CN104902228A - Insect real-time monitoring device and method based on computer vision - Google Patents
Insect real-time monitoring device and method based on computer vision Download PDFInfo
- Publication number
- CN104902228A CN104902228A CN201510249053.1A CN201510249053A CN104902228A CN 104902228 A CN104902228 A CN 104902228A CN 201510249053 A CN201510249053 A CN 201510249053A CN 104902228 A CN104902228 A CN 104902228A
- Authority
- CN
- China
- Prior art keywords
- insect
- camera
- time monitoring
- sampling channel
- computer vision
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Catching Or Destruction (AREA)
Abstract
The invention discloses an insect real-time monitoring device and method based on computer vision, and the device comprises an insect killing apparatus, wherein the bottom of the insect killing apparatus is provided with an insect falling port; an insect collector which is located under the insect falling port; a sampling channel which is connected between the insect falling port and an inlet of the insect collector; a camera which is disposed on a side wall of the sampling channel and used for collecting the real-time image of fallen insects; and a terminal machine which is used for receiving the image from the camera and carrying out recognition. The device and method solves the problems that the collection of an agriculture insect image in the prior art is large in difficulty and is large in workload, and the posture of samples is small in number.
Description
Technical field
The present invention relates to agricultural engineering technology field, particularly relate to a kind of insect real time monitoring apparatus based on computer vision and method.
Background technology
China is a large agricultural country, and the monitoring of agricultural pests, the statistical fluctuation work of insect pest situation disaster are very important.If monitoring and prediction is accurately and timely, just can early Control pests, reduce pesticide dosage, avoid crops to suffer heavy losses.Be generally by experienced peasant and classification of insect expert, insect is identified in conventional method, but manual identified labour intensity is large, efficiency is low.
Therefore, develop some intelligent wireless insect remote automatic monitoring devices, by contributing to the accuracy rate and the efficiency that improve insect identification and counting, reduce the loss that insect pest brings, and then promote the enforcement of precision agriculture, improve the science popularization level of insect knowledge.
In prior art, be generally on the basis that insect is traped, realize identification to insect and remote monitoring.At present, based in the land for growing field crops insect pest situation monitoring of computer vision, can to identify and insect counts to pest species, be a kind of conventional remote control technology.
Land for growing field crops insect pest situation monitoring technique based on computer vision needs to gather a large amount of insect images as sample, carries out the identification of pest species.At present, the collection of agricultural pests sample also mainly relies on artificial shooting, and the angle of shooting is single, can not cover the belly of insect, side angularly, causes the image pattern collected in Intelligent Recognition, very easily produce the phenomenon of over-fitting; And the workload of artificial shooting is large, condition is arduous, and the collection for comprehensive each angle of one group of 20 image pattern often needs nearly two hours.These problems, cause a lot of difficulty to the insect image intelligent Study of recognition in intelligent agriculture process.
Summary of the invention
In order to overcome above-mentioned current agricultural pests IMAQ difficulty, workload is large, and the problem that sample attitude is single, the invention provides a kind of insect real time monitoring apparatus based on computer vision and method.
Based on an insect real time monitoring apparatus for computer vision, comprising:
Insect-killing device, bottom is provided with insect population;
Pest catcher, is positioned at immediately below insect population;
Sampling channel, is connected between insect population and pest catcher entrance;
Camera, is arranged on sampling channel sidewall in order to gather the realtime graphic of worm;
Terminating machine, identifies for the image that receives from camera.
Described sampling channel surrounds hollow triangular prism shape by three blocks of side plates, and described camera is arranged on wherein on side plate, and the inwall of other two blocks of side plates is minute surface.
Insect, after insect-killing device trapping, falls into pest catcher through sampling channel.Dead worm, by the process of sampling channel, is installed in the camera collection image of sampling channel sidewall, and the image of collection carries out the kind that processes to identify dead worm and number through terminating machine.Sampling channel is triangular prism shape, camera is arranged on one of them side plate, two the side plate inside relative with camera are minute surface, such camera can gather dead worm just to the image of camera side, again by the reflection of two minute surfaces, collect the image of dead worm opposite side, can comprehensively, multi-angle IMAQ is carried out to dead worm.
Be provided with the transparent baffle surrounding camera in the installation place of camera, make camera keep clean.The material of transparent baffle is rigid plastics.
As preferably, the angle between two minute surfaces is 45 ° ~ 75 °.
Described minute surface is in the monitoring range of camera.
In order to ensure camera gather the effect of image, be provided with light source in described sampling channel.In order to ensure that dead worm can fall in pest catcher smoothly, described light source is LED ring-type light source.For improving illuminating effect, above and below minute surface, be equipped with above-mentioned ring-type light source.
The lower end of sampling channel connects pest catcher, and described pest catcher comprises:
Be fixed on the housing bottom sampling channel,
The air permeable plate in the housing housing being divided into upper and lower two chambers is installed;
Be arranged on the induced draft fan of bottom chamber, in order to form negative pressure in upper chamber;
What the outlet being positioned at upper chamber and described sampling channel connected fall worm box, this worm box that falls is provided with negative pressure drainage hole.
Pest catcher is divided into upper and lower two chambers, centre is separated by air permeable plate, and the induced draft fan of bottom chamber makes to form negative pressure state in whole device, makes dead worm can enter worm box from insect-killing device through sampling channel smoothly, and avoid the adhesion on insect-killing device of dead worm, also keep the unimpeded of sampling channel simultaneously.
Radio communication between described terminating machine and camera.
Present invention also offers a kind of method for real-time monitoring based on above-mentioned insect real time monitoring apparatus, comprising:
Utilize described insect-killing device insect-killing trapping;
Dead worm falls into pest catcher through dropping down onto sampling channel, the realtime graphic of the worm that fallen by camera collection in dead worm dropping process;
Described terminating machine receives the image from each camera and identifies.
Camera of the present invention is high-speed camera head, carries out IMAQ to the dead worm capture video through sampling channel, and the image of collection, through radioing to terminating machine, identifies the kind of dead worm through terminating machine process.
Compared with prior art, beneficial effect of the present invention is:
Through the insect of insect-killing device trapping, at the sampling channel through triangular prism shape, by being located at the reflex of the minute surface on sampling channel inside, camera can carry out multi-facetedly comprehensively carrying out IMAQ to dead worm.The present invention can automatically capture the image pattern of insect multi-pose, solves the angle changing rate of artificial collecting sample single, the problems such as workload is larger.
Accompanying drawing explanation
Fig. 1 is the structural representation of the insect real time monitoring apparatus that the present invention is based on computer vision;
Fig. 2 is the partial enlarged drawing of Fig. 1;
Fig. 3 is the structural representation of sampling channel of the present invention;
Fig. 4 is the structural representation of pest catcher of the present invention.
Wherein, 1, trapping lamp; 2, sampling channel; 21, minute surface; 22, LED ring-type light source; 23, transparent baffle; 24, shooting mouth; 3, camera; 4, pest catcher; 41, fall worm box; 42, induced draft fan; 5, image processing terminal.
Embodiment
Below in conjunction with drawings and Examples, the present invention is described in further detail, it is pointed out that the following stated embodiment is intended to be convenient to the understanding of the present invention, and any restriction effect is not play to it.
As shown in Figure 1, a kind of insect real time monitoring apparatus based on computer vision, comprises trapping lamp 1, sampling channel 2, camera 3, pest catcher 4, image processing terminal 5.
Trapping lamp 1 in the present embodiment is frequency ventilating type solar insect-killing light, is directly the energy with solar energy.The outlet of the collection worm dish of frequency ventilating type solar insect-killing light lower end is connected with the upper end open of sampling channel 2, and sampling channel 2 lower ending opening is connected with pest catcher 4.Each junction can be connected by screw thread or other existing technology, the present embodiment is in order to easy for installation, shape that the interface of all parts is designed to mutually to mate (cross section of each junction is the shape of the comparison rule such as circular, oval), is connected by the junction button of hood type.
The sidewall of sampling channel is provided with high-speed camera head 3, can record a video to the dead worm by sampling channel 2, and the image document of shooting to image processing terminal 5 by wireless or definite transmission, is processed through image processing terminal 5 pairs of insect images, identifies.
As shown in Figures 2 and 3, sampling channel 2 surrounds hollow triangular prism shape by three blocks of side plates, one block of side plate offers shooting mouth 24, it is inner that camera 3 extend into sampling channel 2 by shooting mouth 24, the inwall at this place of side plate is provided with transparent baffle 23 (can be transparent plastic or glass), camera 3 is surrounded, in order to avoid the insect in sampling channel 2 pollutes camera 3, affects shooting effect.The another two block side plates relative with camera 3 are provided with minute surface 21, minute surface 21 is in the monitoring range of camera 3, angle between two minute surfaces 21 is 60 °, like this, when being entered over-sampling passage 2 by the insect that trapping lamp 1 kills, entered the reflex of minute surface 21, camera 3 can take the omnibearing image information of lower insect.In order to strengthen the shooting effect of camera 3, above and below minute surface 21, be respectively equipped with LED ring-type light source 22.
Camera 3 and controller electrical connection.Controller is fixed on the outer wall of sampling channel, and controller is provided with RAM (random access memory) card and wireless network card.The chip of controller processes the video that camera 3 is taken, and extracts the frame of video having insect image, intercepts the image of insect, is stored by RAM (random access memory) card.Image information in RAM (random access memory) card is derived by data-interface scene or wireless network card is transferred to image processing terminal 5.
The lower end of sampling channel 2 is connected with pest catcher 4.As shown in Figure 4, pest catcher 4 surrounds upper and lower two chambers by housing, is separated between two chambers by air permeable plate.Bottom chamber is provided with induced draft fan 42, in order to make to form negative pressure in trapping lamp 1, sampling channel 2, upper chamber; Be provided with the fall worm box 41 relative with the outlet of sampling channel 2 in upper chamber, the upper box wall of the worm box 41 that falls offers negative pressure drainage hole.Due to the effect of induced draft fan 42, form negative pressure in the inside of the trapping lamp 1 be connected, sampling channel 2, worm collecting box, avoid dead worm to stick on trapping lamp 1, and ensure that dead worm can not blocked by sampling channel 2 smoothly.Dead worm falls into the bottom of worm box 41, and the negative pressure drainage hole of the worm box 41 that falls is opened on box wall, and negative pressure drainage hole is not blocked by dead worm, ensure that in worm box 41 normally to form negative pressure.
Method for real-time monitoring based on above-mentioned insect real time monitoring apparatus is: utilize above-mentioned trapping lamp 1 trapping pests, the insect be entrapped falls in pest catcher 4 through sampling channel 2.When dead worm is by sampling channel 2, camera 3 is to its capture video, and the video of shooting, through chip process, extracts the frame of video having insect image, and intercept the image of insect, truncated picture processes through image processing terminal 5 and carries out the identification of pest species.
Image processing terminal 5 can utilize existing video procession technology process the insect image collected and identify.
Above-described embodiment has been described in detail technical scheme of the present invention and beneficial effect; be understood that and the foregoing is only specific embodiments of the invention; be not limited to the present invention; all make in spirit of the present invention any amendment, supplement and equivalent to replace, all should be included within protection scope of the present invention.
Claims (8)
1., based on an insect real time monitoring apparatus for computer vision, it is characterized in that, comprising:
Insect-killing device, bottom is provided with insect population;
Pest catcher, is positioned at immediately below insect population;
Sampling channel, is connected between insect population and pest catcher entrance;
Camera, is arranged on sampling channel sidewall in order to gather the realtime graphic of worm;
Terminating machine, identifies for the image that receives from camera.
2. the insect real time monitoring apparatus based on computer vision according to claim 1, it is characterized in that, described sampling channel surrounds hollow triangular prism shape by three blocks of side plates, and described camera is arranged on wherein on side plate, and the inwall of other two blocks of side plates is minute surface.
3. the insect real time monitoring apparatus based on computer vision according to claim 2, is characterized in that, the angle between two minute surfaces is 45 ° ~ 75 °.
4. the insect real time monitoring apparatus based on computer vision according to claim 2, it is characterized in that, described minute surface is in the monitoring range of camera.
5. the insect real time monitoring apparatus based on computer vision according to claim 1, is characterized in that, be provided with light source in described sampling channel.
6. the insect real time monitoring apparatus based on computer vision according to claim 1, it is characterized in that, described pest catcher comprises:
Be fixed on the housing bottom sampling channel;
The air permeable plate in the housing housing being divided into upper and lower two chambers is installed;
Be arranged on the induced draft fan of bottom chamber, in order to form negative pressure in upper chamber;
What the outlet being positioned at upper chamber and described sampling channel connected fall worm box, this worm box that falls is provided with negative pressure drainage hole.
7. the insect real time monitoring apparatus based on computer vision according to claim 1, is characterized in that, radio communication between described terminating machine and camera.
8., based on a method for real-time monitoring for the insect real time monitoring apparatus described in any one of claim 1 ~ 7, it is characterized in that, comprising:
Utilize described insect-killing device insect-killing trapping;
Dead worm falls into pest catcher through dropping down onto sampling channel, the realtime graphic of the worm that fallen by camera collection in dead worm dropping process;
Described terminating machine receives the image from each camera and identifies.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510249053.1A CN104902228A (en) | 2015-05-15 | 2015-05-15 | Insect real-time monitoring device and method based on computer vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510249053.1A CN104902228A (en) | 2015-05-15 | 2015-05-15 | Insect real-time monitoring device and method based on computer vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104902228A true CN104902228A (en) | 2015-09-09 |
Family
ID=54034591
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510249053.1A Pending CN104902228A (en) | 2015-05-15 | 2015-05-15 | Insect real-time monitoring device and method based on computer vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104902228A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106489876A (en) * | 2016-09-28 | 2017-03-15 | 浙江大学 | A kind of Agriculture field pests monitoring trap based on high light spectrum image-forming |
| CN107543829A (en) * | 2016-06-29 | 2018-01-05 | 北京邮电大学 | Stored-grain insects detection system |
| CN107996537A (en) * | 2017-12-11 | 2018-05-08 | 江苏飞视文化发展有限公司 | A kind of device for being used to monitor silo China Oil and Food Import and Export Corporation worm |
| CN108651410A (en) * | 2018-05-18 | 2018-10-16 | 仲恺农业工程学院 | A kind of learning-oriented pest control equipment |
| CN110363103A (en) * | 2019-06-24 | 2019-10-22 | 仲恺农业工程学院 | Identifying pest method, apparatus, computer equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1765701A (en) * | 2004-10-26 | 2006-05-03 | 北京航空航天大学 | Device for observing high-flap-rate insect flying motion |
| CN202335307U (en) * | 2011-11-18 | 2012-07-18 | 扬州绿源生物化工有限公司 | Injurious insect trap |
| CN103299969A (en) * | 2013-06-09 | 2013-09-18 | 浙江大学 | Pest trapping device and long-distance remote pest recognizing and monitoring system |
| CN103598169A (en) * | 2013-11-15 | 2014-02-26 | 华南农业大学 | Visual sensor based remote drosophila monitoring system and monitoring method thereof |
| CN203502363U (en) * | 2013-10-08 | 2014-03-26 | 南京萨特科技发展有限公司 | Reflecting image detecting equipment |
| CN103749416A (en) * | 2014-02-15 | 2014-04-30 | 福建农林大学 | Oriented monitoring, preventing and controlling system for pests |
| EP2730165A1 (en) * | 2011-07-06 | 2014-05-14 | Charles Liang | Intelligent theft-proof solar energy insect killer |
-
2015
- 2015-05-15 CN CN201510249053.1A patent/CN104902228A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1765701A (en) * | 2004-10-26 | 2006-05-03 | 北京航空航天大学 | Device for observing high-flap-rate insect flying motion |
| EP2730165A1 (en) * | 2011-07-06 | 2014-05-14 | Charles Liang | Intelligent theft-proof solar energy insect killer |
| CN202335307U (en) * | 2011-11-18 | 2012-07-18 | 扬州绿源生物化工有限公司 | Injurious insect trap |
| CN103299969A (en) * | 2013-06-09 | 2013-09-18 | 浙江大学 | Pest trapping device and long-distance remote pest recognizing and monitoring system |
| CN203502363U (en) * | 2013-10-08 | 2014-03-26 | 南京萨特科技发展有限公司 | Reflecting image detecting equipment |
| CN103598169A (en) * | 2013-11-15 | 2014-02-26 | 华南农业大学 | Visual sensor based remote drosophila monitoring system and monitoring method thereof |
| CN103749416A (en) * | 2014-02-15 | 2014-04-30 | 福建农林大学 | Oriented monitoring, preventing and controlling system for pests |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107543829A (en) * | 2016-06-29 | 2018-01-05 | 北京邮电大学 | Stored-grain insects detection system |
| CN106489876A (en) * | 2016-09-28 | 2017-03-15 | 浙江大学 | A kind of Agriculture field pests monitoring trap based on high light spectrum image-forming |
| CN107996537A (en) * | 2017-12-11 | 2018-05-08 | 江苏飞视文化发展有限公司 | A kind of device for being used to monitor silo China Oil and Food Import and Export Corporation worm |
| CN108651410A (en) * | 2018-05-18 | 2018-10-16 | 仲恺农业工程学院 | A kind of learning-oriented pest control equipment |
| CN108651410B (en) * | 2018-05-18 | 2021-05-18 | 仲恺农业工程学院 | Learning type pest control equipment |
| CN110363103A (en) * | 2019-06-24 | 2019-10-22 | 仲恺农业工程学院 | Identifying pest method, apparatus, computer equipment and storage medium |
| CN110363103B (en) * | 2019-06-24 | 2021-08-13 | 仲恺农业工程学院 | Insect pest identification method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103299969B (en) | Pest trapping device and long-distance remote pest recognizing and monitoring system | |
| CN104902228A (en) | Insect real-time monitoring device and method based on computer vision | |
| CN110235873B (en) | Automatic monitoring and forecasting system for insect pest situation of agricultural and forestry harmful insects | |
| CN101701906B (en) | Method and device for detecting stored-grain insects based on near infrared super-spectral imaging technology | |
| CN205390083U (en) | Intelligence plant diseases and insect pests monitoring and early warning system | |
| CN104918007A (en) | Computer vision-based large field pest situation monitoring sampling device and sampling method | |
| CN105941365A (en) | Automatic monitoring and prevention system for target pests | |
| CN108732175A (en) | Long-range insect identification based on machine vision and counting device | |
| CN203675911U (en) | Insect situation forecasting lamp with real-time monitoring function | |
| CN104137821A (en) | Insect pest situation forecasting management system and forecasting management method | |
| CN106489876B (en) | Hyperspectral imaging-based field pest monitoring and trapping device | |
| CN111109217A (en) | Intelligent insecticidal lamp control system | |
| CN211932193U (en) | Insect pest situation observation and report lamp with functions of screening, separating and filtering insect bodies | |
| CN111449035A (en) | Precise monitoring equipment, system and method for bactrocera dorsalis | |
| CN108490974B (en) | Ecological unmanned aerial vehicle that maintains | |
| CN114460080A (en) | Rice disease and pest intelligent monitoring system | |
| CN101655418A (en) | Grain condition detecting rod | |
| CN112101276A (en) | Forest resource distributed management system | |
| CN205756703U (en) | A kind of target pest monitors preventing mean automatically | |
| CN105868824A (en) | Intelligent identification data statistics system and insect attack monitoring device provided with same | |
| CN109964896B (en) | Insect attracting, identifying and sorting device and pest and disease monitoring and analyzing system based on Internet of things | |
| CN108462855A (en) | A kind of trapping lamp long-distance video monitoring system that can observe desinsection situation in real time | |
| CN108935373A (en) | A kind of method and trap of Grasshopper Population monitoring and prevention and treatment | |
| CN114201636A (en) | Forest pest disaster occurrence prediction method based on big data analysis | |
| CN108552138A (en) | A kind of ecology monitoring and safeguard unmanned plane |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150909 |
|
| WD01 | Invention patent application deemed withdrawn after publication |