CN109622407A - Walnut non-destructive detecting device and method based on image recognition - Google Patents
Walnut non-destructive detecting device and method based on image recognition Download PDFInfo
- Publication number
- CN109622407A CN109622407A CN201811490876.3A CN201811490876A CN109622407A CN 109622407 A CN109622407 A CN 109622407A CN 201811490876 A CN201811490876 A CN 201811490876A CN 109622407 A CN109622407 A CN 109622407A
- Authority
- CN
- China
- Prior art keywords
- walnut
- image
- weighing sensor
- rotating plate
- rotating disk
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/02—Measures preceding sorting, e.g. arranging articles in a stream orientating
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
- B07C5/342—Sorting according to other particular properties according to optical properties, e.g. colour
- B07C5/3425—Sorting according to other particular properties according to optical properties, e.g. colour of granular material, e.g. ore particles, grain
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
- B07C5/38—Collecting or arranging articles in groups
Landscapes
- Image Processing (AREA)
- Sorting Of Articles (AREA)
Abstract
The invention belongs to agricultural mechanical fields, and in particular to walnut non-destructive detecting device and method based on image recognition, for the walnut that detects and classify.Equipment includes storing containers, image device, operation display device, walnut conveying device.Walnut is turned round with rotating disk.Detection method includes the following steps: the first step, the image of the walnut in the acquisition movement of multiple images device；Weighing sensor measures the weight of walnut, and second step calculates walnut weight per unit area, flaw ratio, tone ratio, determines walnut grade according to quality criteria；Third step, different grades of walnut are pushed by two-way push device, are fallen into different storing containers.4th step, waste product walnut automatically fall into collecting receptacle；5th step, arithmetic unit statistic mixed-state result simultaneously save.The present invention can get abundant, comprehensive walnut information, can weaken disturbing factor, improves non-destructive testing precision, obtains careful walnut classification；Detection process convenient for observation, be easy to control, the degree of automation with it is high-efficient.
Description
Technical field
The invention belongs to agricultural mechanical fields, and in particular to walnut non-destructive detecting device and method based on image recognition.
Background technique
Walnut is important one of nut kind, has very high economic value and nutritive value.China's cultivation walnut is gone through
History is long, and cultivated area and yield are at the forefront in the world.In recent years, with agricultural sector structure adjustment, China's industry of planting forest or fruit tress kind
It plants area to expand, wherein the yield of walnut is grown rapidly.Walnut is the increasing peasant income on the ground such as Shanxi, Shaanxi, Xinjiang, outlet
It earns foreign exchange, economic development, social development have played great function.
Currently, being sold after only a small amount of breaking walnut shell, i.e., sold as walnut kernel, a large amount of walnuts are sold with shell.Band
Shell sells walnut, conducive to save the original local flavor of walnut, nutritional ingredient, can also avoid it is damaged, prevent pollution and insect pest.
Agricultural management department needs to evaluate the quality condition of different cultivars walnut, the preferred Walnut Cultivars of science.The marketing
Industry needs rejecting to be not suitable for edible walnut before selling walnut, and needs to be classified walnut.But there are shells for walnut, difficult
To observe walnut internal soundness, therefore, it is necessary to study walnut non-destructive detecting device and methods.
There is internal soundness in many situations in the walnut that consumer is bought in shop or even large supermarket at present.
For example, discovery part walnut is not full enough after shelling, have blackspot, is rotten, integrality is poor or other flaws.
Existing correlative study work achieves certain progress.Xinjiang Agricultural Univ traffic machine institute Liu Hang, Guo Junxian and
Xinjiang Uygur Autonomous Regions Products of The Agricultural And Animal Husbandry Machinery quality supervision and control station Wang Yong, has studied the walnut size based on machine vision
Stage division, and carried out the preliminary test of walnut size fractionation.The research and technology emphasis of the invention are different, are adopted
The principle of detection walnut also has this qualitative difference, but can provide reference for the present invention.
Summary of the invention
The purpose of the present invention: providing walnut non-destructive detecting device and method, is used for agricultural management department or scientific research list
Walnut in position evaluation variety classes or various process, counts walnut quality index, and record product technical data is preferred walnut
Kind, evaluation walnut cultivation technique improvement effect provide technological means and tool.Meanwhile the lossless inspection of walnut based on image recognition
Measurement equipment and method can be used as a kind of advanced, novel technology, for orchard worker's detection, classification high-volume walnut, serve the modern times
Change agricultural production.
The present invention uses following technical scheme.
1. the determination method of walnut quality
Determine three kinds of basic ideas of walnut quality:
First, if some walnut image area is big, and the weight of walnut is small, then this walnut is of poor quality.If walnut figure
Image planes product is small, and the weight of walnut is big, then this walnut is high-quality.If the weight of walnut is G, image area St, then walnut list
Position area weight is G/St。
Second, it is high-quality if some walnut appearance flaw (crackle, defect etc.) is few, conversely, its is of poor quality.If
Shown walnut flaw area is S in walnut imagex, walnut image area is St, then its flaw ratio is Sx/St。
Third, if some walnut exterior color and desired color are close, walnut is high-quality；Conversely, walnut is of poor quality.
If walnut ideal tone is Cl, the practical tone of walnut is Cs, then walnut tone ratio is Cs/Cl。
According to above-mentioned three kinds of basic ideas, walnut quality criteria is represented with P, can define P are as follows:
K in formula1、k2、k3For weighting coefficient, three meets following relationship: k1+k2+k3=1
According to above-mentioned formula (1), available three kinds of methods for determining walnut quality:
First, single factor test criterion
In above-mentioned formula (1), k is enabled1、k2、k3In any two be 0, another is not 0, i.e. acquisition single factor test determines
Method.It is, according to weight per unit area G/St, tone ratio Cs/Cl, flaw ratio Sx/St, select any judgement walnut matter
Amount.
Second, dual factors criterion
In above-mentioned formula (1), k is enabled1、k2、k3In any two be not 0, another is 0, i.e. acquisition dual factors determine
Method.
Third, multifactor criterion
In above-mentioned formula (1), k1、k2、k3It is not 0, that is, obtains multifactor criterion.Formula (1) can also be expanded
For
Formula (2) considers that other are more multifactor, also belongs to multifactor criterion.
2. walnut storing containers, image device, operation display device
The present invention includes walnut storing containers, image device, operation display device, walnut conveying device.
Walnut storing containers include walnut frame 1, Grade A case 13, goods of inferior quality case 23, collecting receptacle 11, also can be set three
The product casees 24 such as equal product case 12, four.
Walnut frame 1 is for storing the walnut 2 not yet detected.Grade A case is used to store the walnut of quality first-class, goods of inferior quality
Case is for storing the second-class walnut of quality.The purposes of the product case such as third grade product case, four and so on.Collecting receptacle is not suitable for for storing
Edible walnut.
The important function of each image device is identification walnut, acquisition and provides walnut image.Image device may be configured as
Six sets or more sets: image device 1, image device 24, image device 35, image device 46, image device 57, image dress
Set 68.
As shown in instruction sheet 1, walnut is dynamically that multiple images device can be from different perspectives in the detection process
Walnut is observed with position, the image that each walnut is located at highest point and walnut is located at highest point two sides can be acquired one by one, be related to core
Posture during peach uphill process and decline.
It is located in each image device, some is respectively S in the area for checking peach image1、S2、S3......Sn, with mean value
St=(S1+S2+……Sn)/n represents the walnut image area.If the walnut flaw area is respectively S in each image device1x、
S2x、S3x......Snx, with mean value Sx=(S1x+S2x+……Snx)/n represents the walnut image flaw area.If each image dress
The walnut tone is respectively C in setting1s、C2s、C3s......Cns, with mean value Cs=(C1s+C2s+C3s+……Cns)/n represents the core
The practical tone of peach.
The image device of actual use can be less than six sets or more than six sets.The quantity of image device depends on wishing
The detection accuracy of the walnut quality of acquisition.Higher walnut internal soundness detection can be obtained using greater number of image device
Walnut can be preferably classified by precision.
Operation display device includes arithmetic unit 16, display 17.Arithmetic unit can divide walnut image obtained
Analysis processing, the area including calculating walnut image compare color, analysis walnut profile and the surface characteristics of walnut, determine flaw
Deng.Display can then show detection and statistic analysis result.
Arithmetic unit can also be analyzed according to mathematical statistics professional technique and handle various detection datas, and record saves testing number
According to various statistic analysis results, convenient at that time and in the future called data.
3. walnut conveying device
Walnut conveying device includes two-way push device 9, rotating disk 14, rotary shaft 15, ring baffle 18, weighing sensor
19, partition 20, rotating plate 21, rotational pin 22.Also two-way push device 9 can be set.
Rotating disk 14 is the main body of the device, it is constantly turned round around rotary shaft 15.It is circumferentially equipped in rotating disk multiple
Partition 20, weighing sensor 19.As shown in instruction sheet 1, the bottom and side of walnut frame 1 are provided with slot, are fixed in rotating disk
Partition, weighing sensor can enter walnut frame slot in.When walnut is located at two adjacent partitions and a weighing sensor institute
When the hemi-closure space surrounded and the upward space openings, walnut is driven to turn round with rotating disk.Walnut is out of walnut frame 1
Portion is carried over, and in its turning course, each image device all identifies this walnut, is acquired and is provided this walnut image；Walnut and
After its flaw information is transmitted to arithmetic unit processing, shown by display.During walnut is turned round with rotating disk, weighing and sensing
Device measures the actual weight G of walnut, is transmitted to arithmetic unit.Ring baffle 18 is located at the two sides of rotating disk.It was turned round in walnut
Cheng Zhong, ring baffle 18 prevent walnut is automatic on the rotating pan from sliding toward lateral, that is, prevent walnut from moving along rotating disk axis direction
It moves and falls off from rotating disk.
Above, the image area S of single walnut had been introduced in image device partt.When weighing sensor measures walnut
Actual weight G after, arithmetic unit calculate walnut weight G and StRatio, i.e. walnut weight per unit area G/St.Arithmetic unit is also counted
Calculate the tone ratio C of walnuts/Cl, flaw ratio Sx/St, and by formula (1), (2) calculate P.
P (or G/StOr other criterions) critical value can be determined by 3 kinds of methods: 1. according to professional standard, by its walnut
Sample parameter determines.2. being determined according to the statistical result of this batch of product sample or other Bulk products by comparing method.3. foundation
Crowd demand determines.
According to P calculated (or G/StOr other criterions), walnut is divided into Grade A, goods of inferior quality, waste product by arithmetic unit；Or
It is subdivided into larger class: for example, the product such as Grade A, goods of inferior quality, third grade product, four, waste product.
It is equipped with a two-way push device in walnut conveying device, or sets there are two two-way push device, i.e. two-way push device
One 9, two-way push device 2 10.
The walnut for belonging to Grade A is pushed under arithmetic unit control by two-way push device one, is transported to the side of rotating disk
It is dynamic.Due to ring baffle 18 can only block walnut half or less than half, two-way push device one can make Grade A core
Peach crosses ring baffle, falls into Grade A case 13.It is similar, belong to the walnut of goods of inferior quality, is pushed by two-way push device one, to
The another side of rotating disk moves, and walnut crosses ring baffle, falls into goods of inferior quality case 23.
When being equipped with third grade product case, the product case such as four, it is not belonging to Grade A, the walnut of goods of inferior quality continues to move ahead with rotating disk,
Move to two-way push device 2 10.The walnut for belonging to third grade product is pushed by two-way push device two, to the flank movement of rotating disk,
Ring baffle is crossed, third grade product case 12 is fallen into.It is similar, belong to the walnut of four equal product, is pushed, crossed by two-way push device two
Ring baffle falls into four equal product casees 24.
Belong to the walnut of waste product, not will receive the effect of two-way push device, turned round with rotating disk, walnut is still located on two
In semi-enclosed space that partition and a weighing sensor are surrounded, but with the revolution of rotating disk, this semi-enclosed sky
Between will will appear downward opening posture, at this moment, under the action of self-weight, walnut will automatically fall into collecting receptacle 11.
It is provided with slot in the bottom of walnut frame 1, thus the partition along rotating disk circle distribution enters in walnut frame 1.In walnut
The position of frame bottom slot is equipped with rotational pin 22, rotating plate 21, and rotating plate can rotate around rotational pin 22 since walnut frame bottom, turn
The movement of plate is driven by the partition rotated, and after the end of rotating plate is separated with partition, rotating plate is rotated down, and movement terminates in walnut
Plane corresponding to frame bottom.Rotating plate 21 cannot be rotated down from walnut frame bottom, to prevent walnut from thus leaking out.21 end of rotating plate
End is respectively less than walnut size with the gap of rotating disk, with the gap of weighing sensor, with the gap of ring baffle, to avoid core
Peach falls from the end of rotating plate.
Two-way push device one, two-way push device two can be shift fork, and walnut is located in two feet of shift fork.Shift fork can be to
Both direction rotation, falls into walnut in different casees.Two-way push device one, two-way push device two are also possible to have two-way fortune
The hydraulic system or baric systerm of dynamic function.
The case where for walnut negligible amounts, can only arrange 1 set of walnut conveying device in the surrounding of walnut frame 1.For
The a fairly large number of situation of walnut is distributed multiple walnut conveying devices and other in the surrounding of walnut frame 1 to improve working efficiency
Device, each device work at the same time.
In conclusion the present invention has following innovative point:
1. the walnut non-destructive detecting device based on image recognition includes walnut storing containers, image device, operation display dress
It sets, walnut conveying device, it is characterised in that: walnut is turned round with rotating disk, and experience rises, reaches the process of highest point and decline；
Multiple images device acquires the image of each walnut in movement one by one from different location respectively；For the single walnut in inspection,
With S1、S2、S3......SnThe walnut image area in each image device is represented, with mean value St=(S1+S2+……Sn)/n generation
The table walnut image area；Rotating disk is turned round, and is circumferentially equipped with multiple partitions and weighing sensor in rotating disk, walnut is located at
In semi-enclosed space that two partitions and a weighing sensor are surrounded, ring baffle prevents walnut automatically toward lateral sliding
It falls, walnut frame is provided with slot, and the partition, weighing sensor in rotating disk can enter in walnut frame slot；When weighing sensor measures
Out when examining the weight G of single walnut, arithmetic unit calculates walnut weight per unit area G/St, according to G/StCritical value determine
Whether the affiliated grade of the walnut is waste product；Different grades of walnut is pushed by the control of arithmetic unit by two-way push device,
Ring baffle only block walnut half or less than half, two-way push device makes different grades of walnut cross ring baffle, fall
Enter in different walnut storing containers；Belong to the walnut of waste product, not by the effect of two-way push device, when two partition and one
Semi-enclosed space openings that weighing sensor is surrounded downward when, waste product walnut automatically falls into collecting receptacle；Pacify on walnut frame
Equipped with rotating plate, rotating plate is driven by the partition rotated, is rotated upwards, and after the end of rotating plate is separated with partition, rotating plate is to backspin
Turn, movement terminates in walnut frame bottom, cannot be rotated down from walnut frame bottom；Rotating plate end and the gap of rotating disk and weighing
The gap of sensor and the gap of ring baffle are respectively less than walnut size, and walnut will not leak down from rotating plate.
2. the walnut non-destructive detecting device based on image recognition includes walnut storing containers, image device, operation display dress
It sets, walnut conveying device, it is characterised in that: walnut is turned round with rotating disk, and experience rises, reaches the process of highest point and decline,
Multiple images device acquires the image of each walnut in movement one by one from different location respectively；For some inspection walnut,
With S1、S2、S3......SnIts image area in each image device is indicated, with mean value St=(S1+S2+……Sn)/n generation
The table walnut image area；With S1x、S2x、S3x......SnxFlaw area of the walnut in each image device is represented, with equal
Value Sx=(S1x+S2x+……Snx)/n represents the walnut image flaw area；With C1s、C2s、C3s......CnsThe walnut is represented to exist
Tone in each image device, with mean value Cs=(C1s+C2s+C3s+……Cns)/n represents the practical tone of the walnut；It is only single
When a image device, S is substituted into n=1t、Sx、CsCalculation formula；Rotating disk is turned round, and is circumferentially equipped in rotating disk multiple
Partition and weighing sensor, walnut are located in semi-enclosed space that two partitions and a weighing sensor are surrounded, annular
Baffle prevents walnut from sliding automatically toward lateral, and walnut frame is provided with slot, and the partition, weighing sensor in rotating disk can enter walnut
In frame slot；Weighing sensor is measured in the weight G for examining single walnut, and arithmetic unit calculates walnut weight per unit area G/St、
Flaw ratio Sx/St, tone ratio Cs/Cl, and according to formulaWalnut quality criteria P is calculated, according to P
And its whether critical value determines the affiliated grade of the walnut or is waste product；Different grades of walnut, by the control of arithmetic unit, by double
Pushed to pusher one or two-way push device two, ring baffle only block walnut half or less than half, walnut crosses annular
Baffle falls into Grade A case, goods of inferior quality case, in third grade product case；Belong to the walnut of waste product, not by the effect of two-way push device, when
Semi-enclosed space openings that two partition and a weighing sensor are surrounded downward when, waste product walnut automatically falls into waste product
Case；Rotating plate is installed on walnut frame, rotating plate is driven by the partition rotated, is rotated upwards, when the end of rotating plate and partition point
From rear, rotating plate is rotated down, and movement terminates in walnut frame bottom, cannot be rotated down from walnut frame bottom；Rotating plate end and rotation
The gap of disk is respectively less than walnut size with the gap of weighing sensor, with the gap of ring baffle, and walnut will not leak from rotating plate
Under；Multiple walnut conveying devices are distributed in the surrounding of walnut frame.
3. the walnut lossless detection method based on image recognition, equipment includes walnut storing containers, image device, operation
Display device, walnut conveying device, it is characterised in that: rotating disk is turned round, and is circumferentially equipped with multiple partitions and title in rotating disk
Sensor is retransmitted, walnut is located in semi-enclosed space that two partitions and a weighing sensor are surrounded；Walnut is with rotating disk
Revolution, experience rise, reach the process of highest point and decline；Ring baffle prevents walnut from sliding automatically toward lateral, ring baffle
Only block walnut half or less than half；Walnut frame is provided with slot, and the partition, weighing sensor in rotating disk can enter walnut
In frame slot；Rotating plate is installed on walnut frame, rotating plate is driven by the partition rotated, upwards rotate, when rotating plate end with every
After piece separation, rotating plate is rotated down, and movement terminates in walnut frame bottom, cannot be rotated down from walnut frame bottom；Rotating plate end with
The gap of rotating disk is respectively less than walnut size with the gap of weighing sensor, with the gap of ring baffle, and walnut will not be from rotating plate
Leak down at place；The step of its detection method, is as follows: the first step, and multiple images device acquires in movement from different location one by one respectively
The image of walnut；For the single walnut in inspection, with S1、S2、S3......SnRepresent the walnut image surface in each image device
Product, with mean value St=(S1+S2+……Sn)/n represents the walnut image area, and weighing sensor, which is measured, is examining single walnut
Weight G；Second step, arithmetic unit calculate walnut weight per unit area G/St, according to G/StCritical value determine belonging to the walnut etc.
Whether grade is waste product；Third step, different grades of walnut are pushed by two-way push device by the control of arithmetic unit, cross ring
Shape baffle is fallen into different walnut storing containers；4th step belongs to the walnut of waste product, not by the effect of two-way push device, when
Semi-enclosed space openings that two partition and a weighing sensor are surrounded downward when, waste product walnut automatically falls into waste product
Case；5th step, arithmetic unit statistic mixed-state result simultaneously save.
4. the walnut lossless detection method based on image recognition, equipment includes walnut storing containers, image device, operation
Display device, walnut conveying device, it is characterised in that: rotating disk is turned round, and is circumferentially equipped with multiple partitions and title in rotating disk
Sensor is retransmitted, walnut is located in semi-enclosed space that two partitions and a weighing sensor are surrounded；Walnut is with rotating disk
Revolution, experience rise, reach the process of highest point and decline；Ring baffle prevents walnut from sliding automatically toward lateral, ring baffle
Only block walnut half or less than half；Walnut frame is provided with slot, and the partition, weighing sensor in rotating disk can enter walnut
In frame slot；Rotating plate is installed on walnut frame, rotating plate is driven by the partition rotated, upwards rotate, when rotating plate end with every
After piece separation, rotating plate is rotated down, and movement terminates in walnut frame bottom, cannot be rotated down from walnut frame bottom；Rotating plate end with
The gap of rotating disk is respectively less than walnut size with the gap of weighing sensor, with the gap of ring baffle, and walnut will not be from rotating plate
Leak down at place；Multiple walnut conveying devices are distributed in the surrounding of walnut frame；The step of its detection method, is as follows: the first step, multiple
Image device acquires the image of each walnut in movement one by one from different location respectively；For some inspection walnut, with S1、
S2、S3......SnIts image area in each image device is indicated, with mean value St=(S1+S2+……Sn)/n is represented should
Walnut image area；With S1x、S2x、S3x......SnxFlaw area of the walnut in each image device is represented, with mean value Sx
=(S1x+S2x+……Snx)/n represents the walnut image flaw area；With C1s、C2s、C3s......CnsThe walnut is represented each
Tone in image device, with mean value Cs=(C1s+C2s+C3s+……Cns)/n represents the practical tone of the walnut；Only arrangement is single
When image device, S is substituted into n=1t、Sx、CsCalculation formula；Weighing sensor is measured in the weight G for examining single walnut；The
Two steps, arithmetic unit calculate walnut weight per unit area G/St, flaw ratio Sx/St, tone ratio Cs/Cl, and according to formulaCalculate walnut quality criteria P, according to P and its critical value determine the affiliated grade of the walnut or whether
For waste product；Third step, different grades of walnut are pushed away by the control of arithmetic unit by two-way push device one or two-way push device two
Dynamic, walnut crosses ring baffle, falls into Grade A case, goods of inferior quality case, in third grade product case；4th step belongs to the walnut of waste product, no
By the effect of two-way push device, when semi-enclosed space openings that two partition and a weighing sensor are surrounded downward
When, waste product walnut automatically falls into collecting receptacle；5th step, arithmetic unit statistic mixed-state result simultaneously save.
5. the walnut non-destructive detecting device according to above 1 or 2 based on image recognition, it is characterised in that two-way push
Device is shift fork or hydraulic system or baric systerm.
6. the walnut lossless detection method according to above 4 based on image recognition, it is characterised in that two-way push device
One or two-way push device second is that shift fork or hydraulic system or baric systerm.
7. the walnut lossless detection method according to above 3 or 4 based on image recognition, it is characterised in that quality criteria
P, weight per unit area G/St, flaw ratio Sx/St, tone ratio Cs/ClCritical value sample as defined in walnut professional standard system
Determination is counted, or is determined by the statistical data of the sample of selected batch, or determined by crowd demand.
Beneficial effects of the present invention: by multiple images identification device, walnut is identified from different location, can get more rich
Rich walnut information；Walnut experience rises, reaches the process of highest point and decline, and walnut image obtained is more fully；Core
Peach is taken out of from walnut frame one by one by two partitions and weighing sensor, falls into different walnut storing containers, and walnut will not heap
Product, will not omit, and detection process is very orderly, convenient for observation；Walnut is from walnut storing containers --- and the top of walnut frame exports,
It is easier to control than horizontal output or lower part output, there is the advantage on position；Ask average value, the tone of walnut image area flat
Mean value, flaw average value are conducive to improve detection accuracy；The image for acquiring walnut different parts from different perspectives, can weaken walnut
Detection error caused by split position seam；The ratio for asking walnut weight Yu walnut image, in the feelings for not destroying walnut shell
Under condition, walnut internal soundness situation may infer that, method is more ingenious, further determines core in conjunction with walnut flaw ratio, tone ratio
Peach quality improves the science and advance of walnut detection.Multiple walnut conveying devices are distributed in the surrounding of walnut frame, it can be real
Existing multistation walnut non-destructive testing and classification, improve walnut detection efficiency；Using shift fork or hydraulic system or baric systerm, certainly
Dynamic that walnut is made to fall into the product casees, collecting receptacle such as Grade A case, goods of inferior quality case, third grade product case, four, it is more careful to be classified, and automates journey
Degree is high.Quality criteria P, weight per unit area G/St, flaw ratio Sx/St, tone ratio Cs/ClCritical value there are many value scheme,
Different demands can be suited.
Detailed description of the invention
Fig. 1 is schematic front view of the invention, and Fig. 2 is schematic top plan view of the invention.In figure: 1- walnut frame, 2- walnut,
3- image device one, 4- image device two, 5- image device three, 6- image device four, 7- image device five, 8- image device
Six, 9- two-way push device one, 10- two-way push device two, 11- collecting receptacle, 12- third grade product case, 13- Grade A case, 14- rotation
Disk, 15- rotary shaft, 16- arithmetic unit, 17- display, 18- ring baffle, 19- weighing sensor, 20- partition, 21- rotating plate,
22- rotational pin, 23- goods of inferior quality case, the equal product case of 24- tetra-.
Specific embodiment
Embodiment of the present invention is described further below in conjunction with attached drawing, specific embodiment is not the present invention any
Limitation.
Embodiment one: there are 4 sets of walnut conveying devices around walnut frame 1.Image device has: image device one, figure
As device two, image device three, image device four, image device five, image device six.Using two-way push device one, Bidirectional pulling
Dynamic device two.Using shift fork as two-way push device.The product casees, collecting receptacle such as Grade A case, goods of inferior quality case, third grade product case, four are set.Meter
Walnut quality criteria P is calculated, foundation P and its critical value determine the affiliated grade of the walnut or whether be waste product.Quality criteria P's is critical
The statistical data of value sample as defined in walnut professional standard determines.
Embodiment two: there are 5 sets of walnut conveying devices around walnut frame 1.Image device has: image device one, figure
As device two, image device three, image device four, image device five, image device six, image device seven.Using two-way push
Device one, using hydraulic system as two-way push device.Grade A case, goods of inferior quality case, collecting receptacle are set.According to G/StCritical value sentence
Whether the fixed affiliated grade of the walnut is waste product.The critical value of quality criteria P is determined by crowd demand.
Claims (7)
1. the walnut non-destructive detecting device based on image recognition includes walnut storing containers, image device, operation display device, core
Peach conveying device, it is characterised in that: walnut is turned round with rotating disk, and experience rises, reaches the process of highest point and decline；Multiple figures
As device acquires the image of each walnut in movement one by one from different location respectively；For the single walnut in inspection, with S1、S2、
S3......SnThe walnut image area in each image device is represented, with mean value St=(S1+S2+……Sn)/n represents the walnut
Image area；Rotating disk is turned round, and is circumferentially equipped with multiple partitions and weighing sensor in rotating disk, walnut is located at two partitions
In the semi-enclosed space surrounded with a weighing sensor, ring baffle prevents walnut from sliding automatically toward lateral, walnut frame
It is provided with slot, the partition, weighing sensor in rotating disk can enter in walnut frame slot；It measures and is being examined individually when weighing sensor
When the weight G of walnut, arithmetic unit calculates walnut weight per unit area G/St, according to G/StCritical value determine belonging to the walnut
Whether grade is waste product；Different grades of walnut is pushed by the control of arithmetic unit by two-way push device, and ring baffle is only
Block walnut half or less than half, two-way push device makes different grades of walnut cross ring baffle, fall into different core
In peach storing containers；Belong to the walnut of waste product, not by the effect of two-way push device, when two partition and a weighing sensor
The semi-enclosed space openings surrounded downward when, waste product walnut automatically falls into collecting receptacle；Rotating plate is installed on walnut frame, is turned
Plate is driven by the partition rotated, is rotated upwards, after the end of rotating plate is separated with partition, rotating plate is rotated down, and movement terminates in
Walnut frame bottom cannot be rotated down from walnut frame bottom；Between rotating plate end and the gap and weighing sensor of rotating disk
The gap of gap and ring baffle is respectively less than walnut size, and walnut will not leak down from rotating plate.
2. the walnut non-destructive detecting device based on image recognition includes walnut storing containers, image device, operation display device, core
Peach conveying device, it is characterised in that: walnut is turned round with rotating disk, and experience rises, reaches the process of highest point and decline, Duo Getu
As device acquires the image of each walnut in movement one by one from different location respectively；For some inspection walnut, with S1、S2、
S3......SnIts image area in each image device is indicated, with mean value St=(S1+S2+……Sn)/n represents the walnut
Image area；With S1x、S2x、S3x......SnxFlaw area of the walnut in each image device is represented, with mean value Sx=
(S1x+S2x+……Snx)/n represents the walnut image flaw area；With C1s、C2s、C3s......CnsThe walnut is represented in each figure
As the tone in device, with mean value Cs=(C1s+C2x+C3s+……Cns)/n represents the practical tone of the walnut；Only single image
When device, S is substituted into n=1t、Sx、CsCalculation formula；Rotating disk revolution, be circumferentially equipped in rotating disk multiple partitions and
Weighing sensor, walnut are located in semi-enclosed space that two partitions and a weighing sensor are surrounded, ring baffle resistance
Only walnut slides toward lateral automatically, and walnut frame is provided with slot, and the partition, weighing sensor in rotating disk can enter walnut frame slot
It is interior；Weighing sensor is measured in the weight G for examining single walnut, and arithmetic unit calculates walnut weight per unit area G/St, flaw
Compare Sx/St, tone ratio Cs/Cl, and according to formulaCalculate walnut quality criteria P, according to P and its
Whether critical value determines the affiliated grade of the walnut or is waste product；Different grades of walnut, by the control of arithmetic unit, by Bidirectional pulling
Dynamic device one or two-way push device two push, ring baffle only block walnut half or less than half, walnut cross ring baffle,
Fall into Grade A case, goods of inferior quality case, in third grade product case；Belong to the walnut of waste product, not by the effect of two-way push device, when two
Semi-enclosed space openings that partition and a weighing sensor are surrounded downward when, waste product walnut automatically falls into collecting receptacle；?
Rotating plate is installed on walnut frame, rotating plate is driven by the partition rotated, it rotates upwards, after the end of rotating plate is separated with partition,
Rotating plate is rotated down, and movement terminates in walnut frame bottom, cannot be rotated down from walnut frame bottom；Between rotating plate end and rotating disk
Gap is respectively less than walnut size with the gap of weighing sensor, with the gap of ring baffle, and walnut will not leak down from rotating plate；?
Multiple walnut conveying devices are distributed in the surrounding of walnut frame.
3. the walnut lossless detection method based on image recognition, equipment includes that walnut storing containers, image device, operation are shown
Device, walnut conveying device, it is characterised in that: rotating disk is turned round, and multiple partitions are circumferentially equipped in rotating disk and weighing passes
Sensor, walnut are located in semi-enclosed space that two partitions and a weighing sensor are surrounded；Walnut is turned round with rotating disk,
Experience rises, reaches the process of highest point and decline；Ring baffle prevents walnut from sliding automatically toward lateral, and ring baffle only blocks
The half of walnut or less than half；Walnut frame is provided with slot, and the partition, weighing sensor in rotating disk can enter walnut frame slot
It is interior；Rotating plate is installed on walnut frame, rotating plate is driven by the partition rotated, is rotated upwards, when the end of rotating plate and partition point
From rear, rotating plate is rotated down, and movement terminates in walnut frame bottom, cannot be rotated down from walnut frame bottom；Rotating plate end and rotation
The gap of disk is respectively less than walnut size with the gap of weighing sensor, with the gap of ring baffle, and walnut will not leak from rotating plate
Under；The step of its detection method, is as follows: the first step, and multiple images device acquires the walnut in movement one by one from different location respectively
Image；For the single walnut in inspection, with S1、S2、S3......SnThe walnut image area in each image device is represented,
With mean value St=(S1+S2+……Sn)/n represents the walnut image area, and weighing sensor is measured in the weight for examining single walnut
Measure G；Second step, arithmetic unit calculate walnut weight per unit area G/St, according to G/StCritical value determine the affiliated grade of the walnut
It or whether is waste product；Third step, different grades of walnut are pushed by two-way push device by the control of arithmetic unit, cross annular
Baffle is fallen into different walnut storing containers；4th step belongs to the walnut of waste product, not by the effect of two-way push device, when it
Semi-enclosed space openings that two partitions and a weighing sensor are surrounded downward when, waste product walnut automatically falls into waste product
Case；5th step, arithmetic unit statistic mixed-state result simultaneously save.
4. the walnut lossless detection method based on image recognition, equipment includes that walnut storing containers, image device, operation are shown
Device, walnut conveying device, it is characterised in that: rotating disk is turned round, and multiple partitions are circumferentially equipped in rotating disk and weighing passes
Sensor, walnut are located in semi-enclosed space that two partitions and a weighing sensor are surrounded；Walnut is turned round with rotating disk,
Experience rises, reaches the process of highest point and decline；Ring baffle prevents walnut from sliding automatically toward lateral, and ring baffle only blocks
The half of walnut or less than half；Walnut frame is provided with slot, and the partition, weighing sensor in rotating disk can enter walnut frame slot
It is interior；Rotating plate is installed on walnut frame, rotating plate is driven by the partition rotated, is rotated upwards, when the end of rotating plate and partition point
From rear, rotating plate is rotated down, and movement terminates in walnut frame bottom, cannot be rotated down from walnut frame bottom；Rotating plate end and rotation
The gap of disk is respectively less than walnut size with the gap of weighing sensor, with the gap of ring baffle, and walnut will not leak from rotating plate
Under；Multiple walnut conveying devices are distributed in the surrounding of walnut frame；The step of its detection method, is as follows: the first step, multiple images
Device acquires the image of each walnut in movement one by one from different location respectively；For some inspection walnut, with S1、S2、
S3......SnIts image area in each image device is indicated, with mean value St=(S1+S2+……Sn)/n represents the walnut
Image area；With S1x、S2x、S3x......SnxFlaw area of the walnut in each image device is represented, with mean value Sx=
(S1x+S2x+……Snx)/n represents the walnut image flaw area；With C1s、C2s、C3x......CnsThe walnut is represented in each figure
As the tone in device, with mean value Cs=(C1s+C2x+C3x+……Cns)/n represents the practical tone of the walnut；The only single figure of arrangement
When as device, S is substituted into n=1t、Sx、CsCalculation formula；Weighing sensor is measured in the weight G for examining single walnut；Second
Step, arithmetic unit calculate walnut weight per unit area G/St, flaw ratio Sx/St, tone ratio Cs/Cl, and according to formulaCalculate walnut quality criteria P, according to P and its critical value determine the affiliated grade of the walnut or whether
For waste product；Third step, different grades of walnut are pushed away by the control of arithmetic unit by two-way push device one or two-way push device two
Dynamic, walnut crosses ring baffle, falls into Grade A case, goods of inferior quality case, in third grade product case；4th step belongs to the walnut of waste product, no
By the effect of two-way push device, when semi-enclosed space openings that two partition and a weighing sensor are surrounded downward
When, waste product walnut automatically falls into collecting receptacle；5th step, arithmetic unit statistic mixed-state result simultaneously save.
5. the walnut non-destructive detecting device according to claim 1 or 2 based on image recognition, it is characterised in that two-way push
Device is shift fork or hydraulic system or baric systerm.
6. the walnut lossless detection method according to claim 4 based on image recognition, it is characterised in that two-way push device
One or two-way push device second is that shift fork or hydraulic system or baric systerm.
7. the walnut lossless detection method according to claim 3 or 4 based on image recognition, it is characterised in that quality criteria
P, weight per unit area G/St, flaw ratio Sx/St, tone ratio Cs/ClCritical value sample as defined in walnut professional standard system
Determination is counted, or is determined by the statistical data of the sample of selected batch, or determined by crowd demand.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811490876.3A CN109622407B (en) | 2018-11-30 | 2018-11-30 | Image recognition-based walnut nondestructive testing equipment and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811490876.3A CN109622407B (en) | 2018-11-30 | 2018-11-30 | Image recognition-based walnut nondestructive testing equipment and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109622407A true CN109622407A (en) | 2019-04-16 |
| CN109622407B CN109622407B (en) | 2020-12-29 |
Family
ID=66071840
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811490876.3A Active CN109622407B (en) | 2018-11-30 | 2018-11-30 | Image recognition-based walnut nondestructive testing equipment and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109622407B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110302974A (en) * | 2019-06-20 | 2019-10-08 | 申丽丽 | A kind of apple orchard volume production fruit weight intelligent detection equipment |
| CN112108393A (en) * | 2020-09-29 | 2020-12-22 | 西南大学 | Pecan shell breakage detection sorting device |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201807545U (en) * | 2010-09-26 | 2011-04-27 | 天津小泉精密金属制品有限公司 | Automatic sorting device |
| CN202591136U (en) * | 2012-05-15 | 2012-12-12 | 塔里木大学 | Automatic walnut sorting machine based on relative specific weight |
| CN203578234U (en) * | 2013-11-19 | 2014-05-07 | 中国科学院沈阳自动化研究所 | Red date appearance online visual detecting and grading system |
| US20140142745A1 (en) * | 2012-11-20 | 2014-05-22 | Bratney Companies | Sorting system for damaged product |
| CN104438133A (en) * | 2014-12-17 | 2015-03-25 | 杭州郝姆斯食品有限公司 | Walnut sorting machine and walnut sorting method |
| CN106442525A (en) * | 2016-08-30 | 2017-02-22 | 石河子大学 | Online detecting method for wizened defects of insides of walnuts |
| JP2017181289A (en) * | 2016-03-30 | 2017-10-05 | 株式会社大林組 | Soil qualities division device and soil qualities division method |
| CN108480232A (en) * | 2018-03-20 | 2018-09-04 | 上海理工大学 | Matrimony vine sorting technique based on shape and plumpness feature |
-
2018
- 2018-11-30 CN CN201811490876.3A patent/CN109622407B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201807545U (en) * | 2010-09-26 | 2011-04-27 | 天津小泉精密金属制品有限公司 | Automatic sorting device |
| CN202591136U (en) * | 2012-05-15 | 2012-12-12 | 塔里木大学 | Automatic walnut sorting machine based on relative specific weight |
| US20140142745A1 (en) * | 2012-11-20 | 2014-05-22 | Bratney Companies | Sorting system for damaged product |
| CN203578234U (en) * | 2013-11-19 | 2014-05-07 | 中国科学院沈阳自动化研究所 | Red date appearance online visual detecting and grading system |
| CN104438133A (en) * | 2014-12-17 | 2015-03-25 | 杭州郝姆斯食品有限公司 | Walnut sorting machine and walnut sorting method |
| JP2017181289A (en) * | 2016-03-30 | 2017-10-05 | 株式会社大林組 | Soil qualities division device and soil qualities division method |
| CN106442525A (en) * | 2016-08-30 | 2017-02-22 | 石河子大学 | Online detecting method for wizened defects of insides of walnuts |
| CN108480232A (en) * | 2018-03-20 | 2018-09-04 | 上海理工大学 | Matrimony vine sorting technique based on shape and plumpness feature |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110302974A (en) * | 2019-06-20 | 2019-10-08 | 申丽丽 | A kind of apple orchard volume production fruit weight intelligent detection equipment |
| CN112108393A (en) * | 2020-09-29 | 2020-12-22 | 西南大学 | Pecan shell breakage detection sorting device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109622407B (en) | 2020-12-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105021529B (en) | Fusion spectrum and the crop disease and insect of image information identify and distinguish between method | |
| CN110455343A (en) | A kind of commercial quality detecting system based on big data | |
| CN102288606B (en) | Pollen viability measuring method based on machine vision | |
| CN102253052B (en) | Grain quality on-line detection apparatus based on field programmable gate array (FPGA) | |
| CN104256882B (en) | Based on reconstituted tobacco ratio measuring method in the pipe tobacco of computer vision | |
| CN109425715B (en) | Full-automatic unmanned grain testing system | |
| CN101013079B (en) | Small-sized material digitalized detecting and grading apparatus | |
| CN104198325B (en) | Stem ratio measuring method in pipe tobacco based on computer vision | |
| CN109622407A (en) | Walnut non-destructive detecting device and method based on image recognition | |
| CN114112932B (en) | Deep learning-based hyperspectral detection method and sorting equipment for ripeness degree of oil tea fruits | |
| CN105975966A (en) | Rice grain mildew nondestructive test method | |
| CN106940292A (en) | Bar denier wood raw material quick nondestructive discrimination method of damaging by worms based on multi-optical spectrum imaging technology | |
| CN111881864A (en) | Wisdom farming crops growth full-period full-dynamic monitoring management system | |
| CN105699304B (en) | A kind of method of material information representated by acquisition spectral information | |
| CN117388292B (en) | Durian internal quality nondestructive rapid detection and classification method and durian internal quality nondestructive rapid detection and classification system | |
| Thong et al. | Mango sorting mechanical system combines image processing | |
| CN107576600A (en) | A kind of quick determination method for smearing tea grain size category | |
| CN105675538A (en) | Method for detection of linseed cake nutrients | |
| CN100449309C (en) | Non destruction detection method of apple internal quality | |
| KR20120021769A (en) | Fruit automatic sorting device using sensor | |
| CN109829387B (en) | Method for determining grabbing priority of stacked serial fruits based on depth of main fruit stalks for parallel robot | |
| CN104865198A (en) | Machine vision-based fresh fat liver quality automatic grading system and method | |
| CN102784762A (en) | Method for rapidly detecting and sorting mouldy grains by applying laser | |
| CN106404779A (en) | Auxiliary detection method for corn seed purity based on form analysis | |
| CN110030927A (en) | Walnut detection device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |