig. 5 is that the volume measuring system of the embodiment of the present invention 2 measures the schematic diagram of object under test volume.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, those skilled in the art's every other embodiment obtained without making creative work,
It shall fall within the protection scope of the present invention.
It should be appreciated that ought use in this specification and in the appended claims, term " includes " and "comprising" instruction
Described feature, entirety, operation, the presence of element and/or component, but one or more of the other feature, entirety, behaviour is not precluded
Make, the presence or addition of element and/or component.
It is also understood that mesh of the term used in this description of the invention merely for the sake of description specific embodiment
And be not intended to limit the present invention.As description of the invention and it is used in the attached claims, unless on
Other situations are hereafter clearly indicated, otherwise " one " of singular, "one" and "the" are intended to include plural form.
It will be further appreciated that the term "and/or" used in description of the invention and the appended claims is
Refer to any combination and all possible combinations of one or more of associated item listed, and including these combinations.
Embodiment 1
The present embodiment provides a kind of volume measurement devices, and referring to Fig.1, which includes depth map receiving unit
101, reference distance value receiving unit 102, target depth acquiring unit 103, compensating unit 104, volume output unit 105.
Depth map receiving unit 101 is used to receive the depth map of the scene containing object under test, and the depth map of scene is with pre-
If on the basis of first level face.Reference distance value receiving unit 102 is used to receive reference distance value, reference from single-point range unit
Distance value is the distance between target detection point and preset first level face, and target detection point is located at the upper surface of object under test.
Target depth acquiring unit 103 is used to obtain target depth corresponding with target detection point according to the depth map of scene.Compensation is single
Member 104 is used to be compensated according to depth value of the compensation factor to the depth map of scene for obtaining compensation factor, to obtain
Compensation depth value;Compensation factor is the ratio of reference distance value and target depth.Volume output unit 105 is used for deep according to compensation
The point cloud coordinate that angle value is compensated, and the volume of object under test is obtained according to the point cloud coordinate of compensation.
The present embodiment also provides a kind of volume measuring system, and referring to Fig. 2, which includes depth camera mould group
2, the volume measurement device 1 of single-point range unit 3 and the present embodiment.Depth camera mould group 2 is used to obtain the depth map of scene.It is single
Point range unit 3 is for obtaining reference distance value.
As an alternative embodiment, single-point range unit 3 includes laser range finder 11.Referring to Fig. 3, laser ranging
Instrument 11 is set on preset first level face 12, and the detection light beam 14 of laser range finder 11 is projeced into object under test 13 straight down
Upper surface, the point that the detection light beam 14 of laser range finder 11 is directed toward is target detection point 15.Depth camera mould group 2 is also provided with
In on preset first level face 12, the coverage of depth camera mould group 2 is illustrated with dotted line in figure.Target depth acquiring unit
103 make for extracting the corresponding depth value of target detection point from the depth map of scene according to the location information of laser range finder 11
For target depth.
During measuring volume, depth camera mould group 2 obtains the depth map of the scene containing object under test 13.The depth
Figure is spent using the depth value of each point as pixel value.Depth map receiving unit 101 receives depth map from depth camera mould group 2.Swash
Optar 11 obtains itself and the distance between 15 h1 of target detection point.Because the detection light beam 14 of laser range finder 11 is along vertical
Direction, so distance h1 is the distance between target detection point 15 and preset first level face 12.Because of laser range finder 11
The distance measured has high precision, therefore, using distance h1 as reference distance value.Reference distance value receiving unit 102 from swash
Optar 11 receives reference distance value.
After the setup, position is known to laser range finder 11.It is according to the position that laser range finder 11 is arranged
It can determine its corresponding target detection point 15 in the position of depth map.Target depth acquiring unit 103 is according to laser range finder 11
Location information the corresponding depth value (being set as h2) of target detection point is extracted from the depth map of scene, as target depth.
Next, compensating unit 104 obtains compensation factor c1.Compensation factor c1=h1/h2.Then, compensating unit 104
It is compensated according to depth value of the compensation factor c1 to the depth map of scene, to obtain compensation depth value that is, in depth map and often
The corresponding depth value of one point is multiplied by compensation factor c1, to obtain the corresponding compensation depth value of each point.
Next, the point cloud coordinate that volume output unit 105 is compensated according to compensation depth value, sits from the point cloud of compensation
The corresponding coordinate point set of object under test 13 is filtered out in mark, is obtained further according to the corresponding coordinate point set of object under test 13 to be measured
The volume of object 13.
As an alternative embodiment, the volume measurement device of the present embodiment is realized using a chip, depth map is connect
Receive unit 101, reference distance value receiving unit 102, target depth acquiring unit 103, compensating unit 104, volume output unit
105 are integrated in the chip.In another optional embodiment, volume measurement device is realized using a processor, deep
It is defeated to spend figure receiving unit 101, reference distance value receiving unit 102, target depth acquiring unit 103, compensating unit 104, volume
The corresponding function of unit 105 is all made of computer program realization out.
As an alternative embodiment, image acquisition unit obtains the image of the scene containing object under test 13.Figure
As acquiring unit includes RGB (RGB) camera mould group, the cromogram of scene of the RGB camera mould group acquisition containing object under test 13
Picture provides visual measurement experience for user.Unit is marked for obtaining profile and the progress of object under test 13 in the picture
Mark.Unit is marked according at least one from RGB phase of bilateral filtering algorithm, Kalman filtering algorithm, Gaussian filter algorithm etc.
The edge contour of object under test 13 is identified in the image that machine mould group obtains, and the profile is marked in eye-catching mode, with convenient
User locks object under test 13.
The present embodiment also provides a kind of volume measuring method, referring to Fig. 4, should the following steps are included:
Step S301, the depth map of the scene containing object under test 13 is received.The depth map of scene is with preset first level
On the basis of face.
Step S302, reference distance value is received from single-point range unit.Reference distance value is target detection point and default the
The distance between one horizontal plane, target detection point are located at the upper surface of object under test 13.
Step S303, target depth corresponding with target detection point is obtained according to the depth map of scene.
Step S304, compensation factor is obtained, is compensated according to depth value of the compensation factor to the depth map of scene, with
To compensation depth value.Compensation factor is the ratio of reference distance value and target depth;
Step S305, the point cloud coordinate compensated according to compensation depth value, and according to the point cloud coordinate of compensation obtain to
Survey the volume of object 13.
In a kind of optional embodiment, which is realized using the volume measurement device of the present embodiment.
In another optional embodiment, the volume measuring method is real using the volume measuring system of the present embodiment
It is existing.The depth map of scene of the acquisition of depth camera mould group 2 containing object under test 13.The depth map using the depth value of each point as
Pixel value.In step S301, depth map receiving unit 101 receives depth map from depth camera mould group 2.
Laser range finder 11 obtains itself and the distance between 15 h1 of target detection point.Because of the detection light of laser range finder 11
Beam 14 along the vertical direction, so distance h1 is the distance between target detection point 15 and preset first level face 12.Because swashing
The distance that optar 11 measures has high precision, therefore, using distance h1 as reference distance value.In step s 302, join
It examines distance value receiving unit 102 and receives reference distance value from laser range finder 11.
After the setup, position is known to laser range finder 11.It is according to the position that laser range finder 11 is arranged
It can determine its corresponding target detection point 15 in the position of depth map.In step S303,103 basis of target depth acquiring unit
The location information of laser range finder 11 extracts the corresponding depth value (being set as h2) of target detection point from the depth map of scene, as
Target depth.
Next, in step s 304, compensating unit 104 obtains compensation factor c1.Compensation factor c1=h1/h2.Then,
Compensating unit 104 is compensated according to depth value of the compensation factor c1 to the depth map of scene, to obtain compensation depth value that is,
Corresponding depth value is put multiplied by compensation factor c1, to obtain the corresponding compensation depth value of each point with each in depth map.
Next, in step S305, the point cloud coordinate that volume output unit 105 is compensated according to compensation depth value,
The corresponding coordinate point set of object under test 13 is filtered out from the point cloud coordinate of compensation, further according to the corresponding coordinate of object under test 13
Point set obtains the volume of object under test 13.
As an alternative embodiment, checking for the ease of user, the volume measuring method of the present embodiment further includes
Following steps: the image of the scene containing object under test 13 is obtained;The profile for obtaining object under test 13 in the picture is gone forward side by side rower
Note.
The numerical value that the present embodiment passes through the accurate distance measured compensates the depth value of depth map, can correct depth
Error existing for the depth value of figure is spent, to improve the precision of the volume of obtained object under test 13.Using one in depth map
Operand can be effectively reduced as target depth in the depth value of a point.
Embodiment 2
The present embodiment provides a kind of volume measurement device, the volume measurement device bases of the volume measurement device and embodiment 1
This is identical, and difference is, the target depth acquiring unit 103 of the volume measurement device of the present embodiment is different.In the present embodiment,
Referring to Fig. 5, object under test 13 is cuboid, and object under test 13 is horizontal positioned.Target depth acquiring unit 103 is according to scene
Pixel set corresponding with object under test 13 is obtained in depth map.Because the corresponding depth value of object under test 13 (is held with background
Carry object under test 13 pedestal) depth value have notable difference therefore can be by object under test 13 and background according to depth value
It distinguishes, to obtain depth data corresponding with the upper surface of object under test from the depth map of scene.The depth data includes
Depth value corresponding with each pixel of the upper surface of object under test in depth map.Then, target depth acquiring unit 103 obtains
Take average depth value as target depth, average depth value is the average value of depth data, the as corresponding institute of object under test 13
There is the average value of the depth value of pixel, that is, after the depth value of the corresponding all pixels of object under test 13 is summed, divided by be measured
The quantity of the corresponding pixel of object 13.Because object under test 13 is cuboid, object under test 13 is horizontal positioned, so, it is average deep
Angle value can characterize between the upper surface and preset first level face 12 of the object under test 13 obtained by depth camera mould group 2
Distance.Using average depth value as target depth, the random error bring that can reduce the depth value of each point influences, and improves
Precision.
The other parts of the volume measurement device of the present embodiment and the volume measurement device of embodiment 1 are essentially identical, herein
It repeats no more.
The present embodiment also provides a kind of volume measuring system, the volume measuring system of the present embodiment and the volume of embodiment 1
Measuring system is essentially identical, and difference is, the volume measuring system of the present embodiment includes the volume measurement device of the present embodiment,
That is, target depth acquiring unit 103 has differences.
During measuring volume, depth camera mould group 2 obtains the depth map of the scene containing object under test 13.The depth
Figure is spent using the depth value of each point as pixel value.Depth map receiving unit 101 receives depth map from depth camera mould group 2.Swash
Optar 11 obtains itself and the distance between 15 h1 of target detection point.Because the detection light beam 14 of laser range finder 11 is along vertical
Direction, so distance h1 is the distance between target detection point 15 and preset first level face 12.Because of laser range finder 11
The distance measured has high precision, therefore, using distance h1 as reference distance value.Reference distance value receiving unit 102 from swash
Optar 11 receives reference distance value.
Target depth acquiring unit 103 is according to acquisition pixel set corresponding with object under test 13 in the depth map of scene.
Because the corresponding depth value of object under test 13 and the depth value of background (carrying the pedestal of object under test 13) have notable difference,
Therefore, object under test 13 and background can be distinguished according to depth value.Then, target depth acquiring unit 103 obtains average deep
For angle value h3 as target depth, average depth value h3 is the average value of the depth value of the corresponding all pixels of object under test 13,
That is, after the depth value of the corresponding all pixels of object under test 13 is summed, divided by the quantity of the corresponding pixel of object under test 13.Cause
Object under test 13 is cuboid, and object under test 13 is horizontal positioned, so, average depth value can be characterized through depth camera mould
The distance between upper surface and preset first level face 12 of the object under test 13 that group 2 obtains.
Next, compensating unit 104 obtains compensation factor c1.Compensation factor c1=h1/h3.Then, compensating unit 104
It is compensated according to the corresponding depth value of compensation factor c1 measuring targets 13, to obtain the first compensation depth value that is, depth map
Each of depth value corresponding with object under test 13 multiplied by compensation factor c1, to obtain corresponding with object under test 13
One compensation depth value.
Next, volume output unit 105 according to first compensation depth value corresponding with object under test 13 obtain with it is to be measured
The point cloud coordinate of the corresponding compensation of object 13 obtains object under test further according to the point cloud coordinate of compensation corresponding with object under test 13
13 volume.
The present embodiment also provides a kind of volume measuring method, the volume measuring method of the present embodiment and the volume of embodiment 1
Measurement method is essentially identical, and difference is that the step of obtaining target depth is distinct.
In a kind of optional embodiment, which is realized using the volume measurement device of the present embodiment.
In another optional embodiment, the volume measuring method is real using the volume measuring system of the present embodiment
It is existing.The depth map of scene of the acquisition of depth camera mould group 2 containing object under test 13.The depth map using the depth value of each point as
Pixel value.In step S301, depth map receiving unit 101 receives depth map from depth camera mould group 2.
Laser range finder 11 obtains itself and the distance between 15 h1 of target detection point.Because of the detection light of laser range finder 11
Beam 14 along the vertical direction, so distance h1 is the distance between target detection point 15 and preset first level face 12.Because swashing
The distance that optar 11 measures has high precision, therefore, using distance h1 as reference distance value.In step s 302, join
It examines distance value receiving unit 102 and receives reference distance value from laser range finder 11.
In step S303, target depth acquiring unit 103 is right with object under test 13 according to obtaining in the depth map of scene
The pixel set answered.Because of the depth value of the corresponding depth value of object under test 13 and background (carrying the pedestal of object under test 13)
With notable difference, therefore, object under test 13 and background can be distinguished according to depth value.Then, target depth acquiring unit
103 obtain average depth value h3 as target depth, and average depth value h3 is the depth of the corresponding all pixels of object under test 13
The average value of value, that is, after the depth value of the corresponding all pixels of object under test 13 is summed, it is corresponding divided by object under test 13
The quantity of pixel.Because object under test 13 is cuboid, object under test 13 is horizontal positioned, so, average depth value can characterize
The distance between upper surface and preset first level face 12 of the object under test 13 obtained by depth camera mould group 2.
Next, in step s 304, compensating unit 104 obtains compensation factor c1.Compensation factor c1=h1/h3.Then,
Compensating unit 104 is compensated according to the corresponding depth value of compensation factor c1 measuring targets 13, to obtain the first compensation depth
Value that is, each of depth map depth value corresponding with object under test 13 multiplied by compensation factor c1, with obtain with it is to be measured
The corresponding first compensation depth value of object 13.
Next, volume output unit 105 is according to first compensation depth corresponding with object under test 13 in step S305
Value obtains the point cloud coordinate of compensation corresponding with object under test 13, further according to the point cloud coordinate of compensation corresponding with object under test 13
Obtain the volume of object under test 13.
Embodiment 3
The present embodiment provides a kind of volume measurement device, the volume measurement device bases of the volume measurement device and embodiment 1
This is identical, and difference is, the target depth acquiring unit 103 of the volume measurement device of the present embodiment is different.In the present embodiment,
Referring to Fig. 5, object under test 13 is cuboid, and object under test 13 is horizontal positioned.Target depth acquiring unit 103 is according to scene
The point cloud coordinate of depth map acquisition scene.Then, the corresponding coordinate points of object under test 13 are filtered out from the point cloud coordinate of scene
Set.That is, target depth acquiring unit 103 obtains the coordinate of each point corresponding with the upper surface of object under test 13.It connects down
Come, target depth acquiring unit 103 is according to the coordinate of the corresponding each point in upper surface of object under test 13, according to least square method
It is fitted the function of plane corresponding with the upper surface of object under test 13, the referred to as function of objective plane.Then, target depth obtains
Unit 103 obtains the distance between objective plane and preset first level face 12 h4.Preset first level face 12 is provided with
Afterwards, corresponding function is known.Therefore, the distance between objective plane and preset first level face 12 are can to obtain
's.Target depth acquiring unit 103 is using distance h4 as target depth.Because object under test 13 is cuboid, object under test 13 is water
Placing flat, so, objective plane tool has high degree of fitting with the plane where the upper surface of object under test 13, and objective plane can
The plane where upper surface to characterize object under test 13, in turn, distance h4 can be characterized to be obtained by depth camera mould group 2
The distance between upper surface and preset first level face 12 of object under test 13.
The other parts of the volume measurement device of the present embodiment and the volume measurement device of embodiment 1 are essentially identical, herein
It repeats no more.
The present embodiment also provides a kind of volume measuring system, the volume measuring system of the present embodiment and the volume of embodiment 1
Measuring system is essentially identical, and difference is, the volume measuring system of the present embodiment includes the volume measurement device of the present embodiment,
That is, target depth acquiring unit 103 has differences.
During measuring volume, depth camera mould group 2 obtains the depth map of the scene containing object under test 13.The depth
Figure is spent using the depth value of each point as pixel value.Depth map receiving unit 101 receives depth map from depth camera mould group 2.Swash
Optar 11 obtains itself and the distance between 15 h1 of target detection point.Because the detection light beam 14 of laser range finder 11 is along vertical
Direction, so distance h1 is the distance between target detection point 15 and preset first level face 12.Because of laser range finder 11
The distance measured has high precision, therefore, using distance h1 as reference distance value.Reference distance value receiving unit 102 from swash
Optar 11 receives reference distance value.
Target depth acquiring unit 103 obtains the point cloud coordinate of scene according to the depth map of scene.Then, from the point of scene
The corresponding coordinate point set of object under test 13 is filtered out in cloud coordinate.That is, target depth acquiring unit 103 obtains and determinand
The coordinate of the corresponding each point in the upper surface of body 13.Next, upper table of the target depth acquiring unit 103 according to object under test 13
The coordinate of the corresponding each point in face is fitted the function of plane corresponding with the upper surface of object under test 13 according to least square method,
The referred to as function of objective plane.Then, target depth acquiring unit 103 obtains between objective plane and preset first level face 12
Distance h4.After being provided with of preset first level face 12, corresponding function is known.Therefore, objective plane and pre-
If the distance between first level face 12 can obtain.Target depth acquiring unit 103 is using distance h4 as target depth.Cause
Object under test 13 is cuboid, and object under test 13 is horizontal positioned, so, objective plane can characterize the upper table of object under test 13
Plane where face, in turn, distance h4 can characterize the upper surface of the object under test 13 obtained by depth camera mould group 2 and pre-
If the distance between first level face 12.
Next, compensating unit 104 obtains compensation factor c1.Compensation factor c1=h1/h4.Then, compensating unit 104
It is compensated according to the corresponding depth value of each point of compensation factor c1 measuring targets 13, to obtain the first compensation depth value
That is, each of depth map depth value corresponding with object under test 13 is multiplied by compensation factor c1, to obtain and object under test
13 corresponding first compensation depth values.
Next, volume output unit 105 according to first compensation depth value corresponding with object under test 13 obtain with it is to be measured
The point cloud coordinate of the corresponding compensation of object 13 obtains determinand further according to and with the point cloud coordinate of the corresponding compensation of object under test 13
The volume of body 13.
The present embodiment also provides a kind of volume measuring method, the volume measuring method of the present embodiment and the volume of embodiment 1
Measurement method is essentially identical, and difference is that the step of obtaining target depth is distinct.
In a kind of optional embodiment, which is realized using the volume measurement device of the present embodiment.
In another optional embodiment, the volume measuring method is real using the volume measuring system of the present embodiment
It is existing.The depth map of scene of the acquisition of depth camera mould group 2 containing object under test 13.The depth map using the depth value of each point as
Pixel value.In step S301, depth map receiving unit 101 receives depth map from depth camera mould group 2.
Laser range finder 11 obtains itself and the distance between 15 h1 of target detection point.Because of the detection light of laser range finder 11
Beam 14 along the vertical direction, so distance h1 is the distance between target detection point 15 and preset first level face 12.Because swashing
The distance that optar 11 measures has high precision, therefore, using distance h1 as reference distance value.In step s 302, join
It examines distance value receiving unit 102 and receives reference distance value from laser range finder 11.
In step S303, target depth acquiring unit 103 obtains the point cloud coordinate of scene according to the depth map of scene.So
Afterwards, the corresponding coordinate point set of object under test 13 is filtered out from the point cloud coordinate of scene.That is, target depth acquiring unit
103 obtain the coordinate of each point corresponding with the upper surface of object under test 13.Next, 103 basis of target depth acquiring unit
The coordinate of the corresponding each point in the upper surface of object under test 13, according to the upper surface pair of least square method fitting and object under test 13
The function for the plane answered, the referred to as function of objective plane.Then, target depth acquiring unit 103 obtains objective plane and presets
The distance between first level face 12 h4.After being provided with of preset first level face 12, corresponding function is known.Cause
This, the distance between objective plane and preset first level face 12 can obtain.Target depth acquiring unit 103 is with distance
H4 is target depth.Because object under test 13 is cuboid, object under test 13 is horizontal positioned, so, objective plane can characterize
Plane where the upper surface of object under test 13, in turn, distance h4 can characterize the determinand obtained by depth camera mould group 2
The distance between upper surface and preset first level face 12 of body 13.
Next, in step s 304, compensating unit 104 obtains compensation factor c1.Compensation factor c1=h1/h4.Then,
Compensating unit 104 is compensated according to the corresponding depth value of compensation factor c1 measuring targets 13, to obtain the first compensation depth
Value that is, each of depth map depth value corresponding with object under test 13 multiplied by compensation factor c1, with obtain with it is to be measured
The corresponding first compensation depth value of object 13.
Next, volume output unit 105 is according to first compensation depth corresponding with object under test 13 in step S305
Value obtains the point cloud coordinate of compensation corresponding with object under test 13, obtains further according to coordinate point set corresponding with object under test 13
The volume of object under test 13.
Although specific embodiments of the present invention have been described above, it will be appreciated by those of skill in the art that this is only
For example, protection scope of the present invention is to be defined by the appended claims.Those skilled in the art without departing substantially from
Under the premise of the principle and substance of the present invention, many changes and modifications may be made, but these change and
Modification each falls within protection scope of the present invention.
Claims (10)
1. a kind of volume measuring method, which comprises the following steps:
The depth map of the scene containing object under test is received, the depth map of the scene is on the basis of preset first level face;
Reference distance value is received from single-point range unit, the reference distance value is target detection point and the preset first level
The distance between face, the target detection point are located at the upper surface of the object under test;
Target depth corresponding with the target detection point is obtained according to the depth map of the scene;
Compensation factor is obtained, is compensated according to depth value of the compensation factor to the depth map of the scene, to be mended
Repay depth value;The compensation factor is the ratio of the reference distance value and the target depth;
The point cloud coordinate compensated according to the compensation depth value, and obtained according to the point cloud coordinate of the compensation described to be measured
The volume of object.
2. volume measuring method as described in claim 1, which is characterized in that the single-point range unit includes laser ranging
Instrument, the laser range finder are set on the preset first level face, and the detection light beam of the laser range finder projects vertically
In the upper surface of the object under test.
3. volume measuring method as claimed in claim 2, which is characterized in that it is described according to the depth map of scene obtain with it is described
The step of target detection point corresponding target depth includes:
It is corresponding that the target detection point is extracted from the depth map of the scene according to the location information of the laser range finder
Depth value is as the target depth.
4. volume measuring method as claimed in claim 1 or 2, which is characterized in that it is described according to the depth map of scene obtain with
The step of target detection point corresponding target depth includes:
Depth data corresponding with the upper surface of the object under test is obtained from the depth map of the scene;
Average depth value is obtained, and using the average depth value as the target depth, the average depth value is the depth
The average value of degree evidence.
5. volume measuring method as claimed in claim 1 or 2, which is characterized in that the object under test be cuboid, it is described to
It is horizontal positioned for surveying object;
Then described obtained according to the depth map of scene with the step of target detection point corresponding target depth includes:
The point cloud coordinate of the upper surface of the object under test is obtained according to the depth map of the scene, is adopted according to described cloud coordinate
An objective plane is fitted with least square method;
It obtains the distance between the objective plane and the preset first level face and is used as the target depth.
6. a kind of volume measurement device, which is characterized in that including depth map receiving unit, reference distance value receiving unit, target
Depth acquiring unit, compensating unit, volume output unit;
The depth map receiving unit is used to receive the depth map of the scene containing object under test, and the depth map of the scene is with pre-
If on the basis of first level face;
The reference distance value receiving unit is used to receive reference distance value from single-point range unit, and the reference distance value is mesh
The distance between test point and the preset first level face are marked, the target detection point is located at the upper table of the object under test
Face;
The target depth acquiring unit is used to obtain target corresponding with the target detection point according to the depth map of scene deep
Degree;
The compensating unit is used for according to the compensation factor for obtaining compensation factor to the depth of the depth map of the scene
Angle value compensates, to obtain compensation depth value;The compensation factor is the ratio of the reference distance value and the target depth
Value;
The volume output unit is used for the point cloud coordinate compensated according to the compensation depth value, and according to the compensation
Point cloud coordinate obtains the volume of the object under test.
7. volume measurement device as claimed in claim 6, which is characterized in that the single-point range unit includes laser ranging
Instrument, the laser range finder are set on the preset first level face, and the detection light beam of the laser range finder projects vertically
In the upper surface of the object under test;
Then the target depth acquiring unit is used for the depth map according to the location information of the laser range finder from the scene
The middle corresponding depth value of the target detection point that extracts is as the target depth.
8. volume measurement device as claimed in claims 6 or 7, which is characterized in that the target depth acquiring unit is also used to
Depth data corresponding with the upper surface of the object under test is obtained from the depth map of the scene;
The target depth acquiring unit is also used to obtain average depth value, and deep using the average depth value as the target
Degree, the average depth value are the average value of the depth data.
9. volume measurement device as claimed in claims 6 or 7, which is characterized in that the object under test be cuboid, it is described to
It is horizontal positioned for surveying object;
The target depth acquiring unit is also used to obtain the upper surface of the object under test according to the depth map of the scene
Point cloud coordinate, and an objective plane is fitted using least square method according to described cloud coordinate;
The target depth acquiring unit is also used to obtain the distance between the objective plane and the preset first level face
As the target depth.
10. a kind of volume measuring system, which is characterized in that the volume measuring system includes depth camera mould group, single-point ranging
Device and the volume measurement device as described in any one in claim 6-9;
The depth camera mould group is used to obtain the depth map of the scene;
The single-point range unit is for obtaining the reference distance value;
The volume measuring system further includes image acquisition unit and mark unit;
Described image acquiring unit is used to obtain the image of the scene containing the object under test;
The mark unit is used to obtain the profile of the object under test in described image and is labeled.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910741396.8A CN110319899A (en) | 2019-08-12 | 2019-08-12 | Volume measuring method, device and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910741396.8A CN110319899A (en) | 2019-08-12 | 2019-08-12 | Volume measuring method, device and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110319899A true CN110319899A (en) | 2019-10-11 |
Family
ID=68126004
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910741396.8A Pending CN110319899A (en) | 2019-08-12 | 2019-08-12 | Volume measuring method, device and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110319899A (en) |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101533529A (en) * | 2009-01-23 | 2009-09-16 | 北京建筑工程学院 | Range image-based 3D spatial data processing method and device |
CN102609941A (en) * | 2012-01-31 | 2012-07-25 | 北京航空航天大学 | Three-dimensional registering method based on ToF (Time-of-Flight) depth camera |
CN103927782A (en) * | 2014-01-06 | 2014-07-16 | 河南科技大学 | Method for depth image surface fitting |
US20150365652A1 (en) * | 2014-06-13 | 2015-12-17 | Lips Corporation | Depth camera system |
CN106767933A (en) * | 2017-02-10 | 2017-05-31 | 深圳奥比中光科技有限公司 | The measuring system of depth camera error, measuring method, evaluation method and compensation method |
CN107255821A (en) * | 2017-06-07 | 2017-10-17 | 旗瀚科技有限公司 | A kind of method for splicing simulated laser radar data based on many depth cameras |
CN107727220A (en) * | 2017-10-11 | 2018-02-23 | 上海展扬通信技术有限公司 | A kind of human body measurement method and body measurement system based on intelligent terminal |
US20180093781A1 (en) * | 2016-09-30 | 2018-04-05 | Sony Interactive Entertainment Inc. | Unmanned aerial vehicle movement via environmental interactions |
CN108010071A (en) * | 2017-12-01 | 2018-05-08 | 中国人民解放军后勤工程学院 | A kind of Luminance Distribution measuring system and method using 3D depth surveys |
-
2019
- 2019-08-12 CN CN201910741396.8A patent/CN110319899A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101533529A (en) * | 2009-01-23 | 2009-09-16 | 北京建筑工程学院 | Range image-based 3D spatial data processing method and device |
CN102609941A (en) * | 2012-01-31 | 2012-07-25 | 北京航空航天大学 | Three-dimensional registering method based on ToF (Time-of-Flight) depth camera |
CN103927782A (en) * | 2014-01-06 | 2014-07-16 | 河南科技大学 | Method for depth image surface fitting |
US20150365652A1 (en) * | 2014-06-13 | 2015-12-17 | Lips Corporation | Depth camera system |
US20180093781A1 (en) * | 2016-09-30 | 2018-04-05 | Sony Interactive Entertainment Inc. | Unmanned aerial vehicle movement via environmental interactions |
CN106767933A (en) * | 2017-02-10 | 2017-05-31 | 深圳奥比中光科技有限公司 | The measuring system of depth camera error, measuring method, evaluation method and compensation method |
CN107255821A (en) * | 2017-06-07 | 2017-10-17 | 旗瀚科技有限公司 | A kind of method for splicing simulated laser radar data based on many depth cameras |
CN107727220A (en) * | 2017-10-11 | 2018-02-23 | 上海展扬通信技术有限公司 | A kind of human body measurement method and body measurement system based on intelligent terminal |
CN108010071A (en) * | 2017-12-01 | 2018-05-08 | 中国人民解放军后勤工程学院 | A kind of Luminance Distribution measuring system and method using 3D depth surveys |
Non-Patent Citations (1)
Title |
---|
商佳尚: "差值法与比值法的误差修正效果比较研究及应用", 《计测技术》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106969706A (en) | Workpiece sensing and three-dimension measuring system and detection method based on binocular stereo vision | |
CN108986070B (en) | Rock crack propagation experiment monitoring method based on high-speed video measurement | |
CN106392304B (en) | A kind of laser assisted weld seam Intelligent tracing system and method | |
US20180108143A1 (en) | Height measuring system and method | |
CN108733053A (en) | A kind of Intelligent road detection method based on robot | |
CN107044821A (en) | A kind of measuring method and system of contactless tubing object | |
CN102063718A (en) | Field calibration and precision measurement method for spot laser measuring system | |
US10290117B2 (en) | System for extracting position information of object in point cloud data by using component | |
CN107869954B (en) | Binocular vision volume weight measurement system and implementation method thereof | |
CN110672020A (en) | Stand tree height measuring method based on monocular vision | |
CN107101582A (en) | Axial workpiece run-out error On-line Measuring Method based on structure light vision | |
CN104034733A (en) | Service life prediction method based on binocular vision monitoring and surface crack image recognition | |
CN103196370A (en) | Measuring method and measuring device of conduit connector space pose parameters | |
CN107356202A (en) | A kind of laser scanning measurement system target sights method automatically | |
KR20130033374A (en) | Surveying method | |
CN105405126A (en) | Multi-scale air-ground parameter automatic calibration method based on monocular vision system | |
EP1459035B1 (en) | Method for determining corresponding points in stereoscopic three-dimensional measurements | |
CN107345789A (en) | A kind of pcb board hole location detecting device and method | |
CN105894511A (en) | Calibration target setting method and device and parking auxiliary system | |
CN107504917B (en) | Three-dimensional size measuring method and device | |
CN107271445B (en) | Defect detection method and device | |
CN103175512B (en) | Shooting measurement method of attitude of tail end of boom of concrete pump truck | |
CN109035343A (en) | A kind of floor relative displacement measurement method based on monitoring camera | |
CN108871185A (en) | Method, apparatus, equipment and the computer readable storage medium of piece test | |
CN110517314A (en) | Method, apparatus and computer readable storage medium are determined based on the pallet pose of TOF 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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20191011 |