CN103808265B - Method, device and system for measuring oilseed rape laminae and forms of sclerotium scabs synchronously - Google Patents
Method, device and system for measuring oilseed rape laminae and forms of sclerotium scabs synchronously Download PDFInfo
- Publication number
- CN103808265B CN103808265B CN201410071688.2A CN201410071688A CN103808265B CN 103808265 B CN103808265 B CN 103808265B CN 201410071688 A CN201410071688 A CN 201410071688A CN 103808265 B CN103808265 B CN 103808265B
- Authority
- CN
- China
- Prior art keywords
- image
- leaf
- shared
- scab
- grid
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
The invention provides a method and a system for measuring oilseed rape laminae and the forms of sclerotium scabs synchronously. The method includes the following steps of: acquiring images of the oilseed rape laminae infected with sclerotiniose; segmenting the acquired image to acquire background gridding images, and calculating pixel counts occupied by unit gridding and pixel counts occupied by unit length; extracting the images of the oilseed rape laminae, and calculating the areas, lengths and widths of the oilseed rape laminae; extracting the images of the sclerotium scabs, and calculating the areas, lengths and widths of the sclerotium scabs; and calculating the proportion of the sclerotium scabs to the integral oilseed rape laminae. The system comprises a gridding background plate, a tectorial membrane, a support, a clamping unit and an image acquisition and processing unit. By the method and the system, the synchronous, quick, accurate and nondestructive measurement of the oilseed rape laminae and the forms of the sclerotium scabs can be realized.
Description
Technical field
The present invention relates to agricultural technology field, and in particular to the method for rape leaf and sclerotium scab pattern synchro measure,
Apparatus and system.
Background technology
Brassica campestris L is one of four big oil crop of the world, and sclerotinia rot of colza is worldwide by the microbial one kind of nuclear disk
Major Diseases, can fall ill at each period of duration of Brassica campestris L, each position, and greatly harm is brought to Rape-seed production.Therefore, accurately comment
The order of severity of valency sclerotiniose morbidity is all very necessary for the preventing and treating that sclerotinia rot of colza does harm to, the identification of anti-source and breeding for disease resistance
's.The method of conventional measurement Sclerotinia Sclerotiorum is come to disease journey according to the size or number of scab on its incidence of leaf
Degree is defined the level.Therefore, it is that rapeseed breeding and species test are special that objective, accurate measurement is carried out to rape leaf and sclerotium scab pattern
Family is carrying out one of sclerotium disease deciding grade and level when institute's problem demanding prompt solution.
Measurement rape leaf and sclerotium scab pattern are mainly using manual ruler measurement at present.By advanced laser scanning
Technology can also realize the measurement of plant leaf blade form.Such as, the CI-202 portable laser leaf areas of U.S. CID companies production
Instrument.When measuring, rape leaf can be placed between thin film and plate, then the thin transparent by scanning device from scanning board
It is inswept on film.The parameters such as the measurable blade areas of CI-202, leaf length, leaf width degree.At present, also Jing is frequently with digital camera
Rape leaf and scab pattern are measured, when measuring, is placed with scab first on the blank of known area
Rape leaf and make leveling blade, using the digital camera directly over being in white board and rape leaf by white board and blade
The image of a picture is shot, then using Photoshop software processes images, image white part, healthy part is read
With the number of the three-channel pixel in scab part, the areal calculation oil of area ratio and white board is equal to finally according to pixel ratio
The area of dish leaf piece and sclerotium scab.
Although the method measurement of existing manual ruler measurement gets up more convenient, this method not only workload
Greatly, and take great energy, owe objectivity and accuracy, the precision to rapeseed breeding and automatic measurement bring delayed for a long time
Impact；Although laser leaf area instrument method can accurately measure the parameters such as the leaf area of rape leaf, leaf length and leaf width, this
The method of kind needs rape leaf is taken from plant in measurement and measures again, and damage is caused to plant, and this
Method can not synchronously calculate the morphological parameters of sclerotiniose speckle；Although digital camera method can measure rape leaf and bacterium simultaneously
Core scab pattern, but this method is difficult to operate in somatometry, easily causes shooting image serious distortion, causes serious
Error, and this method also needs to measure picture importing computer Photoshop, it is impossible to enough output rape leaves in real time
The morphological parameters of piece and sclerotium scab.
The content of the invention
（One）The technical problem of solution
For the deficiencies in the prior art, the invention provides the method for rape leaf and sclerotium scab pattern synchro measure and
Device, can realize rape leaf and sclerotium scab pattern simultaneously and rapidly, accurately, nondestructive measurement.
（Two）Technical scheme
In order to realize object above, the present invention is achieved by the following technical programs：
A kind of method of rape leaf and sclerotium scab pattern synchro measure, the method comprising the steps of：
The rape leaf image of collection infection sclerotiniose；
The image segmentation of collection is obtained into background grid image, the pixel count and unit length institute shared by unit of account grid
The pixel count for accounting for；
Rape leave image is extracted, Brassica campestris L leaf area, leaf length and leaf width degree is calculated；
Sclerotium scab image is extracted, area, the length and width of sclerotium scab is calculated；
Calculate the ratio of whole rape leave shared by sclerotium scab.
Wherein, it is described by collection image segmentation obtain background grid image, the pixel count shared by unit of account grid and
Pixel count shared by unit length, including：
Image to gathering carries out pretreatment, and color notation conversion space is carried out to pretreated image, extracts background net
Table images；
Morphological scale-space is carried out to background grid image, and simple grid image is obtained using Minimum Enclosing Rectangle method；
Simple grid image to obtaining is refined；
The pixel count shared by pixel count and unit length shared by unit of account grid.
Wherein, the image of described pair of collection carries out pretreatment, and carries out color notation conversion space to pretreated image, carries
Background grid image is taken, including：
Image to collecting carries out 5*5 medium filterings；
Color notation conversion space is carried out to the image after medium filtering, image is converted to HSL spaces by rgb space；
Obtain the rectangular histogram of tri- components of H, S, L of described HSL spatial images, the comprehensive Nogata according to three components
Figure, obtains background grid image；
Image is converted to into rgb space coloured image by HSL spaces.
Wherein, Morphological scale-space is carried out to background grid image, and simple grid is obtained using Minimum Enclosing Rectangle method
Image, including：
5*5 medium filterings are carried out to background grid image；
Gray processing is carried out to background grid image, using best threshold method, selected threshold carries out two to the gray level image
Value is processed, and obtains bianry image；
The bianry image is negated；
The Morphological scale-space method combined with corrosion using expansion, burn into expansion is carried out to the image of the inverted
Process so that the part of the prospect blade in image forms a complete connected region；
According to the connected region, the minimum enclosed rectangle in the largest connected region at the part place of blade is found out, is utilized
Minimum enclosed rectangle removes leaf area, obtains simple grid image；
The grid image is negated.
Wherein, the simple grid image of described pair of acquisition is refined, including：
Range conversion is carried out to described simple grid image, the candidate point of skeletal point is determined；
Rule constraint is carried out to the candidate point, single pixel width skeleton is obtained by skeleton growth.
Wherein, the pixel count and the pixel count shared by unit length shared by the unit of account grid, including：
Contours extract is carried out to the image after the refinement；
Obtain all of grid number and the pixel count shared by all of net region；
Using the pixel count shared by all of net region divided by all of grid number, the pixel shared by unit grids is obtained
Number；
Pixel count shared by unit grids is made even root, the pixel count shared by unit length is obtained.
Wherein, the extraction rape leave image, calculates Brassica campestris L leaf area, leaf length and leaf width degree, including：
Rape leaf image to collecting carries out 5*5 medium filterings；
After medium filtering described in obtaining, the rectangular histogram of the green component of coloured image, according to green component rectangular histogram, obtains
Obtain prospect leaf image；
Gray processing is carried out to described prospect leaf image, choosing optimal threshold is carried out at binaryzation to the gray level image
Reason, obtains bianry image；
Using burn into expansion, first corrode the Morphological scale-space method that expands afterwards the bianry image of acquisition is processed, obtain
Take the complete connected region of blade；
Contours extract is carried out to connected region, and acquisition maximum area profile is compared according to contour area size, and will most
Pixel value zero setting outside large area profile, obtains the profile in rape leaf region；
The hole in rape leaf region contour is filled using unrestrained water filling algorithm, is smoothed, complete oil
Dish blade profile；
Count the number of the pixel shared by whole rape leaf；
Using Minimum Enclosing Rectangle method, the pixel count shared by the length and width of rape leaf is calculated；
Pixel count according to shared by unit area and unit length, calculates leaf area, leaf length and the leaf of rape leaf
Width.
Wherein, described extraction sclerotium scab image, calculates area, the length and width of sclerotium scab, including：
Rape leaf image to collecting carries out 5*5 medium filterings；
After medium filtering described in obtaining, the rectangular histogram of R, G, B of coloured image, tentatively obtains scab figure according to rectangular histogram
Picture；
Gray processing is carried out, and chooses optimal threshold binary conversion treatment being carried out to gray level image to described scab image, obtain
To bianry image；
The bianry image for obtaining is processed using the Morphological scale-space method that expands afterwards is first corroded, obtain scab prospect
Image, and calculate the pixel number shared by scab；
According to the scab foreground image, the minimum enclosed rectangle of scab is obtained, and calculate shared by the length and width of scab
Pixel count；
Pixel count according to shared by unit area and unit length, calculates area, the length and width of scab.
A kind of device of rape leaf and sclerotium scab pattern synchro measure, including：Image acquisition units, pretreatment list
Unit, leaf morphology computing unit, scab pattern computing unit, scab ratio computing unit；
Described image collecting unit, for gathering the rape leaf image of infection sclerotiniose；
The pretreatment unit, for the image segmentation of collection is obtained background grid image, shared by unit of account grid
Pixel count and unit length shared by pixel count；
The leaf morphology computing unit, for extracting rape leave image, calculates Brassica campestris L leaf area, leaf length and leaf width
Degree；
The scab pattern computing unit, for extracting sclerotium scab image, calculates area, length and the width of sclerotium scab
Degree；
The scab ratio computing unit, accounts for the ratio of whole rape leave for calculating sclerotium scab.
A kind of system of rape leaf and sclerotium scab pattern synchro measure, including：Grid background board, overlay film, support, folder
Hold the device of unit and rape leaf as claimed in claim 9 and sclerotium scab pattern synchro measure；
The grid background board is used to be placed in parallel rape leaf to be measured；
The overlay film is for pushing down rape leaf to be measured and ensure that blade is not overlapped and crimped；
The support is for supporting, fixed clamp unit, and causes described rape leaf and sclerotium scab pattern synchronization
Fixed distance is formed between the device and grid background board of measurement；
The grip unit is used for the device for clamping, fixing the rape leaf and sclerotium scab pattern synchro measure.
Wherein, the device of described rape leaf and sclerotium scab pattern synchro measure is smart mobile phone.
Wherein, the grip unit, including：Collet, clamping plate, regulation button, stand-off and rotary shaft；
The collet is used to support smart mobile phone and other parts, keeps smart mobile phone photographic head to keep flat with background board
OK；
The clamping plate is used to clamp smart mobile phone；
Adjusting button is used to adjust the clamping plate, and the degree of mobile phone is clamped so as to adjust the clamping plate；
The stand-off is used to hold smart mobile phone bottom, prevents smart mobile phone from dropping；
The rotary shaft is used to adjust the flexible of the stand-off.
（Three）Beneficial effect
The present invention at least has the advantages that：
1st, the method that the present invention is provided carries out pretreatment, grid background segment, leaf by the rape leaf image to gathering
The steps such as piece is extracted, scab extraction are analyzed, and realize rape leaf and the quick, accurate of sclerotium scab pattern, synchronization, nothing
The measurement of damage.Relative to traditional rape leaf and sclerotium scab method for measuring shape of palaemon, have larger in efficiency and accuracy
Raising.
2nd, the system that the present invention is provided is portable design, using main flow on the market smart mobile phone as image acquisition,
Process, display platform, using lightweight acrylic section bar as system support and background board, substantially reduce system weight and
Volume, it is very convenient to use.
Description of the drawings
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
Accompanying drawing to be used needed for having technology description is briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can be with
Other accompanying drawings are obtained according to these figures.
Accompanying drawing 1 is the stream of the method for a kind of rape leaf provided in an embodiment of the present invention and sclerotium scab pattern synchro measure
Cheng Tu；
Fig. 2 is a kind of flow process for obtaining unit grids size and pixel count shared by unit length provided in an embodiment of the present invention
Figure；
Fig. 3 is a kind of area for calculating rape leaf provided in an embodiment of the present invention, the flow chart of length and width；
Fig. 4 is a kind of calculating lesion area provided in an embodiment of the present invention, the flow chart of length and width；
Fig. 5 is the overall structure of the system of a kind of rape leaf provided in an embodiment of the present invention and sclerotium scab synchro measure
Figure.
Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than the embodiment of whole.It is based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of creative work is not made
Embodiment, belongs to the scope of protection of the invention.
Referring to Fig. 1, a kind of method for embodiments providing rape leaf and sclerotium scab pattern synchro measure should
Method comprises the steps：
Step 101：The rape leaf image of collection infection sclerotiniose.
Step 102：The image segmentation of collection is obtained into background grid image, pixel count and unit shared by unit of account grid
Pixel count shared by length.
Referring to Fig. 2, step 102 specifically includes following steps：
Step 201：Image to gathering carries out pretreatment, and carries out color space conversion to pretreated image, carries
Take background grid image.Specifically include：
（1）Image to collecting carries out 5*5 medium filterings, so as to remove the salt-pepper noise in image；
（2）Color space change is carried out to the image after medium filtering, image is converted into HSL spaces by rgb space；
（3）The rectangular histogram of tri- components of H, S, L of described HSL spatial images is obtained, it is comprehensive according to the straight of three components
Side's figure arranges threshold T h, Ts, the scope of Tl, the prospect blade-section pixel value within the scope of threshold value is set to zero, thresholding
Background grid pixel value outside value scope keeps constant, realizes that background grid image is separated with prospect leaf image, obtains
Background grid image.Threshold value process after image g (x, y) be：
Wherein, f (x, y) represents the image array before threshold value process.
（4）Image is converted into into rgb space coloured image by HSL spaces.
Step 202：Applied morphology processing method is processed to background grid image, and adopts Minimum Enclosing Rectangle method
Obtain simple grid image.Specifically include：
（1）5*5 medium filterings are carried out to background grid image, the noise spot in image is removed；
（2）Gray processing is carried out, and chooses optimal threshold binary conversion treatment being carried out to the gray level image to background image, obtain
To bianry image；
（3）The bianry image is negated；
（4）Using expansion, burn into expansion and corrode the Morphological scale-space method that combines to image of the inverted at
Reason so that prospect blade-section forms a complete connected region in image；
（5）According to the connected region, the minimum enclosed rectangle in the largest connected region at blade place is found out, using minimum
The interference that boundary rectangle method is made a return journey except leaf area to extraction grid, obtains simple grid image.Method is:Obtain first most
The coordinate of the top left corner apex of little boundary rectangle（x1,y1）, therefore, the coordinate of minimum enclosed rectangle bottom right angular vertex（x2,y2）
For：
Wherein, width of the w for minimum enclosed rectangle, height of the h for minimum enclosed rectangle.
Then the pixel value of the pixel within four apex coordinate areas encompassed of minimum enclosed rectangle is set to
255, the pixel value of the pixel of the meshing outside region keeps constant, and obtaining single net region image g (x, y) is：
Wherein, the image array before f (x, y) expression is processed.
（6）The grid image is negated.
Step 203：Simple grid image to obtaining is refined.Specifically include：
（1）Range conversion is carried out to described simple grid image, the candidate point of skeleton is determined；
（2）Rule constraint is carried out to the candidate point, single pixel width skeleton is obtained by skeleton growth.
Step 204：The pixel count shared by pixel count and unit length shared by unit of account grid.Specifically include：
（1）Contours extract is carried out to the image after the refinement；
（2）Obtain all of grid number N and pixel count A shared by all of net region；
（3）Using the pixel count shared by all of net region divided by all of grid number, obtain shared by unit grids
Pixel count, i.e., the pixel count C1 shared by each square centimeter；
Wherein, C1 is the pixel count shared by a sq；
（4）Pixel count shared by described each square centimeter is made even root, the picture shared by a side length of element is obtained
Prime number, namely the pixel count C2 shared by each centimetre length；
Wherein, C2 is the pixel count shared by a centimetre length；
Step 103：Rape leave image is extracted, Brassica campestris L leaf area, leaf length and leaf width degree is calculated.
Referring to Fig. 3, step 103 specifically includes following steps：
Step 301：Rape leaf image to collecting carries out pretreatment, i.e., carry out 5*5 medium filterings to image, so as to
Remove the salt-pepper noise in image；
Step 302：The rectangular histogram of the green component of coloured image after medium filtering described in obtaining, according to green component
Rectangular histogram arranges threshold T g, for green component values are less than the pixel of Tg, by the value difference zero setting of its tri- component of R, G, B；
For green component values are more than the pixel of Tg, the value of its tri- component of R, G, B keeps constant；Realize prospect leaf image and background
The separation of grid, it is preliminary to obtain prospect leaf image；
Wherein gR(x,y)、gG(x,y)、gB(x, y) be process after three pixels three component values, fR(x,y)、fG(x,
y)、fB(x, y) is three component values for processing first three pixel.
Step 303：Gray processing is carried out to described prospect leaf image, optimal threshold is chosen and the gray level image is entered
Row binary conversion treatment, obtains bianry image；
Step 304：Using burn into expansion, first corrode the Morphological scale-space method that expands afterwards the bianry image of acquisition is entered
Row is processed, the white grid lines remained in removing image, and causes blade-section to become a complete connected region；
Step 305：Contours extract is carried out to the grid image after Morphological scale-space, and is obtained most according to contour area size
Large area profile, and by the pixel value zero setting outside maximum area profile, obtain the profile in rape leaf region；
Step 306：The hole in rape leaf region contour is filled using unrestrained water filling algorithm, smoothed,
Complete rape leaf profile；
Step 307：The pixel value of the pixel inside blade profile is set to into 255, other pixel pixel values are set to zero,
And calculate the number of white pixel point, that is, number N1 of whole pixel shared by rape leaf；
Step 308：According to the complete rape leaf profile, the minimum enclosed rectangle of profile is obtained, and is calculated most
The length of little boundary rectangle and wide pixel count shared respectively, namely shared by the pixel count N2 and width shared by the length of rape leaf
Pixel count N3；
Pixel count according to shared by each square centimeter and each centimetre length, converse rape leaf leaf area M1,
Leaf length M2 and leaf width degree M3.
Wherein, C1 is the pixel count shared by a sq, and C2 is the pixel count shared by a centimetre length.
Step 104：Sclerotium scab image is extracted, area, the length and width of sclerotium scab is calculated.
Referring to Fig. 4, step 104 specifically includes following steps：
Step 401：Rape leaf image to collecting carries out 5*5 medium filtering pretreatment, removes the spiced salt in image
Noise；
Step 402：After medium filtering described in obtaining, the rectangular histogram of R, G, B of coloured image, determines according to rectangular histogram
When the G components of pixel are all higher than R component and B component, by R, G, B three-component value difference zero setting of this pixel, other pictures
Vegetarian refreshments three-component pixel value keeps constant, realizes that scab image is separated with other backgrounds, preliminary to obtain scab image；
Wherein, gR(x,y)、gG(x,y)、gB(x, y) be process after three pixels three component values, fR(x,y)、fG(x,
y)、fB(x, y) is three component values for processing first three pixel.
Step 403：Gray processing is carried out to the scab image of the preliminary acquisition, and optimal threshold is chosen to the gray-scale maps
As carrying out binary conversion treatment, bianry image is obtained；
Step 404：The bianry image for obtaining is processed using the Morphological scale-space method that expands afterwards is first corroded, removed
The noise block remained in image or noise spot, obtain simple scab foreground image, and calculate the pixel shared by scab
Number D1；
Step 405：According to the scab foreground image, the minimum enclosed rectangle of scab is obtained, and is calculated minimum external
The length of rectangle and wide pixel count shared respectively, namely the pixel count shared by the pixel count D2 and width shared by the length of scab
D3；
Step 406：Pixel count according to shared by each square centimeter and each centimetre length, calculates the area of scab
F1, length F2 and width F3.
Wherein, C1 is the pixel count shared by a sq, and C2 is the pixel count shared by a centimetre length.
Step 105：Calculate the ratio of whole rape leave shared by sclerotium scab.
In this step, with the area F1 of scab divided by the area M1 of blade, the percentage that sclerotium scab accounts for rape leaf is obtained
Compare R.
Method provided in an embodiment of the present invention carries out pretreatment, grid background point by the rape leaf image to gathering
Cut, blade is extracted, scab is extracted etc., and step is analyzed, and realizes the quick, accurate, same of rape leaf and sclerotium scab pattern
Step, lossless measurement.Relative to traditional rape leaf and sclerotium scab method for measuring shape of palaemon, have in efficiency and accuracy
Larger raising.
The device of a kind of rape leaf and sclerotium scab pattern synchro measure is embodiments provided, including：Image
Collecting unit, pretreatment unit, leaf morphology computing unit, scab pattern computing unit, scab ratio computing unit.Wherein,
Described image collecting unit, for gathering the rape leaf image of infection sclerotiniose；The pretreatment unit, for by collection
Image segmentation obtains background grid image, the pixel count shared by pixel count and unit length shared by unit of account grid；It is described
Leaf morphology computing unit, for extracting rape leave image, calculates Brassica campestris L leaf area, leaf length and leaf width degree；The scab shape
State computing unit, for extracting sclerotium scab image, calculates area, the length and width of sclerotium scab；The scab ratio meter
Unit is calculated, for calculating the ratio of whole rape leave shared by sclerotium scab.
The system that the embodiment of the present invention additionally provides a kind of rape leaf and sclerotium scab pattern synchro measure, such as Fig. 5 institutes
Show, the system includes：Grid background board 1, support 2, collet 3, clamping plate 4, stand-off 5, regulation button 6, smart mobile phone 7, rotary shaft
8, overlay film 9.
Wherein, grid background board 1 is used to be placed in parallel rape leaf to be measured；
Overlay film 9 is for pushing down rape leaf to be measured and ensure that blade is not folded and crimped；
Support 2 is for supporting, fixed clamp unit, and causes to be formed between the device and grid background board 1 of smart mobile phone 7
Fixed distance；
Grip unit, including collet 3, clamping plate 4, regulation button 6, stand-off 5 and rotary shaft 8, grip unit are used to clamp, consolidate
Determine smart mobile phone 7.Wherein, collet 3 is used to support smart mobile phone and other parts, keeps 7 photographic head of smart mobile phone and grid
1 keeping parallelism of background board；Clamping plate 4 is used to clamp smart mobile phone 7；Adjusting button 6 is used to adjust clamping plate 4, so as to adjust the folder
Plate 4 clamps the degree of smart mobile phone 7；Stand-off 5 is used to hold 7 bottom of smart mobile phone, prevents smart mobile phone 7 from dropping；Rotary shaft 8 is used
In the flexible of regulation stand-off 5.
During collection image, the user right hand holds the bottom of grid background board 1 so that the right hand can hold tight whole dress
Put, then by device near rape leaf, start overlay film 9, rape leaf is lain on grid background board 1, flattened with handss, so
Afterwards overlay film 9 is put down and is pushed down rape leaf and is ensured that blade is not folded and crimped so that rape leaf and grid background board 1
Occur in mobile phone photograph preview pane simultaneously, meanwhile, 7 display screen of right-hand operated smart mobile phone, the software transfer mobile phone in mobile phone
Photographic head is together taken pictures to rape leaf and grid background board 1, the cromogram that acquisition background is waffle slab, prospect is rape leaf
Picture.
System provided in an embodiment of the present invention is portable design, is adopted as image using the smart mobile phone of main flow on the market
Collection, process, display platform, using lightweight acrylic section bar as the support and background board of system, substantially reduce the weight of system
Amount and volume, it is very convenient to use.
Above example only to illustrate technical scheme, rather than a limitation；Although with reference to the foregoing embodiments
The present invention has been described in detail, it will be understood by those within the art that；Which still can be to aforementioned each enforcement
Technical scheme described in example is modified, or carries out equivalent to which part technical characteristic；And these modification or
Replace, do not make the essence of appropriate technical solution depart from the spirit and scope of various embodiments of the present invention technical scheme.
Claims (7)
1. a kind of method of rape leaf and sclerotium scab pattern synchro measure, it is characterised in that the method comprising the steps of：
The rape leaf image of collection infection sclerotiniose；
The image segmentation of collection is obtained into background grid image, shared by the pixel count and unit length shared by unit of account grid
Pixel count；
Rape leave image is extracted, Brassica campestris L leaf area, leaf length and leaf width degree is calculated；
Sclerotium scab image is extracted, area, the length and width of sclerotium scab is calculated；
Calculate the ratio of whole rape leave shared by sclerotium scab；
The extraction rape leave image, calculates Brassica campestris L leaf area, leaf length and leaf width degree, including：
Rape leaf image to collecting carries out 5*5 medium filterings；
The rectangular histogram of the green component of coloured image after medium filtering described in obtaining, according to green component rectangular histogram, before acquisition
Scape leaf image；
Gray processing is carried out to described prospect leaf image, choosing optimal threshold carries out binary conversion treatment to the gray level image,
Obtain bianry image；
Using burn into expansion, first corrode the Morphological scale-space method that expands afterwards the bianry image for obtaining is processed, acquisition leaf
The complete connected region of piece；
Carry out contours extract to connected region, and acquisition maximum area profile is compared according to contour area size, and by largest face
Pixel value zero setting outside product profile, obtains the profile in rape leaf region；
The hole in rape leaf region contour is filled using unrestrained water filling algorithm, is smoothed, complete rape leave
Piece profile；
Count the number of the pixel shared by whole rape leaf；
Using Minimum Enclosing Rectangle method, the pixel count shared by the length and width of rape leaf is calculated；
Pixel count according to shared by unit area and unit length, calculates leaf area, leaf length and the leaf width degree of rape leaf.
2. method according to claim 1, it is characterised in that the image segmentation by collection obtains background grid figure
Picture, the pixel count shared by pixel count and unit length shared by unit of account grid, including：
Image to gathering carries out pretreatment, and color notation conversion space is carried out to pretreated image, extracts background grid figure
Picture；
Morphological scale-space is carried out to background grid image, and simple grid image is obtained using Minimum Enclosing Rectangle method；
Simple grid image to obtaining is refined；
The pixel count shared by pixel count and unit length shared by unit of account grid.
3. method according to claim 2, it is characterised in that the image of described pair of collection carries out pretreatment, and to pre- place
Image after reason carries out color notation conversion space, extracts background grid image, including：
Image to collecting carries out 5*5 medium filterings；
Color notation conversion space is carried out to the image after medium filtering, image is converted to HSL spaces by rgb space；
The rectangular histogram of tri- components of H, S, L of described HSL spatial images is obtained, the comprehensive rectangular histogram according to three components is divided
Cut acquisition background grid image；
Image is converted to into rgb space coloured image by HSL spaces.
4. method according to claim 2, it is characterised in that described to carry out Morphological scale-space to background grid image, and
Simple grid image is obtained using Minimum Enclosing Rectangle method, including：
5*5 medium filterings are carried out to background grid image；
Gray processing is carried out to background grid image, using best threshold method, selected threshold carries out binaryzation to the gray level image
Process, obtain bianry image；
The bianry image is negated；
The Morphological scale-space method combined with corrosion using expansion, burn into expansion is processed to the image of the inverted,
So that the part of the prospect blade in image forms a complete connected region；
According to the connected region, the minimum enclosed rectangle in the largest connected region at the part place of blade is found out, using minimum
Boundary rectangle removes leaf area, obtains simple grid image；
The grid image is negated.
5. method according to claim 2, it is characterised in that the simple grid image of described pair of acquisition is refined,
Including：
Range conversion is carried out to described simple grid image, the candidate point of skeletal point is determined；
Rule constraint is carried out to the candidate point, single pixel width skeleton is obtained by skeleton growth.
6. method according to claim 2, it is characterised in that the pixel count and unit shared by the unit of account grid is long
The shared pixel count of degree, including：
Contours extract is carried out to the image after the refinement；
Obtain all of grid number and the pixel count shared by all of net region；
Using the pixel count shared by all of net region divided by all of grid number, the pixel count shared by unit grids is obtained；
Pixel count shared by unit grids is made even root, the pixel count shared by unit length is obtained.
7. method according to claim 1, it is characterised in that described extraction sclerotium scab image, calculates the face of sclerotium scab
Product, length and width, including：
Rape leaf image to collecting carries out 5*5 medium filterings；
After medium filtering described in obtaining, the rectangular histogram of R, G, B of coloured image, tentatively obtains scab image according to rectangular histogram；
Gray processing is carried out, and chooses optimal threshold binary conversion treatment being carried out to gray level image to described scab image, obtain two
Value image；
The bianry image for obtaining is processed using the Morphological scale-space method that expands afterwards is first corroded, obtain scab foreground picture
Picture, and calculate the pixel number shared by scab；
According to the scab foreground image, the minimum enclosed rectangle of scab is obtained, and calculates the length and the picture shared by width of scab
Prime number；
Pixel count according to shared by unit area and unit length, calculates area, the length and width of scab.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410071688.2A CN103808265B (en) | 2014-02-28 | 2014-02-28 | Method, device and system for measuring oilseed rape laminae and forms of sclerotium scabs synchronously |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410071688.2A CN103808265B (en) | 2014-02-28 | 2014-02-28 | Method, device and system for measuring oilseed rape laminae and forms of sclerotium scabs synchronously |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103808265A CN103808265A (en) | 2014-05-21 |
| CN103808265B true CN103808265B (en) | 2017-04-26 |
Family
ID=50705384
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410071688.2A Active CN103808265B (en) | 2014-02-28 | 2014-02-28 | Method, device and system for measuring oilseed rape laminae and forms of sclerotium scabs synchronously |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103808265B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106468543B (en) * | 2015-08-21 | 2019-12-31 | 浙江托普云农科技股份有限公司 | Blade area measuring method based on image processing |
| CN110895804A (en) * | 2018-09-10 | 2020-03-20 | 上海市农业科学院 | Fuzzy edge lesion extraction method and device |
| CN109271919B (en) * | 2018-09-12 | 2022-11-01 | 海南省海洋与渔业科学院 | Vegetation coverage measuring method based on grb and grid mode |
| CN109492569A (en) * | 2018-10-31 | 2019-03-19 | 国家电网有限公司 | A kind of cable line insulating layer defect detection method and device |
| CN109610132A (en) * | 2018-11-22 | 2019-04-12 | 无锡小天鹅股份有限公司 | Laundry process, device and controller |
| CN109754423B (en) * | 2018-11-28 | 2020-09-15 | 中国农业科学院农业信息研究所 | Method and equipment for extracting coverage area of leaf scab |
| CN109801235B (en) * | 2018-12-28 | 2023-03-28 | 佛山科学技术学院 | Method and device for detecting disease cause of epipremnum aureum leaves |
| CN109945784B (en) * | 2019-03-22 | 2020-10-09 | 安徽省农业科学院农业经济与信息研究所 | Plant protection image pest size measuring and calculating method based on image Exif metadata |
| CN110580495A (en) * | 2019-06-21 | 2019-12-17 | 南京农业大学 | automatic analysis method for leaf area and leaf surface anthracnose lesion number of pear |
| CN114463245A (en) * | 2020-11-10 | 2022-05-10 | 北京百度网讯科技有限公司 | Image processing method and device and electronic equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007171033A (en) * | 2005-12-22 | 2007-07-05 | Univ Nagoya | Indirect measuring method and system of leaf area index |
| CN101639946A (en) * | 2009-08-26 | 2010-02-03 | 北京农业信息技术研究中心 | Plant leaf three-dimensional model geometric description and curved surface reconstruction method and system |
| CN102200433A (en) * | 2011-02-25 | 2011-09-28 | 北京农业信息技术研究中心 | Device and method for measuring leaf area based on computer vision |
| CN102214306A (en) * | 2011-06-16 | 2011-10-12 | 中国农业大学 | Leaf disease spot identification method and device |
-
2014
- 2014-02-28 CN CN201410071688.2A patent/CN103808265B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007171033A (en) * | 2005-12-22 | 2007-07-05 | Univ Nagoya | Indirect measuring method and system of leaf area index |
| CN101639946A (en) * | 2009-08-26 | 2010-02-03 | 北京农业信息技术研究中心 | Plant leaf three-dimensional model geometric description and curved surface reconstruction method and system |
| CN102200433A (en) * | 2011-02-25 | 2011-09-28 | 北京农业信息技术研究中心 | Device and method for measuring leaf area based on computer vision |
| CN102214306A (en) * | 2011-06-16 | 2011-10-12 | 中国农业大学 | Leaf disease spot identification method and device |
Non-Patent Citations (2)
| Title |
|---|
| 基于Android系统手机的叶面积测量方法;龚爱平等;《农业机械学报》;20130930;第44卷(第9期);第203-208页 * |
| 基于图像处理的作物病害诊断及叶片形态参数测量技术的研究;齐龙;《中国优秀硕士学位论文全文数据库 农业科技辑》;20061015(第10期);正文第12-40页，57-64页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103808265A (en) | 2014-05-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103808265B (en) | Method, device and system for measuring oilseed rape laminae and forms of sclerotium scabs synchronously | |
| CN110363858B (en) | Three-dimensional face reconstruction method and system | |
| CN109360246B (en) | Stereoscopic vision three-dimensional displacement measurement method based on synchronous subarea search | |
| CN105303616B (en) | Embossment modeling method based on single photo | |
| CN102200433B (en) | Device and method for measuring leaf area based on computer vision | |
| CN107886547B (en) | Fisheye camera calibration method and system | |
| CN113177977B (en) | Non-contact three-dimensional human body size measuring method | |
| TW201131512A (en) | Distance evaluation methods and apparatuses, and machine readable medium thereof | |
| CN103630091B (en) | Leaf area measurement method based on laser and image processing techniques | |
| CN102261892A (en) | Rapid width measurement system for tobacco shreds | |
| CN110910431B (en) | Multi-view three-dimensional point set recovery method based on monocular camera | |
| CN109191520B (en) | Plant leaf area measuring method and system based on color calibration | |
| CN103993548A (en) | Multi-camera stereoscopic shooting based pavement damage crack detection system and method | |
| CN105029691B (en) | A kind of cigarette void-end detection method based on three-dimensional reconstruction | |
| CN107016697A (en) | A kind of height measurement method and device | |
| CN109816664A (en) | A kind of three-dimensional point cloud dividing method and device | |
| CN108573503B (en) | Dish quantity measuring and calculating method based on photographing | |
| CN113192206A (en) | Three-dimensional model real-time reconstruction method and device based on target detection and background removal | |
| CN106702870A (en) | Road crack detecting system | |
| CN106871864A (en) | A kind of method that depth of building is automatically extracted based on three-dimensional satellite image | |
| CN105300302B (en) | The measuring method of Brinell circular diameter | |
| CN204202563U (en) | Fish morphological parameters self-operated measuring unit | |
| CN106469300A (en) | A kind of mottle detection recognition method | |
| CN111208138A (en) | Intelligent wood recognition device | |
| CN109978810A (en) | Detection method, system, equipment and the storage medium of mole |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 1107, block a, Nongke building, No. 11, Shuguang garden middle road, Haidian District, Beijing, 100097 Patentee after: Information technology research center of Beijing Academy of agricultural and Forestry Sciences Address before: 100097 318b, block a, agricultural science building, 11 Shuguang Huayuan Middle Road, Haidian District, Beijing Patentee before: BEIJING RESEARCH CENTER FOR INFORMATION TECHNOLOGY IN AGRICULTURE |