CN108921842A - A kind of cereal flow detection method and device - Google Patents
A kind of cereal flow detection method and device Download PDFInfo
- Publication number
- CN108921842A CN108921842A CN201810720353.7A CN201810720353A CN108921842A CN 108921842 A CN108921842 A CN 108921842A CN 201810720353 A CN201810720353 A CN 201810720353A CN 108921842 A CN108921842 A CN 108921842A
- Authority
- CN
- China
- Prior art keywords
- back side
- image
- camera
- pixel
- grain stream
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01F—MEASURING VOLUME, VOLUME FLOW, MASS FLOW OR LIQUID LEVEL; METERING BY VOLUME
- G01F13/00—Apparatus for measuring by volume and delivering fluids or fluent solid materials, not provided for in the preceding groups
- G01F13/001—Apparatus for measuring by volume and delivering fluids or fluent solid materials, not provided for in the preceding groups for fluent solid material
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30128—Food products
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Fluid Mechanics (AREA)
- Indicating And Signalling Devices For Elevators (AREA)
Abstract
The invention discloses a kind of cereal flow detection method, specific steps include:Obtain the direct picture and back side image of grain stream in crop elevator;The direct picture is matched, the front point cloud chart of the grain stream is generated;The back side image is matched, the back side point cloud chart of the grain stream is generated;Based on the front point cloud chart and back side point cloud chart, the volume of grain stream is obtained.It the method achieve the non-cpntact measurement of cereal flow, does not need to handle multi collect image, only a frame image need to be carried out to resolve the volume that cereal can be obtained, and then obtain cereal flow.
Description
Technical field
The present invention relates to field of measuring technique more particularly to a kind of cereal flow detection method and device.
Background technique
Precision agriculture is that the fine farming technique of modern agriculture of crop management is carried out based on information collection and intelligent control.Its
Core is that chemical fertilizer is carried out according to the actual needs of crops, and the investment of pesticide, to reach raising yield, save the cost is reduced dirty
The purpose of dye.The growth situation and its output distribotion difference for specifying crops are the key that precision agricultures, obtain crop yield
Distribution is to implement the first step of precision agriculture.In the mechanized equipment technological innovation for realizing precision agriculture, with fastest developing speed surely belongs to
Close corn mowing machine from the Intelligent joint of belt sensor, so-called joint corn mowing machine be exactly the primary harvesting for completing cereal crops,
The processes such as more than threshing, separation stem, removing are miscellaneous, the cereal harvesting machinery of grain is directly obtained from field.Abbreviation cereal stripper,
Also translations cereal combine, core are to be mounted on the sensor that crop yield is obtained on cereal stripper.
At present have now been developed come flow sensor mainly have impulse type sensor, based on gamma-rays cereal flow pass
The sensors such as sensor and photoelectric type corn flow sensor.The sensor of impulse type is based on theorem of impulse.It is kept off when cereal impacts
Velocity variations will be will lead to when plate, while the variation of power can be experienced on baffle, the variation for measuring power output can extrapolate cereal
Flow.Impulse type cereal flow transducer is widely used using safe, but lacking there are installation and debugging difficulty and low precision
Point, the promotion and application being unfavorable in China Agricultural Structure Adjustment technology.Although being surveyed based on gamma-rays cereal flow transducer
Accuracy of measurement is high, but it involves great expense, and gamma-rays has human body potentially hazardous, these disadvantages limit this sensor
It promotes.
Summary of the invention
The object of the present invention is to provide a kind of cereal flow detection methods, realize the non-cpntact measurement to cereal.
To solve the above problems, the first aspect of the present invention provides a kind of cereal flow detection method, including:Obtain paddy
The direct picture and back side image of grain stream in object elevator;The direct picture is matched, the grain stream is generated
Front point cloud chart;The back side image is matched, the back side point cloud chart of the grain stream is generated;Based on the positive millet cake cloud
Figure and back side point cloud chart, obtain the volume of grain stream.
Further, in crop elevator grain stream direct picture, including:Front LOOK LEFT image and front LOOK RIGHT
Image;The back side image of grain stream in crop elevator, including:Back side LOOK LEFT image and back side LOOK RIGHT image.
Further, direct picture is matched, the front point cloud chart for generating grain stream includes:To front LOOK LEFT figure
Picture and front LOOK RIGHT image are calibrated;Based on preset matching algorithm to each of front LOOK LEFT image pixel
Point finds corresponding pixel in the LOOK RIGHT image of front;Front LOOK LEFT image is calculated based on preset algorithm and front is right
The parallax of each pixel in multi-view image;Based on parallax and parameter is prestored, calculates the depth of each pixel in direct picture
Coordinate;Based on the depth coordinate of pixel each in direct picture, the abscissa and ordinate of each pixel are calculated;Based on just
Abscissa, ordinate and the depth coordinate of all pixels point in the image of face generate front point cloud chart.
Further, it by parallax and prestores parameter substitution formula (1) and calculates, obtain each pixel in direct picture
Depth coordinate;Formula (1) is z1=(f1*T1)/d1;Wherein, z1Indicate the depth coordinate of each pixel in direct picture;f1It indicates
Acquire the focal length of direct picture camera;T1Indicate the baseline length of acquisition front camera;d1Indicate front left images parallax.
Further, back side image is matched, the back side point cloud chart for generating grain stream includes:To back side LOOK LEFT figure
Picture and back side LOOK RIGHT image are calibrated;Based on preset matching algorithm to each of back side LOOK LEFT image pixel
Point overleaf finds corresponding pixel in LOOK RIGHT image;Back side LOOK LEFT image is calculated based on preset algorithm and the back side is right
The parallax of each pixel in multi-view image;Based on parallax and parameter is prestored, calculates the depth of each pixel in back side image
Coordinate;Based on the depth coordinate of pixel each in back side image, the abscissa and ordinate of each pixel are calculated;Based on back
Abscissa, ordinate and the depth coordinate of all pixels point in the image of face generate back side point cloud chart.
Further, it by parallax and prestores parameter substitution formula (2) and calculates, obtain each picture in the LOOK LEFT image of the back side
The depth coordinate of vegetarian refreshments;Formula (2) is z2=(f2*T2)/d2;Wherein, z2Indicate the depth coordinate of each pixel in back side image;
f2Indicate the focal length of acquisition back side image camera; T2Indicate the baseline length of acquisition back side camera;d2Indicate back side left images
Parallax.
Further, it is based on the front point cloud chart and back side point cloud chart, the volume for obtaining grain stream includes:It is sat according to z
Mark and projective transformation can calculate the x of corresponding points, y-coordinate;Cereal appearance is calculated according to the three-dimensional coordinate of cereal surface point
Volume of the face to elevator surrounding;The volume that this section of elevator is calculated in conjunction with the section of elevator, the paddy calculated before subtracting
Exterior surface to elevator surrounding volume to get arrive cereal volume.
Another aspect of the present invention discloses a kind of cereal flow detection device, including:First binocular camera, the second binocular
Camera and the raspberry pie communicated to connect respectively with the first binocular camera and the second binocular camera;First binocular camera is arranged in cereal
The side of elevator is sent to the raspberry for acquiring the direct picture of grain stream in crop elevator, and by direct picture
Group;The side opposite with the first binocular camera of crop elevator is arranged in second binocular camera, for acquiring crop elevator
The back side image of interior grain stream, and back side image is sent to the raspberry pie;Raspberry pie includes memory and processor, storage
The step of being stored with computer program on device, the above method realized when program is executed by the processor.
Further, the first binocular camera includes first shell and setting in the left camera of first shell intracorporal first and the
One right camera;First left camera is used to acquire the front LOOK LEFT image of grain stream in crop elevator;First right camera shooting
Head is for acquiring the front LOOK RIGHT image of grain stream in crop elevator;Second binocular camera includes that second shell and setting exist
The left camera of second shell intracorporal second and the second right camera;Second left camera is for acquiring grain stream in crop elevator
Back side LOOK LEFT image;Second right camera is used to acquire the back side LOOK RIGHT image of grain stream in crop elevator.
Further, first shell and second shell include:Shell, bolt, nut, camera support, window;Left camera shooting
Head and the angle between right camera and horizontal plane are 20-50 degree, further include light compensating lamp.
Above-mentioned technical proposal of the invention has following beneficial technical effect:Realize the non-contact survey of cereal flow
Amount, does not need to handle multi collect image, need to only carry out resolving the volume that cereal can be obtained to a frame image, in turn
Obtain flow.
Detailed description of the invention
Fig. 1 is the flow chart according to the cereal flow detection method of the application embodiment;
Fig. 2 is the flow chart that front point cloud chart is obtained according to the application embodiment;
Fig. 3 is the flow chart that back side point cloud chart is obtained according to the application embodiment;
Fig. 4 is the flow chart that the volume of grain stream is obtained according to the application embodiment;
Fig. 5 is according to the application embodiment cereal flow detection device main view;
Fig. 6 is according to the application embodiment cereal flow detection device top view;
Fig. 7 is according to the application embodiment cereal flow detection device partial sectional view;
Fig. 8 is according to the application embodiment cereal flow detection device explosive view;
Fig. 9 is the view that elevator is installed to according to the application embodiment cereal flow detection device.
Appended drawing reference:
1:Include nut;2:Bolt;3:Shell;4:Bolt;5:Nut;6:Bracket;7:Camera;8:Nut;9:Bolt;10:
Window.
Specific embodiment
In order to make the objectives, technical solutions and advantages of the present invention clearer, With reference to embodiment and join
According to attached drawing, the present invention is described in more detail.It should be understood that these descriptions are merely illustrative, and it is not intended to limit this hair
Bright range.In addition, in the following description, descriptions of well-known structures and technologies are omitted, to avoid this is unnecessarily obscured
The concept of invention.
Obviously, described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on the present invention
In embodiment, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
Fig. 1 is the flow chart according to the cereal flow detection method of the application embodiment.
As shown in Figure 1, according to an aspect of an embodiment of the present invention, a kind of cereal flow detection method is provided, including
Following steps:
S1 obtains the direct picture and back side image of grain stream in crop elevator;
Direct picture and back side image pixel coverage are 2,000,000-1,200 ten thousand, preferably 300-1000 ten thousand, more preferably
500-700 ten thousand.In this pixel coverage, image clearly, and it is convenient for resolving.
S2 matches direct picture, generates the front point cloud chart of the grain stream;
Point cloud chart is the image of the point data set of the object appearance surfaces obtained by measuring instrument.
S3 matches back side image, generates the back side point cloud chart of the grain stream;
S4 is based on front point cloud chart and back side point cloud chart, obtains the volume of grain stream.
The non-cpntact measurement that cereal flow is realized by the above method does not need to handle multi collect image,
Only a frame image need to be carried out to resolve the volume that cereal can be obtained, and then obtain flow.
In an alternative embodiment, the direct picture of grain stream in crop elevator, including:Front LOOK LEFT image and just
Face LOOK RIGHT image;The back side image of grain stream in crop elevator, including:Back side LOOK LEFT image and back side LOOK RIGHT figure
Picture.
Fig. 2 is the flow chart that front point cloud chart is obtained according to the application embodiment.
As shown in Fig. 2, matching in an alternative embodiment to direct picture, the front point cloud chart of grain stream is generated
Specific steps include:
S21 calibrates front LOOK LEFT image and front LOOK RIGHT image, and the specific implementation steps are as follows:
(1) two cameras are demarcated, the calibration of the internal reference including camera itself and two camera position parameters
Calibration.
(2) according to the calibrating parameters of (1), corrected using binocular so that two images to polar curve in the same horizontal line,
Any point and its corresponding points on another piece image, only need to be at these with regard to inevitable line number having the same in this way on piece image
Row, which carries out linear search, can be matched to corresponding points.
S22 is based on preset matching algorithm to each of front LOOK LEFT image pixel, in front LOOK RIGHT figure
Corresponding pixel is found as in;S23 calculates the parallax of each pixel in front LOOK LEFT image and front LOOK RIGHT image;
S24 is based on parallax and prestores parameter, calculates the depth coordinate of each pixel in direct picture;S25 is based on every in direct picture
The depth coordinate of a pixel calculates the abscissa and ordinate of each pixel;S26 is based on all pixels point in direct picture
Abscissa, ordinate and depth coordinate, generate front point cloud chart.
Preset matching algorithm is BM matching algorithm, and BM matching algorithm is the abbreviation of BlockMatch algorithm, is a kind of non-
Normal efficient searching algorithm.
It in an alternative embodiment, by parallax and prestores parameter substitution formula (1) and calculates, obtain each in direct picture
The depth coordinate of pixel;Formula (1) is z1=(f1*T1)/d1;Wherein, z1Indicate that the depth of each pixel in direct picture is sat
Mark;f1Indicate the focal length of acquisition direct picture camera; T1Indicate the baseline length of acquisition front camera;d1Indicate front left and right figure
As parallax.
Fig. 3 is the flow chart that back side point cloud chart is obtained according to the application embodiment.
As shown in figure 3, matching in an alternative embodiment to back side image, the back side point cloud chart of grain stream is generated
Specific steps include:S31 calibrates back side LOOK LEFT image and back side LOOK RIGHT image;S32 is calculated based on preset matching
Method overleaf finds corresponding pixel in LOOK RIGHT image to each of back side LOOK LEFT image pixel;S33 base
The parallax of each pixel in back side LOOK LEFT image and back side LOOK RIGHT image is calculated in preset algorithm;S34 be based on parallax and
Parameter is prestored, the depth coordinate of each pixel in back side image is calculated;Depth of the S35 based on pixel each in back side image
Coordinate calculates the abscissa and ordinate of each pixel;S36 is based on the abscissa of all pixels point, vertical seat in back side image
Mark and depth coordinate generate back side point cloud chart.
In an alternative embodiment, by parallax and prestores parameter substitution formula (2) and calculate, obtain back side LOOK LEFT image
In each pixel depth coordinate;Formula (2) is z2=(f2*T2)/d2;Wherein, z2Indicate each pixel in back side image
Depth coordinate;f2Indicate the focal length of acquisition back side image camera;T2Indicate the baseline length of acquisition back side camera;d2Indicate the back side
Left images parallax.
Fig. 4 is the flow chart that the volume of grain stream is obtained according to the application embodiment.
As shown in figure 4, being based on the front point cloud chart and back side point cloud chart in an alternative embodiment, obtaining grain stream
Volume include:The x of corresponding points, y coordinate can be calculated according to z coordinate and projective transformation;According to the three-dimensional of cereal surface point
Coordinate calculates cereal outer surface to the volume of elevator surrounding;The body of this section of elevator is calculated in conjunction with the section of elevator
The volume of product, the cereal outer surface to the elevator surrounding that calculate before subtracting can be obtained by the volume of cereal.
To resolve convenient for volume, the three-dimensional coordinate of calculating is subjected to a rotation transformation, so that z-axis angle with horizontal plane is
0, calculate outer surface to elevator surrounding volume.It (can be according to the difference of the maximin of ordinate further according to cereal height
Be calculated) calculate the volume of this section of elevator, then subtract outer surface to elevator surrounding volume you can get it cereal reality
Border volume, projective transformation are that the coordinate by a kind of map projection's point that those skilled in the art are understood is transformed to another map projection
The process of the coordinate of point.Rotation transformation be those skilled in the art understood another figure is changed by a figure, changing
In the process, point all in original image all changes same direction around a fixed point, rotates the same angle.
Another aspect of the present invention discloses a kind of cereal flow detection device, as shown in Fig. 5-Fig. 9, including:First pair
Mesh camera, the second binocular camera and the raspberry pie communicated to connect respectively with the first binocular camera and the second binocular camera;First pair
The side of crop elevator is arranged in mesh camera, for acquiring the direct picture of grain stream in crop elevator, and by front elevation
As being sent to the raspberry pie;The side opposite with the first binocular camera of crop elevator is arranged in second binocular camera, uses
In the back side image of grain stream in acquisition crop elevator, and back side image is sent to the raspberry pie;Raspberry pie includes depositing
Reservoir and processor are stored with computer program on memory, and the step of the above method is realized when program is executed by the processor
Suddenly.Raspberry pie installation site can be adjusted voluntarily, subject to wiring installation.It needs to install in the SD card of raspberry pie
Ubuntu Mate operating system, and compile the installation library OpenCV and PCL.
Fig. 9 is perspective view, only shows that the first binocular camera of side, the second binocular camera are blocked no Faxian in the other side
Show,
In the description of the present invention, it should be noted that term " first ", is used for description purposes only " second ", and cannot
It is interpreted as indication or suggestion relative importance.
The present invention depends in house software and resolves to the image acquired in real time, therefore existing by the main of software realization
Process is illustrated:
Firstly, it is necessary to demarcate to two cameras, calibration content includes between the parameter and camera of two cameras itself
Parameter, if change camera between position, adjust camera focal length or replacement camera, then need to re-start calibration.
Secondly, the image of the front and back in two binocular camera acquisition elevators, carries out image judgement, judges whether
There is cereal, if without cereal, output flow 0;If there is cereal, there are two camera acquired images to generate depth
Figure.
Then, the volume of present frame is resolved according to depth map, logging timestamp, following repeatedly above step, acquisition
The image of the next frame and result for resolving volume and logging timestamp and previous frame is compared, if the result of present frame is greater than
Previous frame as a result, then continue acquire next frame image and calculate volume, if the result of present frame be less than previous frame as a result,
Then using the volume of previous frame as final volume, and the cumulative time is calculated according to the timestamp recorded before.
Finally, being output to serial ports or storage to database for the ratio of volume and cumulative time as flow, this time flow
Amount, which resolves, to be terminated, and next round resolving is carried out.
Cereal flow is measured by the above method and device, by binocular camera and software by automatic running, and in serial ports
Corresponding flow value is exported, without manually booting.
Preferably, the first binocular camera and the second binocular camera are arranged in the same horizontal position.It more acurrate can calculate
The volume of cereal.
In an alternative embodiment, the first binocular camera includes that first shell and setting are taken the photograph on intracorporal first left side of first shell
As head and the first right camera;First left camera is used to acquire the front LOOK LEFT image of grain stream in crop elevator;The
One right camera is used to acquire the front LOOK RIGHT image of grain stream in crop elevator;Second binocular camera includes second shell
With setting in the left camera of second shell intracorporal second and the second right camera;Second left camera is for acquiring crop elevator
The back side LOOK LEFT image of interior grain stream;Second right camera is used to acquire the back side LOOK RIGHT figure of grain stream in crop elevator
Picture.
In an alternative embodiment, first shell and second shell include:Shell, bolt, nut, camera support, window
Mouthful;Angle between left camera and right camera and horizontal plane is 20-50 degree, is camera optical axis, the visual angle of such camera
It is bigger than horizontal positioned, it can be observed that more cereal.
The cereal flow detection device further includes light compensating lamp.
Preferably, light compensating lamp is all installed in the two sides of installation binocular camera.Image preferably, which is acquired, for camera supplements light source,
Since illumination condition is poor inside elevator, light is weaker, therefore needs light filling measure.The position of light compensating lamp can be adjusted as needed
It is whole.
Transparent plastic baffle is installed on window, light is allowed to reach the camera lens of camera, and baffle through baffle
Cereal can be stopped to flow to device.
Hereinafter reference will be made to the drawings, and the present invention will be described in more detail.In various figures, identical element is using similar attached
Icon is remembered to indicate.For the sake of clarity, the various pieces in attached drawing are not necessarily to scale.
As long as in addition, the non-structure each other of technical characteristic involved in invention described below different embodiments
It can be combined with each other at conflict.
The present invention is directed to protect a kind of cereal flow detection method, including:Grain image is acquired, cereal in image is obtained
Volume;Judge whether the volume of cereal in image meets the first preset condition;It is preset when the volume of cereal in image meets first
When condition, start timing, continues to acquire grain image, obtain the volume of cereal in image;Judging the volume of cereal in image is
The second preset condition of no satisfaction;When the volume of cereal in image meets the second preset condition, the time is recorded, and obtain outflow knot
Fruit.The non-cpntact measurement for realizing cereal flow does not need to handle multi collect image, need to only carry out to a frame image
The volume that cereal can be obtained is resolved, and then obtains flow.Another aspect provides a kind of grain stream amount detection devices
It sets, which is characterized in that including:Shell is respectively arranged in the two sides of crop elevator;Bracket is placed in shell, and with the shell
Body connection;Binocular camera is connect with the bracket respectively;Light compensating lamp is installed on crop elevator two sides;And resolver.It should
Cereal flow detection device has installation and debugging simple, and precision is high, and to human body without any injury, cheap feature.
It should be understood that above-mentioned specific embodiment of the invention is used only for exemplary illustration or explains of the invention
Principle, but not to limit the present invention.Therefore, that is done without departing from the spirit and scope of the present invention is any
Modification, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.In addition, appended claims purport of the present invention
Covering the whole variations fallen into attached claim scope and boundary or this range and the equivalent form on boundary and is repairing
Change example.
Claims (10)
1. a kind of cereal flow detection method, which is characterized in that including:
Obtain the direct picture and back side image of grain stream in crop elevator;
The direct picture is matched, the front point cloud chart of the grain stream is generated;
The back side image is matched, the back side point cloud chart of the grain stream is generated;
Based on the front point cloud chart and back side point cloud chart, the volume of grain stream is obtained.
2. the method according to claim 1, wherein in the crop elevator grain stream direct picture, packet
It includes:Front LOOK LEFT image and front LOOK RIGHT image;
The back side image of grain stream in the crop elevator, including:Back side LOOK LEFT image and back side LOOK RIGHT image.
3. according to the method described in claim 2, it is characterized in that, described match the direct picture, described in generation
The front point cloud chart of grain stream includes:
The front LOOK LEFT image and front LOOK RIGHT image are calibrated;
Based on preset matching algorithm to each of front LOOK LEFT image pixel, in the front LOOK RIGHT figure
Corresponding pixel is found as in;
The parallax of each pixel in the front LOOK LEFT image and front LOOK RIGHT image is calculated based on preset algorithm;
Based on the parallax and parameter is prestored, calculates the depth coordinate of each pixel in the direct picture;
Based on the depth coordinate of each pixel in the direct picture, the abscissa and ordinate of each pixel are calculated;
Based on abscissa, ordinate and the depth coordinate of all pixels point in the direct picture, front point cloud chart is generated.
4. according to the method described in claim 3, it is characterized in that, by the parallax and prestoring parameter substitution formula (1) and counting
It calculates, obtains the depth coordinate of each pixel in the direct picture;
z1=(f1*T1)/d;Formula (1)
Wherein, z1Indicate the depth coordinate of each pixel in direct picture;
f1Indicate the focal length of acquisition direct picture camera;T1Indicate the baseline length of acquisition front camera;d1Indicate front left and right figure
As parallax.
5. according to the method described in claim 2, it is characterized in that, described match the back side image, described in generation
The back side point cloud chart of grain stream includes:
The back side LOOK LEFT image and back side LOOK RIGHT image are calibrated;
Based on preset matching algorithm to each of back side LOOK LEFT image pixel, in the back side LOOK RIGHT figure
Corresponding pixel is found as in;
The parallax of each pixel in the back side LOOK LEFT image and back side LOOK RIGHT image is calculated based on preset algorithm;
Based on the parallax and parameter is prestored, calculates the depth coordinate of each pixel in the back side image;
Based on the depth coordinate of each pixel in the back side image, the abscissa and ordinate of each pixel are calculated;
Based on abscissa, ordinate and the depth coordinate of all pixels point in the back side image, back side point cloud chart is generated.
6. according to the method described in claim 5, it is characterized in that, by the parallax and prestoring parameter substitution formula (1) and counting
It calculates, obtains the depth coordinate of each pixel in the back side LOOK LEFT image;
z2=(f2*T2)/d2;Formula (1)
Wherein, z2Indicate the depth coordinate of each pixel in back side image;
f2Indicate the focal length of acquisition back side image camera;T2Indicate the baseline length of acquisition back side camera;d2Indicate back side or so figure
As parallax.
7. method according to claim 1-6, which is characterized in that described to be based on the front point cloud chart and the back side
Point cloud chart, the volume for obtaining grain stream include:
The x of corresponding points, y-coordinate can be calculated according to z coordinate and projective transformation;
According to the three-dimensional coordinate of cereal surface point calculate cereal outer surface to elevator surrounding volume;
The volume that this section of elevator is calculated in conjunction with the section of elevator, the cereal outer surface calculated before subtracting to elevator four
The volume in week is to get the volume for arriving cereal.
8. a kind of cereal flow detection device, which is characterized in that including:First binocular camera, the second binocular camera and respectively with
The raspberry pie of first binocular camera and the communication connection of the second binocular camera;
The side of crop elevator is arranged in first binocular camera, for acquiring the front elevation of grain stream in crop elevator
Picture, and the direct picture is sent to the raspberry pie;
The side opposite with first binocular camera of crop elevator is arranged in second binocular camera, for acquiring paddy
The back side image of grain stream in object elevator, and the back side image is sent to the raspberry pie;
The raspberry pie includes memory and processor, and computer program is stored on the memory, and described program is described
The step of processor realizes any one of claims 1 to 7 the method when executing.
9. cereal flow detection device according to claim 8, which is characterized in that
First binocular camera includes that first shell and setting are right in the left camera of the first shell intracorporal first and first
Camera;
The first left camera is used to acquire the front LOOK LEFT image of grain stream in crop elevator;
The first right camera is used to acquire the front LOOK RIGHT image of grain stream in crop elevator;
Second binocular camera includes that second shell and setting are right in the left camera of the second shell intracorporal second and second
Camera;
The second left camera is used to acquire the back side LOOK LEFT image of grain stream in crop elevator;
The second right camera is used to acquire the back side LOOK RIGHT image of grain stream in crop elevator.
10. cereal flow detection device according to claim 8, which is characterized in that
The first shell and second shell include:Shell, bolt, nut, camera support, window;
Angle between the left camera and right camera and horizontal plane is 20-50 degree;
Further include:Light compensating lamp.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810720353.7A CN108921842A (en) | 2018-07-02 | 2018-07-02 | A kind of cereal flow detection method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810720353.7A CN108921842A (en) | 2018-07-02 | 2018-07-02 | A kind of cereal flow detection method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108921842A true CN108921842A (en) | 2018-11-30 |
Family
ID=64425380
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810720353.7A Pending CN108921842A (en) | 2018-07-02 | 2018-07-02 | A kind of cereal flow detection method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108921842A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110595356A (en) * | 2019-09-10 | 2019-12-20 | 湖南海森格诺信息技术有限公司 | Method for measuring solid volume in artificial storage environment |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101720047A (en) * | 2009-11-03 | 2010-06-02 | 上海大学 | Method for acquiring range image by stereo matching of multi-aperture photographing based on color segmentation |
CN102721367A (en) * | 2012-07-02 | 2012-10-10 | 吉林省粮油科学研究设计院 | Method for measuring volume of large irregular bulk grain pile based on dynamic three-dimensional laser scanning |
CN103868460A (en) * | 2014-03-13 | 2014-06-18 | 桂林电子科技大学 | Parallax optimization algorithm-based binocular stereo vision automatic measurement method |
US9257022B2 (en) * | 2012-06-14 | 2016-02-09 | Immersion Corporation | Haptic effect conversion system using granular synthesis |
CN105931238A (en) * | 2016-04-19 | 2016-09-07 | 北京理工大学 | Method and system of granary grain storage volume measurement |
CN106508258A (en) * | 2016-10-11 | 2017-03-22 | 北京农业智能装备技术研究中心 | Photoelectric grain yield metering device |
CN107873233A (en) * | 2017-10-23 | 2018-04-06 | 江苏大学 | A kind of array infrared light electric-type apparatus for measuring cereal flow and method |
CN108200804A (en) * | 2016-12-19 | 2018-06-26 | 上海交通大学 | Visual sensing cereal flow on-line measuring device and detection method |
CN108596948A (en) * | 2018-03-16 | 2018-09-28 | 中国科学院自动化研究所 | The method and device of human body head posture is identified based on depth camera |
-
2018
- 2018-07-02 CN CN201810720353.7A patent/CN108921842A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101720047A (en) * | 2009-11-03 | 2010-06-02 | 上海大学 | Method for acquiring range image by stereo matching of multi-aperture photographing based on color segmentation |
US9257022B2 (en) * | 2012-06-14 | 2016-02-09 | Immersion Corporation | Haptic effect conversion system using granular synthesis |
CN102721367A (en) * | 2012-07-02 | 2012-10-10 | 吉林省粮油科学研究设计院 | Method for measuring volume of large irregular bulk grain pile based on dynamic three-dimensional laser scanning |
CN103868460A (en) * | 2014-03-13 | 2014-06-18 | 桂林电子科技大学 | Parallax optimization algorithm-based binocular stereo vision automatic measurement method |
CN105931238A (en) * | 2016-04-19 | 2016-09-07 | 北京理工大学 | Method and system of granary grain storage volume measurement |
CN106508258A (en) * | 2016-10-11 | 2017-03-22 | 北京农业智能装备技术研究中心 | Photoelectric grain yield metering device |
CN108200804A (en) * | 2016-12-19 | 2018-06-26 | 上海交通大学 | Visual sensing cereal flow on-line measuring device and detection method |
CN107873233A (en) * | 2017-10-23 | 2018-04-06 | 江苏大学 | A kind of array infrared light electric-type apparatus for measuring cereal flow and method |
CN108596948A (en) * | 2018-03-16 | 2018-09-28 | 中国科学院自动化研究所 | The method and device of human body head posture is identified based on depth camera |
Non-Patent Citations (2)
Title |
---|
张力等: "《陆地机器人》", 31 March 2017 * |
谢广明等: ""自主仿生机器鱼"", 《哈尔滨工业大学出版社》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110595356A (en) * | 2019-09-10 | 2019-12-20 | 湖南海森格诺信息技术有限公司 | Method for measuring solid volume in artificial storage environment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11477374B2 (en) | Three dimensional image capture system for imaging building facades using a digital camera, a near-infrared camera, and laser range finder | |
US10151839B2 (en) | Systems and methods for determining crop yields with high resolution geo-referenced sensors | |
US9329683B2 (en) | Method for detecting point of gaze and device for detecting point of gaze | |
EP1672314B1 (en) | Method for preparing a stereo image and corresponding system for preparing three-dimensional data | |
JP2007171033A (en) | Indirect measuring method and system of leaf area index | |
WO2017011595A1 (en) | Modular systems and methods for determining crop yields with high resolution geo-referenced sensors | |
CN108921906A (en) | calibration method and measuring tool | |
CN108650447A (en) | Imaging sensor, depth data measurement head and measuring system | |
CN107884160A (en) | Virtual image photoelectric measuring instrument | |
CN102084228A (en) | Device and method for the space-colorimetric measurement of a three-dimensional object | |
CN108700408A (en) | Three-dimensional shape data and texture information generate system, shooting control program and three-dimensional shape data and texture information generation method | |
CN104568963A (en) | Online three-dimensional detection device based on RGB structured light | |
CN108020175A (en) | A kind of more optical grating projection binocular vision tongue body surface three dimension entirety imaging methods | |
CN107860337A (en) | Structural light three-dimensional method for reconstructing and device based on array camera | |
CN111968074A (en) | Method for detecting and harvesting lodging crops of harvester by combining binocular camera and IMU | |
CN112581545A (en) | Multi-mode heat source recognition and three-dimensional space positioning system, method and storage medium | |
CN108921842A (en) | A kind of cereal flow detection method and device | |
CN107421503B (en) | Single-detector three-linear-array three-dimensional mapping imaging method and system | |
CN109781068A (en) | The vision measurement system ground simulation assessment system and method for space-oriented application | |
CN108550154A (en) | A kind of method of accurately measuring karst earth's surface bare rock accounting | |
JP2006320240A (en) | Method for measuring crop information by remote sensing | |
CN107798668A (en) | The method and system of unmanned plane imaging EO-1 hyperion geometric correction based on RGB images | |
CN108243337A (en) | A kind of imaging sensor dynamic transfer function test phase alignment means | |
CN107636533A (en) | Camera device, endoscope apparatus and image capture method | |
CN109087341B (en) | Fusion method of close-range hyperspectral camera and ranging sensor |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181130 |