CN103884274A - Rice grain type detection system - Google Patents
Rice grain type detection system Download PDFInfo
- Publication number
- CN103884274A CN103884274A CN201310608232.0A CN201310608232A CN103884274A CN 103884274 A CN103884274 A CN 103884274A CN 201310608232 A CN201310608232 A CN 201310608232A CN 103884274 A CN103884274 A CN 103884274A
- Authority
- CN
- China
- Prior art keywords
- rice
- grain
- circumscribed rectangle
- rectangle
- detection system
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention discloses a rice grain type detection system, and relates to a rice grain type detection system based on regional span searching. Grain length and grain width of rice are found by the system via searching of a rice grain regional span, and finally a rice type is obtained via ratio of the grain length to the grain width. An experiment result indicates that a measurement result of the rice grain type by the system is basically consistent with an artificial measurement result with a relative error of 1.73% together with a characteristic of invariance in rice grain rotation. The rice grain type detection system comprises four steps of positioning an initial rotation circumscribed rectangle, clockwise rotation, judging a minimum span value and calculating length-width ratio. The realization mode of the aforementioned steps is that the step of positioning the initial rotation circumscribed rectangle is executed according to the acquired rice image data. Then the step of clockwise rotation is executed with the initial circumscribed rectangle as a starting point. When the step of clockwise rotation is executed, the step of judging the minimum span value is executed. When the minimum span value is reached via rotation, the step of calculating length-width ratio is executed.
Description
Technical field
The present invention relates to rice granule type detection field, more specifically a kind of rice granule type detection system based on region span searching.
Background technology
Rice granule type is a key character weighing Rice Outer-quality.In rice national standard, it refers to the ratio that rice grain length and grain are wide.At present, the method for utilizing automated system to detect rice granule type mainly contains Ellipse Matching mensuration and polar coordinate measuring method.
Ellipse Matching mensuration, by grain of rice sub-elliptical, by the computing of grain of rice region square being obtained to the major and minor axis of sub-elliptical, then represents the grain type of the grain of rice with ratio of semi-minor axis length, therefore measurement result is the approximate value of grain of rice grain type.
Polar coordinate measuring method is take the type heart in grain of rice region as polar coordinates initial point, initial point to the distance of grain of rice edges of regions point is polar coordinates radius, find out the direction of maximum polar coordinates radius by the rotary search of polar coordinates radius as the grain length direction of the grain of rice, what direction was vertical therewith is grain cross direction, because grain of rice region is strictly not symmetrical with the type heart, therefore measurement result can not strictly reflect rice granule type.
Summary of the invention
For addressing the above problem, the invention discloses a kind of rice granule type detection system, the present invention is a kind of rice granule type detection system based on region span searching.This system by the search of grain of rice region span is found out rice grain length and grain wide, finally obtain rice type by the wide ratio of grain length and grain.Experimental result shows: measurement result and the manual measurement result of this invention to rice granule type is basically identical, and relative error is 1.73%, and has the feature of grain of rice rotational invariance.
The present invention takes following technical scheme to realize: a kind of rice granule type detection system, comprises the circumscribed rectangle of the initial rotation in location, turns clockwise, judges span minimum value and calculate four steps of length breadth ratio; The implementation of above-mentioned steps is carried out the circumscribed rectangle step of the initial rotation in location according to the rice view data gathering, then carry out dextrorotation goes to step take initial circumscribed rectangle as starting point, carry out when dextrorotation goes to step and judge span minimum value step, in the time rotating to span minimum value, carry out and calculate length breadth ratio step.
Realization of the present invention also comprises following technical scheme:
The above-mentioned implementation that judges span minimum value step is: take the most left coordinate points in grain of rice region and the rightest coordinate points as rectangle, angle point is done to one and search for the pixel on the each limit of this rectangle with the circumscribed rectangle of initial rotation inclined rectangular into θ angle, belong to the pixel in grain of rice region if having, by this limit outside the grain of rice region translation until tangent with grain of rice edges of regions, and being extended to the circumscribed rectangle in grain of rice region, this circumscribed solid-line rectangle is just decided to be the circumscribed rectangle behind initial circumscribed rectangle rotation θ angle.
Along with the variation of rotation angle θ, length and width span also can correspondingly change, and the minimum value of getting length and width span is calculated the length breadth ratio of the grain of rice.
Advantage of the present invention and beneficial effect are embodied in the following aspects:
1. use measurement result and the manual measurement result of the present invention to rice granule type basically identical, relative error is 1.73%.
2. the present invention has the feature of grain of rice rotational invariance.
Accompanying drawing explanation
Fig. 1 is execution step schematic diagram of the present invention;
Fig. 2 is the rotary course schematic diagram of the circumscribed rectangle in grain of rice region.
Embodiment
Below in conjunction with Figure of description 1 and explanation accompanying drawing 2, enforcement of the present invention is further described:
A kind of rice granule type detection system, comprises the circumscribed rectangle of the initial rotation in location, turns clockwise, judges span minimum value and calculates four steps of length breadth ratio; The implementation of above-mentioned steps is carried out the circumscribed rectangle step of the initial rotation in location according to the rice view data gathering, then carry out dextrorotation goes to step take initial circumscribed rectangle as starting point, carry out when dextrorotation goes to step and carry out and judge span minimum value step, in the time rotating to span minimum value, carry out and calculate length breadth ratio step.
Due to the circumscribed rectangle in region and edges of regions tangent, therefore can be represented by the width of the circumscribed rectangle in region and length the span in grain of rice region.At present, the circumscribed rectangle in region is mainly used in zone location, and each limit of the circumscribed rectangle in region is conventionally parallel with image coordinate axle.The length breadth ratio of this circumscribed rectangle is difficult to accurately reflect the grain type with the grain of rice of putting arbitrarily direction, only have the circumscribed rectangle in this location of the grain of rice is suitably rotated according to the placement orientation of the grain of rice, when the length and width direction that makes circumscribed rectangle is consistent with the length and width direction of the grain of rice respectively, the length and width Bizet of the circumscribed rectangle of the grain of rice likely reflects the grain type of the grain of rice.The spinning solution of the circumscribed rectangle of the grain of rice can be illustrated by Figure of description 2.If coordinate is image coordinate in figure, fill area R represents grain of rice region, the each limit respectively grain of rice region circumscribed solid-line rectangle parallel with image coordinate axle is the locating rectangle in grain of rice region, and the most left coordinate of grain of rice edges of regions is (xL, yL), the rightest coordinate is (xR, yR), going up coordinate is most (xT, yT), descending most coordinate is (xB, yB).This is located to circumscribed rectangle and be decided to be the circumscribed rectangle of initial rotation, the circumscribed rectangle in grain of rice region rotates in the direction of the clock, and spinning solution is:
Take the most left coordinate points (xL, yL) in grain of rice region and the rightest coordinate points (xR, yR) as rectangle, angle point is made to an inclined rectangular into θ angle with the circumscribed rectangle of initial rotation, as shown in the dashed rectangle that Figure of description 2 tilts.Search for the pixel on the each limit of this rectangle, belong to the pixel of R if having, the translation outside grain of rice region of this limit until tangent with grain of rice edges of regions, and is extended to the circumscribed rectangle in grain of rice region, as shown in the circumscribed solid-line rectangle in grain of rice region of Figure of description 2 medium dips.This circumscribed solid-line rectangle is just decided to be the circumscribed rectangle behind initial circumscribed rectangle rotation θ angle.In this rectangle, the two rectangular edges equations into θ angle with image y direction of principal axis are:
y=-xctgθ+xLctgθ+yL-kθy=-xctgθ+xRctgθ+yR+lθ(1)
Wherein, k θ, l θ are translation pixel count.Spacing (the being span) D1 on these both sides is:
D1=|(xR-xL)ctgθ+(yR-yL)+kθ+lθ|1+ctg2θ(2)
In rectangle, the straight-line equation on the other both sides vertical with upper both sides is:
y?=?xtgθ-xLtgθ+yL+mθy?=?xtgθ-xRtgθ+yR-nθ(3)
Wherein, m θ, n θ are translation pixel count.Spacing (the being span) D2 on these both sides is:
D2=|?(xR-xL)tgθ+yL-yR+mθ+nθ|1+tg2θ(4)
Along with the variation of rotation angle θ, k θ, l θ, m θ, n θ, D1, D2 respective change, get the wide imaging value W (=min (D1min of grain that the minimum value in D1 and D2 is the grain of rice, D2min)), distance values on the other direction that cross direction is vertical is therewith the grain length imaging value L of the grain of rice, and the length breadth ratio of the grain of rice is just L/W.
Utilize technical solutions according to the invention, or those skilled in the art being under the inspiration of technical solution of the present invention, designs similar technical scheme, and reaching above-mentioned technique effect, is all to fall into protection scope of the present invention.
Claims (5)
1. a rice granule type detection system, comprises the circumscribed rectangle of the initial rotation in location, turns clockwise, judges span minimum value and calculates four steps of length breadth ratio; The implementation of above-mentioned steps is carried out the circumscribed rectangle step of the initial rotation in described location according to the rice view data gathering, then carrying out described dextrorotation take initial circumscribed rectangle as starting point goes to step, carry out when described dextrorotation goes to step and carry out described judgement span minimum value step, in the time rotating to span minimum value, carry out described calculating length breadth ratio step.
2. a kind of rice granule type detection system according to claim 1, it is characterized in that: the implementation of the circumscribed rectangle step of the initial rotation in described location is: image coordinate, fill area R represents grain of rice region, the each limit respectively grain of rice region circumscribed solid-line rectangle parallel with image coordinate axle is the locating rectangle in grain of rice region, the most left coordinate of grain of rice edges of regions is (xL, yL), the rightest coordinate is (xR, yR), going up most coordinate is (xT, yT), and descending coordinate is most (xB, yB), this is located to circumscribed rectangle and be decided to be the circumscribed rectangle of initial rotation.
3. a kind of rice granule type detection system according to claim 1, it is characterized in that: described dextrorotation goes to step to such an extent that implementation is: with the most left coordinate points (xL in grain of rice region, and the rightest coordinate points (xR yL), yR) be that rectangle is made an inclined rectangular into θ angle with the circumscribed rectangle of initial rotation to angle point, search for the pixel on the each limit of this rectangle, belong to the pixel of R if having, by this limit outside the grain of rice region translation until tangent with grain of rice edges of regions, and be extended to the circumscribed rectangle in grain of rice region, this circumscribed solid-line rectangle is just decided to be the circumscribed rectangle behind initial circumscribed rectangle rotation θ angle.
4. a kind of rice granule type detection system according to claim 1, it is characterized in that: when the described dextrorotation of described execution goes to step, carrying out the described implementation that judges span minimum value step is: along with the variation of rotation angle θ, k θ, l θ, m θ, n θ, D1, D2 respective change, get the wide imaging value W (=min (D1min of grain that the minimum value in D1 and D2 is the grain of rice, D2min) distance values on the other direction that), cross direction is vertical is therewith the grain length imaging value L of the grain of rice.
5. a kind of rice granule type detection system according to claim 1, is characterized in that: the implementation of described calculating length breadth ratio step is: the length breadth ratio of the grain of rice is L/W.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310608232.0A CN103884274A (en) | 2013-11-25 | 2013-11-25 | Rice grain type detection system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310608232.0A CN103884274A (en) | 2013-11-25 | 2013-11-25 | Rice grain type detection system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN103884274A true CN103884274A (en) | 2014-06-25 |
Family
ID=50953303
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310608232.0A Pending CN103884274A (en) | 2013-11-25 | 2013-11-25 | Rice grain type detection system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103884274A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105954284A (en) * | 2016-05-06 | 2016-09-21 | 安徽贝通智能科技有限公司 | Rice quality detection method |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5590060A (en) * | 1992-03-20 | 1996-12-31 | Metronics, Inc. | Apparatus and method for an object measurement system |
CN101109743A (en) * | 2007-09-10 | 2008-01-23 | 肯特大学 | Portable cereal analyzer based on digital picture processing technique |
CN101275824A (en) * | 2008-05-16 | 2008-10-01 | 中国农业大学 | Method for detecting rice granule type |
WO2013168457A1 (en) * | 2012-05-07 | 2013-11-14 | 株式会社ニコン | Surface position measurement device, surface position measurement method, exposure device, and device production method |
-
2013
- 2013-11-25 CN CN201310608232.0A patent/CN103884274A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5590060A (en) * | 1992-03-20 | 1996-12-31 | Metronics, Inc. | Apparatus and method for an object measurement system |
CN101109743A (en) * | 2007-09-10 | 2008-01-23 | 肯特大学 | Portable cereal analyzer based on digital picture processing technique |
CN101275824A (en) * | 2008-05-16 | 2008-10-01 | 中国农业大学 | Method for detecting rice granule type |
WO2013168457A1 (en) * | 2012-05-07 | 2013-11-14 | 株式会社ニコン | Surface position measurement device, surface position measurement method, exposure device, and device production method |
Non-Patent Citations (2)
Title |
---|
陈鲤江等: "基于区域跨度搜索的大米粒型检测方法", 《光电子• 激光》 * |
陈鲤江等: "基于区域跨度搜索的大米粒型检测方法", 《光电子• 激光》, vol. 18, no. 1, 31 January 2007 (2007-01-31) * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105954284A (en) * | 2016-05-06 | 2016-09-21 | 安徽贝通智能科技有限公司 | Rice quality detection method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR102037820B1 (en) | High precision map data processing methods, devices, storage media and devices | |
CN112966587B (en) | Training method of target detection model, target detection method and related equipment | |
US20200081125A1 (en) | Method and robot of mapping | |
CN102663725B (en) | A visible light and SAR image registration method based on linear features and control points | |
CN102620742B (en) | Steering identification method based on road form point section angle accumulation and road relative position | |
CN104331689B (en) | The recognition methods of a kind of cooperation mark and how intelligent individual identity and pose | |
CN101718551B (en) | Flexible cable movement measuring method and measuring device | |
CN104809689A (en) | Building point cloud model and base map aligned method based on outline | |
CN105809702A (en) | Improved position and orientation estimation method based on Tsai algorism | |
CN104197934B (en) | A kind of localization method based on earth magnetism, apparatus and system | |
CN105912971A (en) | Regular graphic code array for AGV navigation and code reading method thereof | |
CN103353758A (en) | Indoor robot navigation device and navigation technology thereof | |
CN107636679A (en) | A kind of obstacle detection method and device | |
CN110595478A (en) | Robot full-coverage path planning method, device and equipment based on off-line map | |
CN109655805A (en) | A kind of laser radar positioning method being overlapped length estimate based on scan lines | |
CN103644904A (en) | Visual navigation method based on SIFT (scale invariant feature transform) algorithm | |
CN102938147A (en) | Low-altitude unmanned aerial vehicle vision positioning method based on rapid robust feature | |
CN104360743A (en) | Body posture data acquisition method and system, and data processing device | |
CN104913775B (en) | Measurement method, unmanned plane localization method and the device of unmanned plane distance away the ground | |
CN105116886A (en) | Robot autonomous walking method | |
CN109856649A (en) | Error analysis method, device and the storage medium of navigator fix | |
CN106682674B (en) | A kind of pointer instrument automatic reading method based on intelligent inspection robot | |
CN103871072A (en) | Method for automatic extraction of orthoimage embedding line based on projection digital elevation models | |
CN103768792A (en) | Partitioning and recognition method applied to wide-range video space object positioning system | |
CN106652028A (en) | Environment three-dimensional mapping method and apparatus |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140625 |