CN103345634B - The automatic identifying method of Vegetables insecticide on a kind of yellow plate - Google Patents
The automatic identifying method of Vegetables insecticide on a kind of yellow plate Download PDFInfo
- Publication number
- CN103345634B CN103345634B CN201310321494.9A CN201310321494A CN103345634B CN 103345634 B CN103345634 B CN 103345634B CN 201310321494 A CN201310321494 A CN 201310321494A CN 103345634 B CN103345634 B CN 103345634B
- Authority
- CN
- China
- Prior art keywords
- insecticide
- image identification
- insect
- dark
- insect image
- 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.)
- Expired - Fee Related
Links
Landscapes
- Catching Or Destruction (AREA)
Abstract
The invention discloses 8 class Vegetables insecticide recognition methodss on a kind of yellow plate, the steps include: (1), image acquisition: utilize imaging device to gather the image of vegetable insecticide on yellow plate;(2), vegetable insect image identification on yellow plate zoomed in and out, rotate, shear, target image separating treatment;(3) vegetable insect image identification area, rectangular degree and elongation on yellow plate, is calculated;(4), by insect image identification dark color position and background separation;(5), calculate insect image identification dark color position image area, rectangular degree and dark position profile and protrude quantity;(6), the result of calculation obtained by morphological characteristic, step (3) and the step (5) being had according to inhomogeneity insecticide, it is judged that and show the title of vegetable insecticide on yellow plate.The present invention is that Vegetables insecticide on yellow plate can be known method for distinguishing by one automatically.The present invention have principle simple, easy and simple to handle, can be automatically to the advantages such as Vegetables insecticide is identified on yellow plate.
Description
Technical field
Present invention relates generally to the field of insecticide recognition methods in insecticide research, refer in particular to the automatic identifying method of Vegetables insecticide on a kind of yellow plate.
Background technology
At present, on green house of vegetables, yellow plate becomes increasingly popular to be applied to monitoring and pest control.Yellow plate can attract various insects, including insect and natural enemy, therefore identifies that on yellow plate, the kind of insecticide is the thorny problem that peasant household often is faced with.Identifying that mistake will take the prophylactico-therapeutic measures of mistake, cause production cost to increase, the agricultural product underproduction, pesticide residues rise.
Judge the kind of insecticide on yellow plate, one of existing method is: peasant household can ask entomology person skilled to be identified the insecticide on yellow plate, entomology person skilled often stays in bureau of agriculture, plant protection unit and agricultural colleges and universities and R&D institution, this approach can delay the time of peasant household, spends many travel charges and correlative charges.The two of existing method: peasant household is by browsing books and online enquiries judge the kind of insecticide on yellow plate, and the method needs peasant household to possess certain cultural competence, needs possess books and reference materials or online condition, and to spend the longer time.In prior art, similar insect image identification automatic identification technology has the method and system that butterfly is identified automatically, but the method does not have versatility.
Summary of the invention
The technical problem to be solved in the present invention is that the technical problem existed for prior art, and the present invention provides a kind of method that principle is simple, easy and simple to handle, automatically can be identified Vegetables insecticide on yellow plate.
For solving above-mentioned technical problem, the present invention is by the following technical solutions.
On a kind of yellow plate, the automatic identifying method of Vegetables insecticide, the steps include:
(1) image acquisition, process and insect image identification extract: gather insect image identification on vegetable Radix Rehmanniae plate;Adjust the resolution of insect image identification;And make the head of targeted insect image the most upward;Then, insect image identification region is sheared;Complete the extraction of insect image identification;
(2) insect image identification character numerical value calculates: calculate the area of insect image identification, rectangular degree and elongation;
(3) image zooming-out at insect body dark color position this characteristics of image numerical value is calculated: calculate dark site area (insect body dark color position pixel quantity), rectangular degree, the dark position concavo-convex quantity of profile;
(4) insecticide classification identification: compare according to the data that step (2) and step (3) obtain, it is judged that the classification of insecticide.
As a further improvement on the present invention, the particular content of described step (4) is:
A. it is zero when dark site area, when insect image identification elongation is less than 2, display insecticide entitled " aleyrodid ";
B. it is zero when dark site area, when insect image identification elongation is more than 2.3, display insecticide entitled " leafhopper ";
C. when dark position profile protrusion quantity is 4 ~ 10, display insecticide entitled " parasitic wasp ";
D. when dark position rectangular degree is between 5 ~ 20, display insecticide entitled " moth class ";
E. subtracting each other less than 6 when insect image identification rectangular degree with dark position rectangular degree, the logarithm of insect image identification area is between 7~9, when the logarithm of dark site area is between 4.5~7, and display insecticide entitled " aphid ";
F. when the logarithm of insect image identification area is between 6.1~8.2, the logarithm of insecticide dark color site area is between 3~6.7, insect image identification rectangular degree and dark position rectangular degree are between 25~55, and when dark position rectangular degree is more than insect image identification rectangular degree, show insecticide entitled " mosquito ";
G. when the logarithm of insect image identification area is between 8~12, the logarithm of the area at insect image identification and dark position subtracts each other less than 2, when the ratio of insect image identification rectangular degree and dark position rectangular degree is between 1.7~2.3, and display insecticide entitled " wasp fly ";
H. subtract each other less than 7 when the logarithm of insect image identification area and the logarithm of dark site area, and insect image identification rectangular degree and dark position rectangular degree are between 61~90, display insecticide entitled " chrysomelid ";
I., when numerical value does not meets above-mentioned all conditions, the entitled None-identified of insecticide is shown.
As a further improvement on the present invention, carrying out insect image identification when extracting in described step (1), first the yellow of yellow plate bleached, blackening again, and other colors are constant, and insect image identification and other background areas are separated, such insect image identification is extracted.
As a further improvement on the present invention, when described step (3) carries out the image zooming-out at insect body dark color position, first that image is graying, regulate gray threshold, insect body dark color station diagram picture is extracted.
Compared with prior art, advantages of the present invention is that
1, the automatic identifying method of Vegetables insecticide on the yellow plate of the present invention, principle is simple, easy and simple to handle rapidly;Utilize vegetable caste on this method identification Huang plate, can complete within half an hour to obtaining result from image acquisition.
2, the present invention can save expense after application, reduce the actual cost identifying operation.The equipment that this method utilizes is common digital camera and computer, consumes energy less than 1 degree of electricity, and expense is negligible.
3, the method for the present invention is easily mastered, easily promotes, and most of people can read alone this method operation instruction within half an hour and grasp this method.
Accompanying drawing explanation
Fig. 1 is present invention schematic flow sheet in a particular embodiment.
Fig. 2 is the schematic flow sheet that picture is rotated by the present invention in a particular embodiment.
Fig. 3 is the schematic flow sheet that the present invention shears targeted insect image-region in a particular embodiment.
Fig. 4 is the schematic flow sheet that insect image identification is extracted by the present invention in a particular embodiment.
Fig. 5 is the schematic flow sheet that insect body dark color station diagram picture is extracted by the present invention in a particular embodiment.
Fig. 6 is the schematic flow sheet that 8 class insecticides are differentiated by the present invention in a particular embodiment according to result of calculation.
Detailed description of the invention
Below with reference to specific embodiment and Figure of description, the present invention is described in further details.
As it is shown in figure 1, the automatic identifying method of Vegetables insecticide on the yellow plate of the present invention, the steps include:
(1) image acquisition: utilize imaging device to gather the image of vegetable insecticide on yellow plate.As: utilize digital camera by the image acquisition of vegetable insecticide on yellow plate and to be transported in computer.
(2) image scaling: during digital camera shooting image, owing to the resolution of digital camera different model is variant, during shooting, after target distance also leads to target imaging, resolution is variant, and insecticide size is the present invention judges an important evidence of caste, accordingly, it would be desirable to the image collected to be zoomed in and out the resolution reaching to be more or less the same.
(3) image rotation: on a yellow plate, insect numbers is more, most insecticide is that different directions is bonded on yellow plate on all fours, need the targeted insect head identified towards the most indefinite, unified head is oriented ensuing parameter and calculates and provide comparability.Accordingly, it would be desirable to rotate the image collected, postrotational picture makes targeted insect picture headers the most upward.
(4) image cut: a common size of yellow plate is between 200mm × 300mm to 300mm × 400mm, when field shooting picture, the dark degree of whole picture is different, relatively big with the difficulty of background image by Threshold segmentation insect image identification from whole pictures, nor is it necessary that.In order to reduce the interference of other insect image identifications, need artificially to cut by the less rectangular area at targeted insect place.
(5) insect image identification extracts: Vegetables insecticide color is all different with yellow plate color, utilizes this point the yellow of yellow plate to be bleached, blackening again, and other colors are constant.Computer R (red) redness, G (green) green, the various combination of blue three numerical value of B (blue) represent different colors, and tri-numerical value of RGB of the yellow of yellow plate are satisfied be respectively may be about: 200,180 and 50.If tri-numerical value of the RGB of the pixel in image file are respectively plus about 55,75,205, yellow i.e. becomes white, and white blackening again, such insect image identification is extracted.
(6) insect image identification character numerical value calculates: under equal resolution, different size of insecticide has different image pixel point quantity, the insecticide of formed objects causes the rectangular area cause not of uniform size at insect image identification place due to the difference of form and form of spreading the wings, the ratio of long axis to short axis of insect image identification is the most variant, calculate the area of insect image identification, rectangular degree and elongation, judge to provide foundation for insects is other.
(7) insect body dark color position image zooming-out: the wing of insecticide is often translucent, head and chest then color is the most dark, also has some insecticides decorative pattern with different depth colors, so, first that image is graying, regulate gray threshold, insect body dark color station diagram picture can be extracted.
(8) insect body dark color station diagram is as character numerical value calculating: different types of insecticide, its health dark color position is not of uniform size, profile is the most variant, calculates health dark color position pixel quantity, rectangular degree, dark position profile protrusion quantity, judges to provide foundation for insects is other.
(9) insecticide classification identification: on yellow plate, Vegetables insecticide mainly has flies, mosquito, parasitic wasp, aphid, aleyrodid, leafhopper, wasp fly, chrysomelid, moth class 9 class.The form of flies is similar to housefly, well known insecticide, the present invention does not provide recognition methods, the present invention provides the recognition methods of other 8 class insecticide, when the result shown when step (6) and (8) corresponds with the morphological characteristic of mosquito, parasitic wasp, aphid, aleyrodid, smaller green leaf hopper, wasp fly, chrysomelid, moth class, show the item name of this targeted insect, when not meeting above-mentioned 8 class Morphology of entomology feature, show None-identified.
The distinguishing characteristics of above-mentioned 8 class insecticides mainly has: a. aleyrodid and smaller green leaf hopper are without dark position;B. the body length of aleyrodid differs less than 1 times with body is wide;C. smaller green leaf hopper body length is more than wide 2 times of body;D. there is significantly depression between parasitic wasp health cephalothorax;E. aphid wing expanse is bigger;F. the rectangular degree of mosquito health and wing expanse is less than normal;G. wasp fly abdominal part has the decorative pattern that the depth is alternate;The most chrysomelid wing is opaque;I. the whole health of moth class is covered by the wing of the alternate decorative pattern of depth color.
In concrete the present embodiment, detailed operating process is as follows:
(1) image acquisition: utilize digital camera by the image acquisition of vegetable insecticide on yellow plate and to be transported in computer.
(2) image scaling: the actual size in kind that this picture is gathered is about 200mm × 300mm, the total pixel of this picture is 4000 × 3000, average every mm2Pixel be 196, approximate 200, therefore, this picture without scaling.
(3) image rotation: by this picture through 118 degree of rotations, targeted insect head is upward (such as Fig. 2).
(4) image cut: (such as Fig. 3) is got off in the cutting from whole pictures of targeted insect region with rectangle frame.
(5) insect image identification extracts: adjust tri-numerical value of RGB so that the yellow of yellow plate bleaches, and white blackening again, such insect image identification is extracted (such as Fig. 4).
(6) insect image identification character numerical value calculates: delete the row and column that insect image identification periphery is all black, accurately locks the minimum rectangular area at insect image identification place, insect image identification area, rectangular degree and elongation are calculated and be shown.
(7) insect body dark color position image zooming-out: will be graying through the insect image identification after step (5), regulation gray threshold, by constant less than the insect body dark color position of this gray threshold for brightness, brightness is higher than insecticide light color position and the whole blackening of background yellow of this gray threshold, insect body dark color station diagram picture is extracted (Fig. 5), as fruit insect brightness is not below the position of yellow plate brightness, then all pixels of this shearogram bleach.
(8) insect body dark color station diagram is as character numerical value calculating: delete step (7) gained image periphery afterwards is all black row and column, accurately locks the minimum enclosed rectangle region at place, insecticide dark color position, and dark position image area, rectangular degree are calculated and be shown.Statistics dark position profile protrudes quantity.
(9) caste identification: according to the result of calculation (such as Fig. 6) of step (8) and step (6).
A. it is zero when dark site area, when insect image identification elongation is less than 2, display insecticide entitled " aleyrodid ";
B. it is zero when dark site area, when insect image identification elongation is more than 2.3, display insecticide entitled " leafhopper ";
C. when dark position profile protrusion quantity is 4-10, display insecticide entitled " parasitic wasp ";
D. when dark position rectangular degree is between 5~20, display insecticide entitled " moth class ";
E. subtracting each other less than 6 when insect image identification rectangular degree with dark position rectangular degree, the logarithm of insect image identification area is between 7-9, when the logarithm of dark site area is between 4.5-7, and display insecticide entitled " aphid ";
F. when the logarithm of insect image identification area is between 6.1-8.2, the logarithm of insecticide dark color site area is between 3-6.7, insect image identification rectangular degree and dark position rectangular degree are between 25-55, and when dark position rectangular degree is more than insect image identification rectangular degree, show insecticide entitled " mosquito ";
G. when the logarithm of insect image identification area is between 8-12, the logarithm of the area at insect image identification and dark position subtracts each other less than 2, when the ratio of insect image identification rectangular degree and dark position rectangular degree is between 1.7-2.3, and display insecticide entitled " wasp fly ";
H. subtract each other less than 7 when the logarithm of insect image identification area and the logarithm of dark site area, and insect image identification rectangular degree and dark position rectangular degree are between 61-90, display insecticide entitled " chrysomelid ";
I., when the numerical value shown when step (6) and (8) does not meets above-mentioned condition, the entitled None-identified of insecticide is shown.
The above is only the preferred embodiment of the present invention, and protection scope of the present invention is not limited merely to above-described embodiment, and all technical schemes belonged under thinking of the present invention belong to protection scope of the present invention.It should be pointed out that, for those skilled in the art, some improvements and modifications without departing from the principles of the present invention, these improvements and modifications also should be regarded as protection scope of the present invention.
Claims (2)
1. the automatic identifying method of Vegetables insecticide on a yellow plate, it is characterised in that step is:
(1) image acquisition, process and insect image identification extract: gather insect image identification on vegetable Radix Rehmanniae plate;Adjust the resolution of insect image identification;And make the head of targeted insect image the most upward;Then, insect image identification region is sheared;Complete the extraction of insect image identification;
(2) insect image identification character numerical value calculates: calculate the area of insect image identification, rectangular degree and elongation;
(3) image zooming-out at insect body dark color position this characteristics of image numerical value is calculated: calculate dark site area, rectangular degree, dark position profile protrude quantity;When carrying out the image zooming-out at insect body dark color position, first that image is graying, regulate gray threshold, brightness is extracted less than the insect body dark color station diagram picture of this gray threshold;
(4) insecticide classification identification: compare according to the data that step (2) and step (3) obtain, it is judged that the classification of insecticide;
The particular content of described step (4) is:
A. it is zero when dark site area, when insect image identification elongation is less than 2, display insecticide entitled " aleyrodid ";
B. it is zero when dark site area, when insect image identification elongation is more than 2.3, display insecticide entitled " leafhopper ";
C. when dark position profile protrusion quantity is 4 ~ 10, display insecticide entitled " parasitic wasp ";
D. when dark position rectangular degree is between 5 ~ 20, display insecticide entitled " moth class ";
E. subtracting each other less than 6 when insect image identification rectangular degree with dark position rectangular degree, the logarithm of insect image identification area is between 7~9, when the logarithm of dark site area is between 4.5~7, and display insecticide entitled " aphid ";
F. when the logarithm of insect image identification area is between 6.1~8.2, the logarithm of insecticide dark color site area is between 3~6.7, insect image identification rectangular degree and dark position rectangular degree are between 25~55, and when dark position rectangular degree is more than insect image identification rectangular degree, show insecticide entitled " mosquito ";
G. when the logarithm of insect image identification area is between 8~12, the logarithm of the area at insect image identification and dark position subtracts each other less than 2, when the ratio of insect image identification rectangular degree and dark position rectangular degree is between 1.7~2.3, and display insecticide entitled " wasp fly ";
H. subtract each other less than 7 when the logarithm of insect image identification area and the logarithm of dark site area, and insect image identification rectangular degree and dark position rectangular degree are between 61~90, display insecticide entitled " chrysomelid ";
I., when numerical value does not meets above-mentioned all conditions, the entitled None-identified of insecticide is shown.
The automatic identifying method of Vegetables insecticide on yellow plate the most according to claim 1, it is characterized in that, when described step (1) carries out insect image identification extraction, first the yellow of yellow plate is bleached, blackening again, and other colors are constant, insect image identification and other background areas are separated, and such insect image identification is extracted.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310321494.9A CN103345634B (en) | 2013-07-29 | 2013-07-29 | The automatic identifying method of Vegetables insecticide on a kind of yellow plate |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310321494.9A CN103345634B (en) | 2013-07-29 | 2013-07-29 | The automatic identifying method of Vegetables insecticide on a kind of yellow plate |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103345634A CN103345634A (en) | 2013-10-09 |
CN103345634B true CN103345634B (en) | 2016-12-28 |
Family
ID=49280429
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310321494.9A Expired - Fee Related CN103345634B (en) | 2013-07-29 | 2013-07-29 | The automatic identifying method of Vegetables insecticide on a kind of yellow plate |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103345634B (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105894131A (en) * | 2016-04-28 | 2016-08-24 | 广东石油化工学院 | Fruit-piercing moth rapid early-warning method |
CN107066938B (en) * | 2017-02-08 | 2020-02-07 | 清华大学 | Video analysis apparatus, method and computer program product |
CN108040997B (en) * | 2017-12-25 | 2020-09-11 | 仲恺农业工程学院 | Insect pest monitoring method based on machine vision |
CN110276278A (en) * | 2019-06-04 | 2019-09-24 | 刘嘉津 | Insect image identification entirety and the recognition methods of multiple clips comprehensive automation |
CN112949625A (en) * | 2021-01-29 | 2021-06-11 | 西安电子科技大学 | Target identification method and system based on centroid contour distance |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102239793A (en) * | 2011-04-18 | 2011-11-16 | 浙江大学 | Real-time classification method and system of rice pests |
CN102819748A (en) * | 2012-07-19 | 2012-12-12 | 河南工业大学 | Classification and identification method and classification and identification device of sparse representations of destructive insects |
CN102930249A (en) * | 2012-10-23 | 2013-02-13 | 四川农业大学 | Method for identifying and counting farmland pests based on colors and models |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030026484A1 (en) * | 2001-04-27 | 2003-02-06 | O'neill Mark | Automated image identification system |
-
2013
- 2013-07-29 CN CN201310321494.9A patent/CN103345634B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102239793A (en) * | 2011-04-18 | 2011-11-16 | 浙江大学 | Real-time classification method and system of rice pests |
CN102819748A (en) * | 2012-07-19 | 2012-12-12 | 河南工业大学 | Classification and identification method and classification and identification device of sparse representations of destructive insects |
CN102930249A (en) * | 2012-10-23 | 2013-02-13 | 四川农业大学 | Method for identifying and counting farmland pests based on colors and models |
Non-Patent Citations (4)
Title |
---|
Automated insect identification through concatenated histograms of local appearance features: feature vector generation and region detection for deformable objects;Natalia Larios等;《Machine Vision and Applications》;20070707;全文 * |
基于图像的昆虫自动识别与计数研究进展;姚青等;《中国农业科学》;20110716(第14期);全文 * |
快速几何可变形彩色昆虫图像分割算法;黄世国等;《计算机应用》;20081201;第28卷(第12期);全文 * |
数学形态学在储量昆虫图像识别中的应用;张学庆等;《武汉工业学院学报》;20040330;第23卷(第1期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN103345634A (en) | 2013-10-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103345634B (en) | The automatic identifying method of Vegetables insecticide on a kind of yellow plate | |
Kulkarni et al. | Applying image processing technique to detect plant diseases | |
Wei et al. | Automatic method of fruit object extraction under complex agricultural background for vision system of fruit picking robot | |
Doitsidis et al. | Remote monitoring of the Bactrocera oleae (Gmelin)(Diptera: Tephritidae) population using an automated McPhail trap | |
CN106203265B (en) | A kind of Construction Fugitive Dust Pollution source monitors automatically and coverage forecasting system and method | |
Miranda et al. | Pest detection and extraction using image processing techniques | |
CA3172012C (en) | Insect monitoring system and method | |
CN109919966A (en) | Area determination method, device, storage medium and processor | |
Bhadane et al. | Early pest identification in agricultural crops using image processing techniques | |
CN106845415A (en) | A kind of pedestrian based on deep learning becomes more meticulous recognition methods and device | |
CN102768757A (en) | Remote sensing image color correcting method based on image type analysis | |
Zhang et al. | Individual leaf identification from horticultural crop images based on the leaf skeleton | |
CN106650822A (en) | Identification method and device for diseases and insect pests | |
CN107977631A (en) | The generation method and device of plant protection map | |
CN109784239A (en) | The recognition methods of winged insect quantity and device | |
CN102339378A (en) | Method and device for automatically extracting cotton seeds | |
CN110287822A (en) | The snail pest control method of view-based access control model target detection in a kind of officinal dendrobium stem plantation | |
Mortensen et al. | The oil radish growth dataset for semantic segmentation and yield estimation | |
CN105894131A (en) | Fruit-piercing moth rapid early-warning method | |
CN114022656A (en) | Intelligent air-suction type LED characteristic spectrum pest trapping and killing system and method | |
CN104616283B (en) | A kind of pest detection method based on low-rank representation algorithm | |
CN110796148A (en) | Litchi insect pest monitoring and identifying system and litchi insect pest monitoring and identifying method | |
Tripathy et al. | Image processing techniques aiding smart agriculture | |
Wadhai et al. | Agriculture pest detection using video processing technique | |
CN108304818A (en) | A kind of mosquito matchmaker automatic distinguishing method for image |
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 |
Granted publication date: 20161228 Termination date: 20170729 |
|
CF01 | Termination of patent right due to non-payment of annual fee |