CN101701915A - Device and method for detecting stored-grain insects based on visible light-near infrared binocular machine vision - Google Patents
Device and method for detecting stored-grain insects based on visible light-near infrared binocular machine vision Download PDFInfo
- Publication number
- CN101701915A CN101701915A CN200910235075A CN200910235075A CN101701915A CN 101701915 A CN101701915 A CN 101701915A CN 200910235075 A CN200910235075 A CN 200910235075A CN 200910235075 A CN200910235075 A CN 200910235075A CN 101701915 A CN101701915 A CN 101701915A
- Authority
- CN
- China
- Prior art keywords
- grain
- worm
- near infrared
- stored
- visible light
- 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
Images
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to a device and a method for detecting stored-grain insects, which consists of a stored-grain insect separation part, a stored-grain insect transmission part and a machine vision part, wherein the stored-grain insect separation part efficiently separates stored-grain insects from a grain sample by utilizing a screening motor to drive a vibrating screen; the stored-grain insect transmission part can transmit an acquisition box to an appointed position and can automatically clear screen underflows in the acquisition box; the machine vision part consists of a control module, a computer, a lighting box, a near infrared camera, a visible light CCD, a light source and corresponding software. The method for detecting stored-grain insects comprises the following steps of sequentially acquiring near infrared images and visible light images of the screen underflows for information fusion, extracting entire morphological characteristics and local morphological characteristics of the stored-grain insects, and automatically determining the kinds and the quantity of the live stored-grain insects by utilizing an optimized stored-grain insects characteristic space and identification software, thus realizing the real-time, precise and automatic detection of normal stored-grain insects.
Description
Technical field
The present invention relates to a kind of pick-up unit and method, refer in particular to storage pest automatic testing method and device thereof based on visible light-near infrared binocular computer vision at storage pest.
Background technology
China is maximum in the world grain-production, storage and consumption big country, does foodstuff preservation well and has even more important meaning undoubtedly.The grain loss in weight after the whole world results is about 10-15%, and the grain of annual storage period has at least 5% to be ravaged by insect.The national treasury grain storage loss percentage of China is about 0.2%, and this is a real accomplishment.But the cost of for this reason paying also is huge, particularly a large amount of uses of pesticide.In order to ensure the safe storage of grain, grain storage wants use of insecticide to carry out fumigant insect killing one time every year at least, and many places will be fumigated twice, even it is more, this has not only increased spending, the pollution of grain and environment also is on the rise, and the resistance to the action of a drug level of insect improves fast.A main cause that causes this situation is that the control of insect decision-making lacks science, and one of control of insect decision-making important scientific basis is exactly the accurate detection of grain storage pest.China's " grain and oil storage technology standard " stipulates that clearly what the pest density in the worm grain classification standard was added up is the quantity of worm of living, so need accurately detect the worm that lives.
The detection method of storage pest can divide the grain worm to separate and two links of grain worm identification.
Grain worm separation link mainly contains sample and crosses sieve method and mass trapping.Wherein, it is most widely used, the most traditional method of present China grain depot that sample is crossed sieve method, it be by distinguish layer fixed point artificial/electronic skewer gets grain samples, each check point extracts at least that 1kg grain manually sieves, the back artificial cognition of sieving.This method labour intensity is bigger, efficiency ratio is lower, but this owned by Francely gets worm in passive, as long as the grain worm that exists all can be removed, and with low cost.Mass trapping is to utilize trap or grain worm biological habit, traps as chemotaxis, photoaxis, hypsotaxis etc., and using many is the inserting tube trapper of band worm trapping hole.This method is subjected to that such environmental effects such as temperature worm kind big, trapping is more single, the unstable result measured and lack corresponding relation accurately with the screening pest density, needs to place a large amount of trappers in addition, and cost is than higher, thereby has limited its application.But it is less that the labour intensity of this method is lower, workload compares.
In the identification link of grain worm, artificial cognition method and machine vision method are arranged.
The artificial cognition method is traditional artificial differential method, this method is that the grain worm differentiates that the expert by microscope or directly utilize architectural feature and the color characteristic of insect by sense organ, differentiates if any no wing, head size, elytrum shape, speckle shape and color etc.Yao Wei etc. utilize grain storage pest to like selecting the characteristics of control environment to research and develop foodstuff pest pick-up unit (utility model patent number: 90209011.9 and patent of invention number: 92102125.9), after entering the pest-induced tube that has worm trapping hole when the grain worm, photoelectric pulse signal that produces when utilizing insect through special modality or insect fall to knocking the pulse signal that is produced, and by data acquisition system (DAS) insect are counted.Owing to this method does not have camera head, therefore must take out the ability artificial cognition by handle assembly, so its artificial cognition is relatively more difficult.Shen Dong etc. have researched and developed grain storage pest monitor (utility model patent number: 96232419.1), when the grain worm is entered grain worm monitor by worm trapping hole after, the optical system of scioptics combination is with grain worm amplification imaging, and a situation arises to make the staff pile inner grain worm from grain out-pile layer artificial visually examine to grain.02269339.4) and Grainhouse injurious insect detector (utility model patent number: 200420075415.7) Duan Jingzhi etc. have researched and developed Grainhouse injurious insect sampler (utility model patent number:, this device is drawn into the grain worm in the pest storage by the pipeline that links to each other with negative pressure source, carry out the insect counting by optic coupling element, by camera head the grain worm that enters in it is made a video recording, and view data sent to data processing unit, be convenient to the staff grain worm is discerned classification.
Above-mentioned all open file major parts all adopt the inserting tube trapper to lure collection grain worm, and some has also realized the automatic tally function of grain worm, but must carry out the artificial visual classification to the grain worm by the expert.In fact, artificial cognition grain worm needs the testing staff to have very professional taxonomy knowledge, and in addition, the build of grain worm own is very little, and kind is very many, and similarity is very high between some grain worm, therefore error in classification can occur inevitably.In addition, the efficiency ratio of artificial cognition is lower, is unfavorable for the robotization that the grain worm detects.
The machine vision method be adopt automatically/manually extract the grain sample, and obtain grain worm image automatically, use technology such as computer vision, Flame Image Process, pattern-recognition to discern the grain worm automatically then.This method has the accuracy height, labor capacity is little, efficient is high, grain worm image viewing, be convenient to and advantages such as the existing computer management system of grain depot is connected.Be the research focus in grain worm field during the nearly last ten years, the researchist has carried out a large amount of research in this respect always, has obtained very big progress.Utilizations such as He Guiming are inner is equipped with the feeler picked-up grain of camera and grain worm image and has developed grain insects in grain depot intelligent monitor system and method (patent of invention number: 01125651.6), this device judges whether have the grain worm to occur in the image by the threshold value of asking for difference image (background image and target image), can be to grain worm counting and statistics.Zhen Tong, Zhang Hongmei, Zhou Long etc. have studied the automatic classification of a kind or 3 kinds grain worm with digital camera or CCD, but all do not relate to the front end of grain worm recognition system---the automatic separation of grain worm.Qiu Daoyin, Zhang Hongtao etc. have developed grain worm on-line intelligence detection system at 9 class grain worms, system can be to grain worm real-time counting and classification automatically at present, but the worm that lives can duplicate the phenomenon of counting, and the grain sample does not have to sieve the efficient of directly carrying out image sampling and having influenced system in addition.
Machine vision detection method can be realized the automatic classification of grain worm, it is the direction that the grain worm detects development, but the multipotency of present Vision Builder for Automated Inspection carries out Classification and Identification to 9 class grain worms, can't distinguish " extremely " " work " of the worm of putting out cereal, can't overcome " seemingly-dead " problem, Shi Bie kind awaits further increase in addition.Therefore, be necessary to study storage pest automatic testing method and device thereof,, realize detecting automatically accurately in real time of common storage pest to determine kind and the quantity of efficient worm automatically.
Summary of the invention
The objective of the invention is to provide a kind of storage pest automatic testing method and device thereof based on visible light-near infrared binocular computer vision, can from the grain sample, effectively separate the worm of putting out cereal fast, and be transferred to binocular vision system automatically and carry out image acquisition, on the basis of visible images and near-infrared image information fusion, automatically determine kind and the quantity of efficient worm, realize that the real-time, accurate, automatic of common storage pest detects.
Technical scheme of the present invention is as follows:
Pick-up unit of the present invention partly is made up of grain worm separating part, grain worm hop and machine vision.Grain worm separating part and machine vision part all be installed in grain worm hop above.
The function that described grain worm separating part is realized is that the grain worm is separated from the grain sample quickly and efficiently, and removes the dust in the grain screen underflow automatically, mainly comprises feeder, screening motor, crank connecting link, sieve and dedusting mechanism.Existing grain worm Vision Builder for Automated Inspection does not possess electronic screening dedusting function, and grain worm separating part of the present invention detects automatically and designs at grain worm machine vision.The screening motor drives final drive shaft by the transmission of single V band, and the transmission shaft other end connects crank disc.Connecting rod one end is hinged on the crank disc, and the other end is hinged on the screen frame (unit).Sieve is fixed on the screen frame (unit), and screen frame (unit) adopts underslung type to be hinged on the frame, thereby forms the linear vibration screening mechanism.Feeder be installed in sieve directly over, dedusting mechanism be installed in sieve under, comprise dust cleaning case and be positioned at the dedusting fan and the dust collection bag of its both sides.Therefore, after the grain sample is poured feeder into, sieve, and utilize dust in the dedusting mechanism cleaning screen underflow by the motor-driven toggle of above-mentioned screening.
The function that described grain worm hop is realized accurately receives screen underflow, carry collecting cassette under the machine vision part for image acquisition, and the screen underflow in the cleaning collecting cassette, mainly comprise control module, collecting cassette, travelling belt, photoelectric sensor, collection box and transmission motor.Control module links to each other with photoelectric sensor, screening motor, dedusting motor, transmission motor and computing machine by data line, computing machine passes through the trigger pip of its reception from photoelectric sensor, and by it screening motor, dedusting motor and transmission motor is sent steering order.Travelling belt adopts double-stranded bar chain transmission transport sector, by the running of transmission driven by motor.The collecting cassette front end is hinged on the double-stranded bar, and the rear end is a free end, can rotate flexibly around hinged end.3 photoelectric sensors are installed in the below of dedusting mechanism, near infrared camera and visible light CCD respectively, the inboard of travelling belt, and be lower than the position of collecting cassette height slightly.When they detect collecting cassette move to dedusting mechanism, near infrared camera and visible light CCD under the time, the trigger pip of grain worm screening and image taking is provided to control module.When collecting cassette ran to extreme position, collecting cassette was carried out button and is stirred work, topples over the screen underflow in the collecting cassette, and was collected by the collection box that is installed under it.
Described machine vision part, major function are to gather and analyze the near-infrared image and the visible images of screen underflow, are made up of computing machine, lighting box, near infrared camera, visible light CCD, light source and corresponding software.Lighting box be installed in travelling belt directly over, there is the light source of the visible light-near-infrared band of even illumination the inside, diffuses for the screen underflow in the collecting cassette provides uniformly.Near infrared camera and visible light CCD are installed in the lighting box, be vertically mounted on travelling belt directly over, all with collecting cassette direction of motion in same plane, can distinguish the near infrared figure and the visible images of screen underflow in the clear shooting, collecting box.Two cameras link to each other with computing machine through data line, and the image of taking is transferred to computing machine.
Automatic testing method of the present invention is characterized in that:
(1) grain worm separating part sieves the grain sample automatically, and the dust in the grain screen underflow is removed in cleaning, and makes screen underflow fall into collecting cassette automatically.
(2) grain worm hop collecting cassette be sent to respectively near infrared camera and visible light CCD under, machine vision part is gathered near-infrared image and visible images respectively to the screen underflow in the collecting cassette, and transfers to computing machine.
(3) near-infrared image of being gathered is carried out Flame Image Process, obtain containing the binary image of grain worm worm alive, extract characteristic parameters such as its area, complexity, invariant moments, differentiate the worm of putting out cereal worm alive, and determine each worm coordinate information in image of living by grain worm worm identification software alive.Simultaneously, the number of all grain worms worm alive in the statistical picture.
(4) visible images of gathering is carried out Flame Image Process, the near-infrared image of coupling and fusion screen underflow, the worm alive of orienting motion.In conjunction with the coordinate information of above-mentioned all worms that live, determine the coordinate information of " seemingly-dead " worm alive in visible images, and be partitioned into the subregion of all grain worms worm alive in the visible images.
(5) extract the local feature parameters such as ratio of global feature parameter such as grain worm area, girth and elytrum length breadth ratio, elytrum length and grain polypide length, form and optimize the primitive character space of grain worm, utilization grain worm kind identification software determine the to put out cereal kind of information of worm worm alive.
(6) grain worm hop is after the screen underflow image acquisition finishes, and the screen underflow in the collecting cassette is removed in cleaning automatically, and enters next circulation.
Described Flame Image Process comprises Image Acquisition, goes background, filtering strengthens image, split image.
All include high-precision model of cognition in described grain worm worm identification software alive and the grain worm kind identification software, can pass through technology such as neural network classifier, support vector machine classifier or fuzzy classification device, the relational model of near infrared grain worm characteristics of image parameter that foundation is extracted and grain worm worm alive, and grain worm near infrared spectrum characteristic parameter and other relational model of grain insects, and guarantee that the Model Identification precision is more than 95%.
The intensity of light source in the lighting box and position can be adjusted, and choose the light source and the locus of suitable frequency spectrum, and make and form uniform illumination in the visual field of camera, make camera can obtain screen underflow image clearly.
Effect of the present invention is: (1) the present invention adopts visible light, near infrared Vision Builder for Automated Inspection that the grain worm is analyzed simultaneously, merge near-infrared image and the visible light image information of grain worm, can determine the more specific location information of efficient worm in visible images automatically, and utilizing the efficient worm of high resolving power visible images reliable recognition, this was not all relating in file in the past.(2) the present invention utilize visible images realized worm alive automatically, accurate counting, the counting accuracy rate is 100%, solved grain worm " seemingly-dead " phenomenon and detected grain worm that the grain worm the brought worm that lives automatically for the machine vision method to count an inaccurate difficult problem.(3) the present invention reaches more than 95% the real-time grading accuracy of worm alive by extracting the global feature and the local feature of grain worm visible images under the high resolving power.(4) the present invention combines fast automatic screening plant of grain worm and the Vision Builder for Automated Inspection that includes high speed processing software, has improved the efficient that the grain worm detects, and has reduced labour intensity.
Description of drawings
Fig. 1 is a structural representation of the present invention;
Fig. 2 is the dedusting mechanism structural representation;
Among the figure, the 1-crank connecting link, 2-sieves motor, 3-control module, the 4-feeder, 5-sieve, 6-light source, 7-dedusting fan, the 8-dust cleaning case, 9-lighting box, 10-optical filter, 11-near infrared camera, 12-visible light CCD, the 13-computing machine, 14-transmits motor, and 15-collects box, 16-optoelectronic switch 1,17-optoelectronic switch 2,18-travelling belt, 19-optoelectronic switch 3,20-collecting cassette, 21-dust collection bag.
Embodiment
Below in conjunction with Fig. 1 concrete enforcement of the present invention is described.Pick-up unit of the present invention partly is made up of grain worm separating part, grain worm hop and machine vision.Grain worm separating part and machine vision part all be installed in grain worm hop above.It is characterized in that:
The function that described grain worm separating part is realized is that the grain worm is separated from the grain sample quickly and efficiently, and removes the dust in the grain screen underflow automatically, mainly comprises feeder 4, screening motor 2, crank connecting link 1, sieve 5 and dedusting mechanism.Screening motor 2 drives final drive shaft by the transmission of single V band.The final drive shaft other end connects crank disc.Connecting rod one end is hinged on the crank disc, and the other end is hinged on the screen frame (unit).Sieve 5 is fixed on the screen frame (unit), and screen frame (unit) adopts underslung type to be hinged on the frame, thereby forms the linear vibration screening mechanism.Feeder 4 be installed in sieve 5 directly over, dedusting mechanism be installed in sieve 5 under, comprise dust cleaning case 8, dedusting fan 7 and dust collection bag 21, wherein, dedusting fan 7 and dust collection bag 21 are installed in the both sides of dust cleaning case 8.Therefore, after the grain sample is poured feeder 4 into, sieve, and utilize dust in the dedusting mechanism cleaning screen underflow by the motor-driven toggle of above-mentioned screening.
The function that described grain worm hop is realized accurately receives screen underflow, carry collecting cassette 20 under the machine vision part for image acquisition, and the screen underflow in the cleaning collecting cassette 20, mainly comprise control module 3, collecting cassette 20, travelling belt 18, photoelectric sensor 1, photoelectric sensor 2, photoelectric sensor 3, collect box 15 and transmission motor 14.Control module 3 links to each other with photoelectric sensor 1, photoelectric sensor 2, photoelectric sensor 3, screening motor 2, dedusting motor 7, transmission motor 14 and computing machine 13 by data line, computing machine 13 passes through the trigger pip of its reception from photoelectric sensor 1, photoelectric sensor 2, photoelectric sensor 3, and by it screening motor 2, dedusting motor 7 and transmission motor 14 is sent steering order.Travelling belt 18 adopts double-stranded bar chain transmission transport sector, drives running by transmission motor 14.Collecting cassette 20 front ends are hinged on the double-stranded bar, and the rear end is a free end, can rotate flexibly.3 photoelectric sensors are installed in the below of dedusting mechanism, near infrared camera 11 and visible light CCD12 respectively, the inboard of travelling belt 18, and be lower than collecting cassette 20 position highly slightly.When they detect collecting cassette 20 move to dedusting mechanism, near infrared camera 11 and visible light CCD12 under the time, the trigger pip of screening of grain worm and image taking is provided for control module 3.When collecting cassette 20 ran to extreme position, collecting cassette 20 was carried out button and is stirred work, topples over the screen underflow in the collecting cassette 20, and was collected by the collection box 15 that is installed under it.
Described machine vision part, major function are to gather and analyze the near-infrared image and the visible images of screen underflow, are made up of computing machine 13, lighting box 9, near infrared camera 11, visible light CCD12, light source 6 and corresponding software.Lighting box 9 be installed in travelling belt 18 directly over, there is the light source of the visible light-near-infrared band of even illumination the inside, diffuses for the screen underflow in the collecting cassette 20 provides uniformly.Near infrared camera 11 and visible light CCD12 are installed in the lighting box 9, be vertically mounted on travelling belt 18 directly over, all with collecting cassette 20 direction of motion in same plane, can distinguish the near infrared figure and the visible images of screen underflow in the clear shooting, collecting box 20.The spectral range of near infrared camera 11 is 900-1700nm, and its front end has near infrared filter 10 (half-band width is less than 10nm).Two cameras link to each other with computing machine 13 through data line, and the image of taking is transferred to computing machine 13.
During work, after the grain sample was sent into feeder 4, computing machine 13 sent steering order for control module 3 and starts screening motor 2, drives the vibration of sieve 5 by crank connecting link 1.Through dedusting mechanism, dedusting fan 7 blows to the dust in the screen underflow in the dust collection bag 21 when screen underflow falls, and the screen underflow after the dedusting falls into the collecting cassette 20 of dedusting mechanism below.Transmission motor 14 drives travelling belts 18 and rotates, and after photoelectric sensor 17 detected collecting cassette 20 and puts in place, computing machine 13 control near infrared cameras 11 carried out image acquisition, gather finish after, travelling belt 18 moves on.After photoelectric sensor 16 detected collecting cassette 20 and puts in place, computing machine 13 control visible light CCD12 carried out image acquisition, gather finish after, travelling belt 18 moves on.The near-infrared image of being gathered is carried out Flame Image Process, obtain containing the binary image of grain worm worm alive, extract morphological feature parameters such as its area, complexity, invariant moments, differentiate the worm of putting out cereal worm alive, and determine each worm coordinate information in image of living by identification software.Simultaneously, the number of all grain worms worm alive in the statistical picture.The visible images of gathering is carried out Flame Image Process, the near-infrared image of coupling and fusion screen underflow, the worm alive of orienting motion.In conjunction with the coordinate information of above-mentioned all worms that live, determine " seemingly-dead " worm coordinate information in visible images of living, and be partitioned into live subregion in the visible images at worm places of all grain worms.Extract global feature such as grain worm area, girth and elytrum length breadth ratio, elytrum length is learned characteristic parameter with the local forms such as ratio of grain polypide length, forms and optimize the primitive character space of grain worm, utilization identification software determine the to put out cereal kind of information of worm worm alive.Spin upside down when travelling belt 18 moves to rightmost, screen underflow falls into automatically collects box 15.After photoelectric sensor 19 detected collecting cassette and puts in place, 14 stalls of transmission motor detected to wait for next time.
Claims (4)
1. based on the grain worm pick-up unit of visible light-near infrared binocular machine vision, it is characterized in that pick-up unit of the present invention is made up of grain worm separating part, grain worm hop, machine vision part and computing machine (13).Grain worm separating part and machine vision part all be installed in grain worm hop above;
Described grain worm separating part comprises feeder (4), screening motor (2), crank connecting link (1), sieve (5), dedusting fan (7), dust cleaning case (8) and dust collection bag (21); Described screening motor (2) drives crank connecting link (1) by final drive shaft, and crank connecting link (1) is hinged with screen frame (unit), and sieve (5) is fixed on the screen frame (unit), and screen frame (unit) adopts underslung type to be hinged on the frame; Described feeder (4) be installed in sieve (5) directly over, described dust cleaning case (8) be installed in sieve (5) under, the both sides of dust cleaning case (8) are provided with dedusting fan (7) and dust collection bag (21);
Described grain worm hop comprises control module () 3, collecting cassette (20), travelling belt (18), photoelectric sensor (1), photoelectric sensor (2), photoelectric sensor (3), collects box (15) and transmission motor (14); Described transmission motor (14) drives travelling belt (18), and travelling belt (18) adopts double-stranded bar chain transmission transport sector; Described collecting cassette (20) front end is hinged on the double-stranded bar, and the rear end is a free end; 3 photoelectric sensors are installed in the below of dedusting mechanism, near infrared camera (11) and visible light CCD (12) respectively, and the inboard of travelling belt (18) is lower than collecting cassette (20) position highly; Collect box (15) be positioned at travelling belt (18) terminal under; Described photoelectric sensor (1), photoelectric sensor (2), photoelectric sensor (3), screening motor (2), dedusting motor (7), transmission motor (14) link to each other with described control module (3) by data line, control module (3) links to each other with described computing machine (13), computing machine (13) is by the trigger pip of control module (3) reception from photoelectric sensor (1), photoelectric sensor (2), photoelectric sensor (3), and computing machine (13) sends steering order by control module (3) to screening motor (2), dedusting motor (7) and transmission motor (14).
Described machine vision part is made up of lighting box (9), near infrared camera (11), visible light CCD (12), light source (6) and corresponding software; Described lighting box (9) be installed in travelling belt (18) directly over, described light source (6), near infrared camera (11) and visible light CCD (12) are positioned at lighting box (9); Near infrared camera (11) and visible light CCD (12) be vertically mounted on travelling belt (18) directly over, with collecting cassette (20) direction of motion in same plane, link to each other with described computing machine (13) through data line; The spectral range of near infrared camera (11) is 900-1700nm, and its front end has near infrared filter (10), and infrared fileter (10) half-band width is less than 10nm.
2. implement the described grain worm pick-up unit method of claim 1, it is characterized in that the step that comprises is based on visible light-near infrared binocular machine vision:
(1) grain worm separating part sieves the grain sample automatically, and the dust in the grain screen underflow is removed in cleaning, and makes screen underflow fall into collecting cassette automatically.
(2) grain worm hop collecting cassette be sent to respectively near infrared camera and visible light CCD under, machine vision part is gathered near-infrared image and visible images respectively to the screen underflow in the collecting cassette, and transfers to computing machine;
(3) near-infrared image of being gathered is carried out Flame Image Process, obtain containing the binary image of grain worm worm alive, extract characteristic parameters such as its area, complexity, invariant moments, differentiate the worm of putting out cereal worm alive, and determine each worm coordinate information in image of living by grain worm worm identification software alive; Simultaneously, the number of all grain worms worm alive in the statistical picture;
(4) visible images of gathering is carried out Flame Image Process, the near-infrared image of coupling and fusion screen underflow, the worm alive of orienting motion; In conjunction with the coordinate information of above-mentioned all worms that live, determine the coordinate information of " seemingly-dead " worm alive in visible images, and be partitioned into the subregion of all grain worms worm alive in the visible images;
(5) extract the local feature parameters such as ratio of global feature parameter such as grain worm area, girth and elytrum length breadth ratio, elytrum length and grain polypide length, form and optimize the primitive character space of grain worm, utilization grain worm kind identification software determine the to put out cereal kind of information of worm worm alive;
(6) grain worm hop is after the screen underflow image acquisition finishes, and the screen underflow in the collecting cassette is removed in cleaning automatically, and enters next circulation.
3. the grain worm pick-up unit method based on visible light-near infrared binocular machine vision according to claim 2 is characterized in that, the described Flame Image Process of step (3) comprises Image Acquisition, goes background, filtering strengthens image, split image.
4. the grain worm pick-up unit method based on visible light-near infrared binocular machine vision according to claim 2, it is characterized in that, all include high-precision model of cognition by technology such as neural network classifier, support vector machine classifier or fuzzy classification devices in described grain worm worm identification software alive and the grain worm kind identification software, the relational model of near infrared grain worm characteristics of image parameter that foundation is extracted and grain worm worm alive, and grain worm near infrared spectrum characteristic parameter and other relational model of grain insects.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009102350757A CN101701915B (en) | 2009-11-13 | 2009-11-13 | Device and method for detecting stored-grain insects based on visible light-near infrared binocular machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009102350757A CN101701915B (en) | 2009-11-13 | 2009-11-13 | Device and method for detecting stored-grain insects based on visible light-near infrared binocular machine vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101701915A true CN101701915A (en) | 2010-05-05 |
| CN101701915B CN101701915B (en) | 2011-08-10 |
Family
ID=42156835
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009102350757A Expired - Fee Related CN101701915B (en) | 2009-11-13 | 2009-11-13 | Device and method for detecting stored-grain insects based on visible light-near infrared binocular machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101701915B (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103177266A (en) * | 2013-04-07 | 2013-06-26 | 青岛科技大学 | Intelligent stock pest identification system |
| CN104155312A (en) * | 2014-08-11 | 2014-11-19 | 华北水利水电大学 | Method for detecting pests in food grains based on near infrared machine vision, and apparatus thereof |
| CN104749189A (en) * | 2015-02-28 | 2015-07-01 | 华北水利水电大学 | Grain interior insect pest detection device based on multi-spectral imaging technology |
| CN105021563A (en) * | 2015-07-14 | 2015-11-04 | 河南科技大学 | Tobacco information acquisition device based on near infrared spectroscopy |
| CN105300991A (en) * | 2015-10-15 | 2016-02-03 | 国家粮食局科学研究院 | Biological detection system and method for screening insecticidal active substances |
| CN106596207A (en) * | 2016-12-23 | 2017-04-26 | 郑州贝博电子股份有限公司 | Grain condition detection probe |
| CN107593200A (en) * | 2017-10-31 | 2018-01-19 | 河北工业大学 | A kind of trees plant protection system and method based on visible ray infrared technique |
| US20180068164A1 (en) * | 2016-09-08 | 2018-03-08 | Wal-Mart Stores, Inc. | Systems and methods for identifying pests in crop-containing areas via unmanned vehicles |
| CN107894419A (en) * | 2017-12-29 | 2018-04-10 | 南京艾龙信息科技有限公司 | A kind of original grain pest detection means and method |
| CN107909575A (en) * | 2017-12-30 | 2018-04-13 | 煤炭科学研究总院唐山研究院 | For the binocular vision on-line measuring device and detection method of vibrating screen operating status |
| CN108240982A (en) * | 2017-12-30 | 2018-07-03 | 天津博硕东创科技发展有限公司 | A kind of automatic processing device of solid waste |
| CN108362326A (en) * | 2018-01-03 | 2018-08-03 | 江苏大学 | A kind of outstanding rail greenhouse integrated information automatic cruising monitoring device |
| CN111765974A (en) * | 2020-07-07 | 2020-10-13 | 中国环境科学研究院 | Wild animal observation system and method based on miniature refrigeration thermal infrared imager |
| CN111830029A (en) * | 2020-01-02 | 2020-10-27 | 刘金涛 | Pesticide preparation concentration field analysis system and method |
| CN114030907A (en) * | 2022-01-10 | 2022-02-11 | 安徽高哲信息技术有限公司 | Feeding system |
| CN114063179A (en) * | 2021-10-27 | 2022-02-18 | 福建省农业科学院植物保护研究所 | Biological cross-border intelligent monitoring equipment of invasion |
| CN115035515A (en) * | 2022-06-15 | 2022-09-09 | 电子科技大学 | Nang identification method |
| CN116758467A (en) * | 2023-05-05 | 2023-09-15 | 广州白云国际机场建设发展有限公司 | Monitoring alarm method and device in civil aviation security equipment field |
| CN117664975A (en) * | 2024-02-01 | 2024-03-08 | 黑龙江八一农垦大学 | Corn quality detection device and method based on machine vision |
-
2009
- 2009-11-13 CN CN2009102350757A patent/CN101701915B/en not_active Expired - Fee Related
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103177266A (en) * | 2013-04-07 | 2013-06-26 | 青岛科技大学 | Intelligent stock pest identification system |
| CN104155312B (en) * | 2014-08-11 | 2017-06-16 | 华北水利水电大学 | Grain intragranular portion pest detection method and apparatus based on near-infrared computer vision |
| CN104155312A (en) * | 2014-08-11 | 2014-11-19 | 华北水利水电大学 | Method for detecting pests in food grains based on near infrared machine vision, and apparatus thereof |
| CN104749189A (en) * | 2015-02-28 | 2015-07-01 | 华北水利水电大学 | Grain interior insect pest detection device based on multi-spectral imaging technology |
| CN105021563A (en) * | 2015-07-14 | 2015-11-04 | 河南科技大学 | Tobacco information acquisition device based on near infrared spectroscopy |
| CN105300991A (en) * | 2015-10-15 | 2016-02-03 | 国家粮食局科学研究院 | Biological detection system and method for screening insecticidal active substances |
| CN105300991B (en) * | 2015-10-15 | 2018-06-19 | 国家粮食局科学研究院 | A kind of biological detection system and detection method for insecticide active substance screening |
| US20180068164A1 (en) * | 2016-09-08 | 2018-03-08 | Wal-Mart Stores, Inc. | Systems and methods for identifying pests in crop-containing areas via unmanned vehicles |
| CN106596207A (en) * | 2016-12-23 | 2017-04-26 | 郑州贝博电子股份有限公司 | Grain condition detection probe |
| CN106596207B (en) * | 2016-12-23 | 2023-05-05 | 郑州贝博电子股份有限公司 | Grain condition detecting probe |
| CN107593200A (en) * | 2017-10-31 | 2018-01-19 | 河北工业大学 | A kind of trees plant protection system and method based on visible ray infrared technique |
| CN107593200B (en) * | 2017-10-31 | 2022-05-27 | 河北工业大学 | Tree plant protection system and method based on visible light-infrared technology |
| CN107894419A (en) * | 2017-12-29 | 2018-04-10 | 南京艾龙信息科技有限公司 | A kind of original grain pest detection means and method |
| CN107894419B (en) * | 2017-12-29 | 2023-10-27 | 南京艾龙信息科技有限公司 | Device and method for detecting raw grain pests |
| CN107909575A (en) * | 2017-12-30 | 2018-04-13 | 煤炭科学研究总院唐山研究院 | For the binocular vision on-line measuring device and detection method of vibrating screen operating status |
| CN107909575B (en) * | 2017-12-30 | 2023-09-15 | 煤炭科学研究总院唐山研究院 | Binocular vision on-line detection device and detection method for running state of vibrating screen |
| CN108240982A (en) * | 2017-12-30 | 2018-07-03 | 天津博硕东创科技发展有限公司 | A kind of automatic processing device of solid waste |
| CN108362326B (en) * | 2018-01-03 | 2020-12-18 | 江苏大学 | Suspension rail type greenhouse comprehensive information automatic cruise monitoring device |
| CN108362326A (en) * | 2018-01-03 | 2018-08-03 | 江苏大学 | A kind of outstanding rail greenhouse integrated information automatic cruising monitoring device |
| CN111830029A (en) * | 2020-01-02 | 2020-10-27 | 刘金涛 | Pesticide preparation concentration field analysis system and method |
| CN111830029B (en) * | 2020-01-02 | 2023-10-20 | 河北盛鹏化工有限公司 | Pesticide preparation concentration on-site analysis system and method |
| CN111765974A (en) * | 2020-07-07 | 2020-10-13 | 中国环境科学研究院 | Wild animal observation system and method based on miniature refrigeration thermal infrared imager |
| CN111765974B (en) * | 2020-07-07 | 2021-04-13 | 中国环境科学研究院 | Wild animal observation system and method based on miniature refrigeration thermal infrared imager |
| CN114063179A (en) * | 2021-10-27 | 2022-02-18 | 福建省农业科学院植物保护研究所 | Biological cross-border intelligent monitoring equipment of invasion |
| CN114063179B (en) * | 2021-10-27 | 2023-10-27 | 福建省农业科学院植物保护研究所 | Intelligent monitoring equipment for invasion organism cross-border |
| CN114030907A (en) * | 2022-01-10 | 2022-02-11 | 安徽高哲信息技术有限公司 | Feeding system |
| CN115035515A (en) * | 2022-06-15 | 2022-09-09 | 电子科技大学 | Nang identification method |
| CN116758467A (en) * | 2023-05-05 | 2023-09-15 | 广州白云国际机场建设发展有限公司 | Monitoring alarm method and device in civil aviation security equipment field |
| CN117664975A (en) * | 2024-02-01 | 2024-03-08 | 黑龙江八一农垦大学 | Corn quality detection device and method based on machine vision |
| CN117664975B (en) * | 2024-02-01 | 2024-04-26 | 黑龙江八一农垦大学 | Corn quality detection device and method based on machine vision |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101701915B (en) | 2011-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101701915B (en) | Device and method for detecting stored-grain insects based on visible light-near infrared binocular machine vision | |
| CN101701906A (en) | Method and device for detecting stored-grain insects based on near infrared super-spectral imaging technology | |
| Bjerge et al. | A computer vision system to monitor the infestation level of Varroa destructor in a honeybee colony | |
| CN104155312B (en) | Grain intragranular portion pest detection method and apparatus based on near-infrared computer vision | |
| US10664726B2 (en) | Grain quality monitoring | |
| CN107578089A (en) | A kind of crops lamp lures the automatic identification and method of counting for observing and predicting insect | |
| CN105938571B (en) | Insect identifies number system and method | |
| US7916951B2 (en) | System and method for detecting and classifying objects in images, such as insects and other arthropods | |
| CN103179993B (en) | Insecticide real-time monitoring device | |
| Chen et al. | A deep learning-based vision system combining detection and tracking for fast on-line citrus sorting | |
| Pegoraro et al. | Automated video monitoring of insect pollinators in the field | |
| CN104813993B (en) | Small agricultural pests automated watch-keeping facility and method based on machine vision | |
| CN107094729A (en) | The machine visual detection device and method of counting of insect inside silo | |
| CN203324781U (en) | Pest trapping apparatus and pest remote identifying and monitoring system | |
| CN104949998A (en) | Online visual inspection device and method for surface dirt of group origin eggs | |
| WO2011041924A1 (en) | Device and method for identifying ripe oranges in nature scene by filter spectral image technology | |
| Annamalai et al. | Citrus yield mapping system using machine vision | |
| CN112106747A (en) | Intelligent agricultural insect pest remote automatic monitoring system | |
| König | IndusBee 4.0–integrated intelligent sensory systems for advanced bee hive instrumentation and hive keepers' assistance systems | |
| Giakoumoglou et al. | White flies and black aphids detection in field vegetable crops using deep learning | |
| de Castro Pereira et al. | Detection and classification of whiteflies and development stages on soybean leaves images using an improved deep learning strategy | |
| Cao et al. | Automated Pollen Detection with an Affordable Technology. | |
| CN112580604A (en) | Mouse condition monitoring method and system and intelligent terminal | |
| CN208639408U (en) | The small pest that migrates traps monitoring device | |
| Miranda et al. | Pest identification using image processing techniques in detecting image pattern through neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20110810 Termination date: 20161113 |