CN205538564U - Broken on -line monitoring system of cereal in combine grain tank - Google Patents
Broken on -line monitoring system of cereal in combine grain tank Download PDFInfo
- Publication number
- CN205538564U CN205538564U CN201620233431.7U CN201620233431U CN205538564U CN 205538564 U CN205538564 U CN 205538564U CN 201620233431 U CN201620233431 U CN 201620233431U CN 205538564 U CN205538564 U CN 205538564U
- Authority
- CN
- China
- Prior art keywords
- corn
- image capture
- capture unit
- cylinder
- conveyer belt
- 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.)
- Active
Links
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The utility model provides a broken on -line monitoring system of cereal in combine grain tank, including cereal transport mechanism, image capture unit and treater ARM, cereal transport mechanism includes cylinder, oblique conveying belt and vibration exciter, the cylinder is located the granary exit, and oblique conveying belt is located the below of cylinder, and the vibration exciter is installed to oblique conveying belt's bottom, the image capture unit is located the oblique conveying belt top, and in perpendicular to oblique conveying belt's direction, the image capture unit includes CCD camera, camera lens, light source and data collection card in proper order, the image capture unit is used for carrying out periodic absorbing to cereal to to absorb the picture conveys to treater ARM, cylinder, CCD camera, light source all are connected with treater ARM, treater ARM is used for opening and closing of index drum, CCD camera and light source to analysis processes image capture unit conveying absorb the picture. On -line monitoring cereal easy operation is convenient in real time to utilize this device, and the rate of accuracy is high.
Description
Technical field
This utility model belongs to combined harvester service behaviour automated watch-keeping facility field, especially relates to a kind of united reaper grain
In case, corn crushes on-line monitoring system, for on-line monitoring corn percentage of damage.
Background technology
In current combined harvester running parameter and transaction capabilities study on monitoring, it it is the most only the monitoring for single homework parameter
Or the research of forecast model, not according to the current work parameter value monitored, associated components is carried out feedback control and to many works
The research that industry Parameter fusion controls is the most relatively fewer.For the monitoring of combine operation performance, do not consider a weight
The parameter percentage of broken grain wanted.
During combine operation, when occurring that the running parameters such as excesssive gap between cutting table auger and base plate, rotating speed be too fast are abnormal
Situation, can cause the percentage of damage of corn to increase, and according to National agricultural standard NY/T 498-2002, have specified below:
After the seed samples removing magazine is mixed, it is taken out 5 parts of sample, every part of 100g by quartering, picks out therein
Broken seed also weighs, and calculates percentage of damage according to the following formula, takes the meansigma methods of 5 parts of sample broke rates.
Z (%)=m1/m2*100
In formula, Z percentage of damage, %；
Broken quality grain, g in m1 sample；
M2 sample quality, g.
Under the conditions of general job, the harvest operation percentage of damage of Oryza sativa L. full-feeding combine harvester should be less than 2.5%.Percentage of damage is higher
Corn, on the one hand when selling at grain distribution station, price is low, can bring direct economic loss to peasant, and on the other hand, peasant is with breaking
During the relatively low corn breeding of broken rate, germination percentage is relatively low, and severe patient can affect the plantation in next season, and research shows, in results process
In to reduce corn percentage of damage be one of important channel solving the problems referred to above as far as possible.
At present, the corn percentage of damage detection of China generally rests on to be identified by gross visualization, and this work is sentenced mainly by subjectivity
Disconnected, loaded down with trivial details repetitive rate is high, and error is big, and efficiency is low, the longest, adds the biggest burden for agricultural research staff, and result is but
Unsatisfactory, have a strong impact on the working performance of united reaper and current grain can not be obtained during combine operation in real time
Grain in case crushes situation, and united reaper human pilot also cannot crush situation according to current corn and take corresponding measure.
Utility model content
Can not obtain the grain in current tanker in real time for existence in prior art and crush the deficiencies such as situation, this utility model provides
In a kind of united reaper tanker, corn crushes on-line monitoring system, utilizes this method monitoring corn simple to operation and accurate
Really rate is high.
This utility model realizes above-mentioned technical purpose by techniques below means.
In a kind of united reaper tanker, corn crushes on-line monitoring system, including corn connecting gear, image capture unit and place
Reason device ARM；Described corn connecting gear includes cylinder, tilts conveyer belt and vibrator；Described cylinder is positioned at silo exit,
Described inclination conveyer belt is positioned at the lower section of cylinder, and the bottom tilting conveyer belt is provided with vibrator；Described corn connecting gear is used for
What tanker corn was sent to image capture unit takes the photograph district；
Described image capture unit is positioned at above inclination conveyer belt, and on the direction being perpendicular to conveyer belt；Described image is taken the photograph
Take unit and include ccd video camera, camera lens, light source and data collecting card successively；Image capture unit is for carrying out week to corn
The picked-up of phase property, and picked-up picture is sent to processor ARM；Described cylinder, ccd video camera, light source are all and processor
ARM connects；Described processor ARM is for controlling the opening and closing of cylinder, ccd video camera and light source, and analyzes and processes
The picked-up picture that image capture unit transmits.
Further, described vibrator is positioned at the centre position of conveyer belt；Described image capture unit is positioned at inclination conveyer belt
Centre position.
In above-mentioned united reaper tanker, corn crushes on-line monitoring method, comprises the steps:
S1: processor ARM controls corn connecting gear takes the photograph district by what sampling corn was periodically sent to graphics processing unit,
Image capture unit periodically gathers grain image；
S2: grain image is sent to image pre-processing unit by image capture unit, and image pre-processing unit is to Image semantic classification；
Pretreated image is transferred to processor ARM by S3: image pre-processing unit, and processor ARM is further to image
Analyze, process, subsequently output corn percentage of damage；
S4: the numerical result of corn percentage of damage is fed back to driver's cabin, driver carries out associative operation.
Further, image pre-processing unit described in step S2 is as follows to Image semantic classification process: gray level image processes, and leads to
Cross medium filtering and it is removed noise, form denoising view data；Extracted in denoising image by maximum variance between clusters
Appropriate threshold, the corn that will detect is separated from background, forms the monitoring bianry image removing background；Utilize morphology
The tiny burr of corn in bianry image is removed in operation makes profile polish, and fills up the cavity that grain is medium and small, is formed smooth complete
Corn bianry image, calculate contained corn total quantity in image.
Further, processor ARM described in step S3 is as follows to image processing process: each to pretreated image zooming-out
The input of the size of cereal-granules, shape, colourity three class morphological feature parameter is set up at neural network classifier, such classify
Device sub-argument from various cereal-granules goes out broken grains granular size, shape, colourity three class morphological feature parameter；Separate
The size of broken grains granule, shape, the input of colourity three class morphological feature parameter set up feature on the basis of fuzzy clustering
Fusion device, is normalized each class multidimensional morphological feature parameter by this feature fusion device, obtains one-dimensional characteristic vector；
The one-dimensional characteristic vector of the described cereal-granules after normalization is sent into neural network classifier, identifies crushing in cereal-granules
Cereal-granules, and be calculated broken grains account for corn sum percentage ratio, i.e. percentage of damage.
Further, described size characteristic parameter includes area, girth, equivalent diameter, major axis, short axle；Described shape facility
Parameter includes circle rate, eccentricity；Described chromaticity parameter includes based on HSV target area color histogram, based on probability window
Regional aim Dominant Color Features extract.
Further, described neural network classifier is specific as follows: use probabilistic neural network PNN method to set up forecast model,
Use three-layer neural network structure, including 1 input layer, 1 hidden layer node and 1 output layer；By corn sample size,
Shape, colourity three aspect characteristic ginseng value as the input value of neural network classifier, after training repeatedly and test,
Carry out probabilistic neural network training, obtain the state of corn, obtain forecast model；Corn is the most broken as output valve；Neural
The design parameter of network is: the learning efficiency=0.001；Frequency of training=2000；Target minimum error=0.0001；Factor of momentum=0.9；
The maximum frequency of failure=1000.
The beneficial effects of the utility model:
In a kind of united reaper tanker described in the utility model, corn crushes on-line monitoring system and method, utilizes image capture
Unit is to being sent to image pre-processing unit and processor ARM analyzing and processing after sampling corn picked-up, by selected characteristic parameter
With set up neural network classifier etc. and obtain the quantity of broken grains, and then obtain the percentage of damage of broken grains granule in tanker, logical
Cross and analyze the percentage of damage height obtained, it is possible to the duty of united reaper is made and feeds back timely；Compare tradition and utilize people
The method that work, drying are weighed obtains percentage of damage, and this utility model method is simple to operation, quick and accuracy rate is high.
Accompanying drawing explanation
Fig. 1 is that in united reaper tanker described in the utility model, corn crushes on-line monitoring system flow chart.
Fig. 2 is that in united reaper tanker, corn crushes on-line monitoring system structural representation.
Fig. 3 is the neural network structure figure of embodiment of the present utility model.
Description of reference numerals is as follows:
1-CCD video camera, 2-camera lens, 3-light source, 4-vibrator, 5-tilts conveyer belt, 6-cylinder.
Detailed description of the invention
Below in conjunction with the accompanying drawings and this utility model is further described by specific embodiment, but protection domain of the present utility model
It is not limited to this.
As depicted in figs. 1 and 2, in a kind of united reaper tanker, corn crushes on-line monitoring system, including corn connecting gear,
Image capture unit and processor ARM；This package unit system is positioned in associating corn mowing machine tanker, can be by monitoring grain
In case, the percentage of damage of corn judges the working condition of united reaper.
Corn connecting gear includes cylinder 6, tilts conveyer belt 5 and vibrator 4；Cylinder 6 is positioned at silo exit, described inclination
Conveyer belt 5 entrance is positioned at the lower section of cylinder 6, and the bottom middle position tilting conveyer belt 5 is provided with vibrator 4；Described paddy
Thing connecting gear takes the photograph district for what tanker corn was sent to image capture unit.Described image capture unit is positioned at inclination conveyer belt
Above 5 centre positions and perpendicular with the direction tilting conveyer belt 5；Described image capture unit includes ccd video camera successively
1, camera lens 2, light source 3 and data collecting card；Image capture unit is for periodically absorbing corn, and will absorb
Picture is sent to processor ARM7；Described cylinder 6, ccd video camera 1, light source 3 are all connected with processor ARM；Described
Processor ARM is for controlling cylinder 6, ccd video camera 1 and the opening and closing of light source 3, and analyzes and processes image capture
The picked-up picture that unit transmits.
Utilize the detection method of this on-line monitoring system, comprise the steps:
S1: processor ARM controls corn connecting gear takes the photograph district by what sampling corn was periodically sent to graphics processing unit,
4 i.e. every 15 seconds cycles once per minute are used to take the photograph district by what corn was sent to image capture unit by corn connecting gear,
Sampling 200 every time；Carry out in airtight space, it is to avoid the inevitably factor impact such as dust, wind speed；Image capture
Unit periodically gathers grain image.
Grain image is sent to image pre-processing unit by S2: image capture unit, and image is carried out as follows by image pre-processing unit
Pretreatment；Gray level image processes, and by medium filtering, it is removed noise, forms denoising view data；Pass through maximum kind
Between variance method extract the appropriate threshold in denoising image, the corn that will detect is separated from background, is formed and removes background
Monitoring bianry image；Utilizing morphological operation to remove the tiny burr of corn in bianry image makes profile polish, and fills up paddy
The cavity that grain is medium and small, forms smooth complete corn bianry image, calculates contained corn total quantity in image.
Pretreated image is transferred to processor ARM by S3: image pre-processing unit, and processor ARM is further to image
Analyze, process: the size of cereal-granules each to pretreated image zooming-out, shape, colourity three class morphological feature parameter are defeated
Enter to set up at neural network classifier, by such grader sub-argument from various cereal-granules go out broken grains granular size, shape,
Colourity three class morphological feature parameter；Concrete: size characteristic parameter includes area, girth, equivalent diameter, major axis, short axle；
Parameters for shape characteristic includes circle rate, eccentricity；Chromaticity parameter includes based on HSV target area color histogram, based on generally
The regional aim Dominant Color Features of rate window is extracted.The size of broken grains granule separated, shape, colourity three class morphology are special
Levy parameter input and set up the Feature Fusion device on the basis of fuzzy clustering, by this feature fusion device by each class multidimensional morphological feature
Parameter is normalized, and obtains one-dimensional characteristic vector；The one-dimensional characteristic vector of the described cereal-granules after normalization is sent into
Neural network classifier, identifies the broken grains granule in cereal-granules, and is calculated broken grains and accounts for the hundred of corn sum
Proportion by subtraction, i.e. percentage of damage.
S4: subsequently percentage of damage numerical value is transmitted driving cabin of combined-harvester, feed back to driver and carry out associative operation.
Wherein, described calculating corn percentage of damage, it is characterised in that including: calculate corn total quantity N contained in image,
Obtained quantity n of broken grains by described percentage of damage on-line monitoring system, corn percentage of damage: Z (%)=n/m*100 can be obtained；
In formula, Z percentage of damage, %；
In n sample, broken seed number, individual；
N grain sample total quantity, individual.
Neural network classifier, particularly as follows: use probabilistic neural network PNN method to set up forecast model, uses three-layer neural network
Structure, including 1 input layer, 1 hidden layer node and 1 output layer；By corn sample size, shape, colourity tripartite
The characteristic ginseng value in face, as the input value of neural network classifier, after training repeatedly and test, carries out probabilistic neural net
Network training, obtains the state of corn, obtains forecast model；Corn is the most broken as output valve；The design parameter of neutral net
For: the learning efficiency=0.001；Frequency of training=2000；Target minimum error=0.0001；Factor of momentum=0.9；Maximum failure time
Number=1000.This foundation grader on neural net base includes a feedback procedure, this feedback procedure be to classification out can
Doubt target and identify false target carry out refining, classify, complementary features parameter, and set up corresponding mathematical model, to nerve net
Network is trained, and neutral net learns and remember those characteristic parameter entrance model databases refining, classifying, supplement automatically,
Return again to grader based on neutral net and carry out cereal-granules classification.
Described embodiment is that of the present utility model preferred embodiment but this utility model is not limited to above-mentioned embodiment,
In the case of flesh and blood of the present utility model, those skilled in the art can make any conspicuously improved,
Replace or modification belongs to protection domain of the present utility model.
Claims (2)
1. in a united reaper tanker, corn crushes on-line monitoring system, it is characterised in that include corn connecting gear, figure
As picked-up unit and processor ARM (7)；Described corn connecting gear includes cylinder (6), tilts conveyer belt (5) and exciting
Device (4)；Described cylinder (6) is positioned at silo exit, and described inclination conveyer belt (5) is positioned at the lower section of cylinder (6), tilts
The bottom of conveyer belt (5) is provided with vibrator (4)；Described corn connecting gear is for being sent to image capture list by tanker corn
Yuan She district；
Described image capture unit is positioned at inclination conveyer belt (5) top, and on the direction being perpendicular to conveyer belt (5)；
Described image capture unit includes ccd video camera (1), camera lens (2), light source (3) and data collecting card successively；Image
Picked-up unit is for periodically absorbing corn, and picked-up picture is sent to processor ARM；Described cylinder (6),
Ccd video camera (1), light source (3) are all connected with processor ARM；Described processor ARM be used for controlling cylinder (6),
Ccd video camera (1) and the opening and closing of light source (3), and analyze and process the picked-up picture that image capture unit transmits.
In a kind of united reaper tanker the most according to claim 1, corn crushes on-line monitoring system, it is characterised in that
Described vibrator (4) is positioned at the centre position of conveyer belt (5)；Described image capture unit is positioned at inclination conveyer belt (5)
Centre position.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201620233431.7U CN205538564U (en) | 2016-03-24 | 2016-03-24 | Broken on -line monitoring system of cereal in combine grain tank |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201620233431.7U CN205538564U (en) | 2016-03-24 | 2016-03-24 | Broken on -line monitoring system of cereal in combine grain tank |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN205538564U true CN205538564U (en) | 2016-08-31 |
Family
ID=56783535
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201620233431.7U Active CN205538564U (en) | 2016-03-24 | 2016-03-24 | Broken on -line monitoring system of cereal in combine grain tank |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN205538564U (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105806751A (en) * | 2016-03-24 | 2016-07-27 | 江苏大学 | On-line monitoring system and method for crushing of cereals in grain tank of combine harvester |
| CN109299736A (en) * | 2018-09-18 | 2019-02-01 | 北京农业信息技术研究中心 | Product batches degree of mixing computing device and method |
| CN109548467A (en) * | 2017-09-26 | 2019-04-02 | 雷沃重工股份有限公司 | A kind of cropper monitoring device, cropper silo monitoring method and cropper |
| CN115175768A (en) * | 2020-05-13 | 2022-10-11 | 碎石大师Hmh有限公司 | Method for suppressing dust in a crusher with a spraying device |
-
2016
- 2016-03-24 CN CN201620233431.7U patent/CN205538564U/en active Active
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105806751A (en) * | 2016-03-24 | 2016-07-27 | 江苏大学 | On-line monitoring system and method for crushing of cereals in grain tank of combine harvester |
| CN109548467A (en) * | 2017-09-26 | 2019-04-02 | 雷沃重工股份有限公司 | A kind of cropper monitoring device, cropper silo monitoring method and cropper |
| CN109299736A (en) * | 2018-09-18 | 2019-02-01 | 北京农业信息技术研究中心 | Product batches degree of mixing computing device and method |
| CN115175768A (en) * | 2020-05-13 | 2022-10-11 | 碎石大师Hmh有限公司 | Method for suppressing dust in a crusher with a spraying device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105806751A (en) | On-line monitoring system and method for crushing of cereals in grain tank of combine harvester | |
| CN205538564U (en) | Broken on -line monitoring system of cereal in combine grain tank | |
| EP3646703B1 (en) | Controlling a machine based on cracked kernel detection | |
| CN113820325B (en) | Corn grain direct-harvest impurity-containing rate and breakage rate online detection system and method | |
| CN107027650A (en) | A kind of boar abnormal state detection method and device based on PSO SVM | |
| CN108664874A (en) | Underground coal flow rate testing methods based on image recognition | |
| CN108107049A (en) | Combined harvester tanker seed percentage of impurity and percentage of damage real-time monitoring device and method | |
| Chen et al. | Online field performance evaluation system of a grain combine harvester | |
| CN108120715A (en) | A kind of corn kernel percentage of damage online test method and device based on machine vision | |
| CN113298085B (en) | Corn impurity-crushing identification method and system based on Mask R-CNN | |
| Mahirah et al. | Monitoring harvested paddy during combine harvesting using a machine vision-Double lighting system | |
| CN114255394A (en) | Method, device and processor for adjusting operation parameters of agricultural machine | |
| CN202083645U (en) | FPGA (field programmable gate array)-based food quality on-line detector | |
| WO2021131317A1 (en) | Threshing state management system, threshing state management method, threshing state management program, recording medium recording threshing state management program, harvester management system, harvester, harvester management method, harvester management program, recording medium recording harvester management program, work vehicle, work vehicle management method, work vehicle management system, work vehicle management program, recording medium recording work vehicle management program, management system, management method, management program, and recording medium recording management program | |
| CN206406126U (en) | A kind of intelligent Jujun grasses cut-out sorting equipment | |
| CN103411929B (en) | A kind of rice transparency detection method | |
| Ji et al. | Apple color automatic grading method based on machine vision | |
| Siebald et al. | Referencing acoustic monitoring of cutting knives sharpness in agricultural harvesting processes using image analysis | |
| Rocha et al. | Assessing kernel processing score of harvested corn silage in real-time using image analysis and machine learning | |
| JP2021062315A (en) | Grain inspection device and culture management system using the device | |
| E. Angelia et al. | Dried Robusta Coffee Bean Quality Classification Using Convolutional Neural Network Algorithm | |
| CN109993228A (en) | Plant protection drone rice sheath blight disease recognition methods based on machine vision | |
| CN116899896A (en) | Red ginseng sorting device and sorting method based on machine vision | |
| CN112001283B (en) | Corn kernel breakage rate online detection device and detection method | |
| Wu et al. | Development of a lightweight online detection system for impurity content and broken rate in rice for combine harvesters |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |