CN109059869A - A method of detection plant protection drone is to fruit tree spraying effect - Google Patents
A method of detection plant protection drone is to fruit tree spraying effect Download PDFInfo
- Publication number
- CN109059869A CN109059869A CN201810847304.XA CN201810847304A CN109059869A CN 109059869 A CN109059869 A CN 109059869A CN 201810847304 A CN201810847304 A CN 201810847304A CN 109059869 A CN109059869 A CN 109059869A
- Authority
- CN
- China
- Prior art keywords
- sprayed
- target
- spray
- medicament
- visible images
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/04—Interpretation of pictures
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/28—Measuring arrangements characterised by the use of optical techniques for measuring areas
- G01B11/285—Measuring arrangements characterised by the use of optical techniques for measuring areas using photoelectric detection means
Abstract
The present invention relates to plant protection to spray detection field, more particularly to a kind of detection plant protection drone is to the method for fruit tree spraying effect, the unmanned plane is equipped with machine vision photographic device and three-dimensional laser scanner, including carrying out toning processing to medicament, make the spray region for spraying target and the region that do not spray forms color difference, reuse the threedimensional model that three-dimensional laser scanner scans and establishes entirely goal tree hat to be sprayed, the visible images of target are sprayed using the acquisition of machine vision photographic device, the color difference generated on wait the target that sprays is sprayed further according to medicament obtains spray region, the threedimensional model being preced with again to visible images and goal tree to be sprayed is registrated and is merged, fused image is handled, calculate separately out the area of target to be sprayed and the region that sprays, accounting is sprayed to obtain.This method can realize real-time detection, have the characteristics that detection speed is fast, small to environmental factor dependence.
Description
Technical field
The present invention relates to plant protection to spray detection field more particularly to a kind of detection plant protection drone to fruit tree spraying effect
Method.
Background technique
Currently, there are mainly two types of the detection methods of plant protection drone medical fluid spraying effect: one is airborne machine vision is examined
Survey method, after this method uses the scheme of UAV flight's industry camera, shooting medical fluid to be atomized ejection from plant protection drone
It is freely scattering into atomized drop in the sky to hit the mark or the overall process on ground, therefrom extracts the airflight of droplet group after atomization
Track, prediction spray the arrival region of medical fluid with judgement.The detection method sprays medical liquid atomizing droplet partial size due to that can not overcome
Too small, medical fluid be colourless transparent liquid and hit by spraying spray target after the problems such as disappearing at once, cause it not detect accurately
Situation of being scattered after to medical liquid atomizing, detection error is very big, and the detection for spraying region to medical fluid is not known where to begin more.
The second is the method for place mat wet sensitive paper, the method for place mat wet sensitive paper on spraying target, although can be used for analyzing
It sprays the droplet to hit the mark after medical liquid atomizing to be scattered situation, but this method is a kind of Indirect Detecting Method, so can not return
Its on-line checking as a result, and detection time-consuming.
And lack in control method because existing plant protection drone medical fluid sprays to this skill of the real-time detection of its spraying effect
Art link, plant protection drone medical fluid spray control method and be forced based on opened loop control, can only be using continuously uninterruptedly spraying
It is single to spray scheme, due to opened loop control is to target accurate spraying effect difference, it is dirty to will cause great liquid waste, environment
Dye and inefficient the problems such as spraying.
Summary of the invention
It is in the prior art more insufficient it is an object of the invention to overcome, a kind of detection plant protection drone pair is provided
The method of fruit tree spraying effect, this method energy real-time detection spraying effect, small to the dependence of environment, detection accuracy is high.
To achieve the above object, the technical solution adopted by the present invention is that:
A kind of detection plant protection drone is provided to the method for fruit tree spraying effect, the unmanned plane is equipped with machine vision camera shooting dress
It sets and three-dimensional laser scanner, comprising the following steps:
S1: carrying out toning processing to medicament, and the medicament and target to be sprayed after making toning form color difference, and by the medicament after toning
It is placed on unmanned plane；
S2: unmanned plane is hovered over above target to be sprayed, autorotational flight one week, passes through the three-dimensional laser scanner on unmanned plane
The acquisition point information that entirely goal tree to be sprayed is preced with surface establishes what entirely goal tree to be sprayed was preced with according to the point information of acquisition
Threedimensional model, to obtain the canopy profile of target to be sprayed；
S3: the medicament of toning processing is sprayed on wait spray in target by the sprayer apparatus on unmanned plane, obtains being sprayed with medicament
Spray target, spray in target spray region and do not spray region formed color difference；
S4: after spray, by unmanned plane around 360 ° of target flights are sprayed, spray is acquired by the machine vision photographic device on unmanned plane
The visible images for applying target spray the color difference that generates on target to be sprayed (5) further according to medicament and obtain spray region, to can
The threedimensional model of light-exposed image and goal tree to be sprayed hat is registrated and is merged, and is handled fused image, is calculated
The area of target to be sprayed out；Color difference on fused image is distinguished, the area in spray region is calculated, to obtain
Accounting must be sprayed；Since medicament is different from the color for spraying target, spray region can be differentiated in visible images according to color
On position.
Carrying out Visualization to spatial information usually has two class methods: first is that the method based on image or photo；Second is that
Method based on geometry obtains point cloud data by three-dimensional laser scanner to establish threedimensional model.Method based on geometry
Working principle are as follows: the laser range finder on three-dimensional laser scanner actively emits laser pulse signal, while receiving by being tested
The signal of target surface reflection carries out ranging, to obtain the three dimensional point cloud on measured target surface, carries out to point cloud data
The processing such as noise remove, multi-veiw registration, data compaction and surface reconstruction obtain the threedimensional model on measured target surface.Wherein noise
Removal refers to the data removed in point cloud data other than measured target surface, that is, removes the point cloud data of background；Multi-veiw registration is
Finger takes multiple scan measured target surface from multi-angle of view, in order to avoid cannot one due to the shape on measured target surface is excessively complicated
It is secondary to measure all data.
In above scheme, toning processing first is carried out to medicament, the medicament and target to be sprayed after making toning are formed significantly
Color difference, convenient for distinguishing that medicament is placed on unmanned plane by the distribution for spraying medicament after toning by color, then by unmanned plane
It flies around 360 ° of target to be sprayed, by the three-dimensional laser scanner acquisition on unmanned plane, entirely goal tree to be sprayed is preced with surface
Point information carries out the processing such as noise remove, multi-veiw registration, data compaction and surface reconstruction to the point information of acquisition, obtains entire
The threedimensional model of goal tree hat to be sprayed, is divided into several sections for threedimensional model, by simply comparing, can be obtained every
Then all adjacent extreme value sides are found out, can be obtained threedimensional model according to the connectivity on extreme value side in the extreme value side in one section
Contour line obtains the canopy profile of target to be sprayed.
There are apparent color difference with target to be sprayed for the medicament of toning, so spraying spray region in target and not spraying
Region also forms apparent color difference, is handled according to the color difference in spray region and the region that do not spray visible images, is obtained
Spray region；The distribution in target is being sprayed according to spray accounting and the medicament of acquisition, is adjusting sprayer apparatus to the region that do not spray
It is sprayed, sprays error to reduce.Three-dimensional laser scanner can quickly establish complicated three to the point information of acquisition
Visualization Model is tieed up, time of measuring is saved, improves detection speed, further, since spraying spray region in target and not spraying
Region is there are apparent color difference, and the high resolution of machine vision photographic device, can be from acquisition even if the weather for having mist
The distribution for spraying medicament is identified in visible images, it is small to the dependence of environment in this way.
Preferably, it in the step S1, is mixed colours by the way that the pigment of energy self-discoloration is added into medicament medicament
Processing.The pigment of energy self-discoloration is added into medicament, stirs evenly, medicament is made to change colour, the medicament of toning is sprayed on to be painted
In medicine target, obtain being sprayed with medicament sprays target, and the spray region for spraying target will catch color, and the color in the region that sprays is held
After continuous a period of time, color will take off completely, so pigment will not be impacted to target is sprayed.
It is further preferred that the color of the pigment of the self-discoloration is red or blue in the step S1.Wait spray
The color of target is green, so the color of the pigment of self-discoloration is preferably that biggish red or blue is differed with green, so as to
The distribution in target is being sprayed in the medicament for distinguishing toning.
It preferably, further include the step that initial work processing is carried out to three-dimensional laser scanner before the step S2
Suddenly.The electricity for checking three-dimensional laser scanner establishes the item file of scanning, then carries out initial chemical industry to three-dimensional laser scanner
It deals with, its running parameter is adjusted to required state, select office operation or outdoor according to the handling situations of scanning first
Then operation selects resolution ratio, quality and scanning area, then selects chromoscan, three-dimensional laser scanner is adjusted to horizontal position,
Preparation before carrying out scanning keeps the point accuracy of information of scanning higher.
Preferably, in the step S2 three-dimensional laser scanner acquire point information include point a three-dimensional coordinate, reflectivity and
Tree crown texture information.Three-dimensional laser scanner believes the point three-dimensional coordinate, reflectivity and tree crown texture of the target to be sprayed of acquisition
Breath carries out the processing such as noise remove, multi-veiw registration, data compaction and surface reconstruction, obtains the three-dimensional of entirely goal tree hat to be sprayed
Model, the point information that three-dimensional laser scanner scanning obtains is comprehensive, the threedimensional model accuracy of resulting goal tree hat to be sprayed
It is high.
Preferably, before the step S4, further include through scaling method obtain machine vision photographic device internal reference and
Outer ginseng, and the step of distortion correction is carried out to the visible images that machine vision photographic device acquires according to internal reference and outer ginseng.It is logical
It crosses scaling method and obtains internal references and spin matrix and the translations such as focal length, picture centre, the distortion factor of machine vision photographic device
Matrix etc. is outer to join, and demarcates to machine vision photographic device, then abnormal by visible images progress of the calibration result to acquisition
Become correction.Since visible images are generating and colour cast, fuzzy, the inclination of geometric distortion, geometry etc. in transmittance process, are easily occurring no
Apparent, more true visible images can be obtained by carrying out distortion correction to visible images with the distortion of degree.Mark
Determining method can be selected gridiron pattern scaling method.
Preferably, in the step S4, environmental correction processing first is carried out to it after obtaining visible images, further according to medicament
Spraying the color difference that generates on target to be sprayed (5), treated that visible images are handled to environmental correction, is sprayed
Region.Unmanned plane, because of the interference of environmental factor, causes the visible images of acquisition unintelligible in flight course, so need pair
Visible images are pre-processed, and the influence of environmental factor is eliminated.
It is further preferred that the environmental correction processing includes being become to visible images using small echo in the step S4
The multi-resolution decomposition changed obtains high frequency section and low frequency part, and is filtered to high frequency section, carries out to low frequency part
Correction process based on color difference, to obtain clearly visible images.The visible light propagated in an atmosphere easily by cloud and mist, raindrop,
Solid particle etc. absorb, scattering and reflection, make from spray target surface reach machine vision photographic device visible light signal hair
Raw decaying, in addition, atmospheric scattering and the radiation energy of diffusion increase ambient noise to visible light signal, reduce spray target with
The contrast of environmental background makes the visible images obtained deformation, fuzzy and distortion, and therefore, it is necessary to the visible light figures to acquisition
As carrying out necessary noise smoothing and Background clutter suppression processing.The high frequency that can be obtained by the wavelet transformation to visible images
Part carries out the median filter process of local overlapped reference transformation, eliminates general noise and motion artifacts, obtains to wavelet transformation
Low frequency part, using be based on color difference blur correcting method, realize LFC low-frequency correction, effectively remove environmental colors to visual light imaging
Influence, in addition, environmental correction processing may also include to visible images carry out jitter compensation processing, that is, use two-dimension fourier
It converts the blurred picture that forms the shake in shooting process to handle, obtained result is normalized and binaryzation
Transformation, and blurred picture actual conditions are combined, it is fuzzy to estimate shake using maximum (MRT) algorithm converted based on Radon
Direction, finally obtain clearly visible images.
Preferably, in the step S4, it is seen that the threedimensional model registration and fusion of light image and goal tree to be sprayed hat
The specific steps are the threedimensional model being preced with respectively to the visible images for spraying target and goal tree to be sprayed carries out SIFT feature and mentions
It takes, searches key point on different feature spaces, match to obtain visible images using key point and goal tree to be sprayed is preced with
Threedimensional model match point alignment, to obtain each spray region on visible images and entirely goal tree to be sprayed is preced with
Then the rotation and translation matrix parameter of threedimensional model carry out figure to the threedimensional model of visible images and goal tree to be sprayed hat
As fusion, binary conversion treatment is carried out to fused image according to the canopy profile of target to be sprayed, to calculate mesh to be sprayed
Target area；Color difference differentiation is carried out to fused image, to calculate the area in spray region, sprays accounting to obtain.
By the processes such as SIFT feature extraction, matching and fusion process, binary conversion treatment, color difference differentiation, target to be sprayed is calculated
The area of area and spray region the spraying effects information such as sprays accounting, sprays distribution and spray error to obtain, to realize
Offer reference data is precisely provided.
Compared with prior art, the beneficial effect of technical solution of the present invention is: the detection method in advance adjusts medicament
Color processing makes spray region and the region formation color difference that do not spray, reuses three-dimensional laser scanner and scans and establish entire to be painted
The threedimensional model of medicine target tree crown sprays the visible images of target using the acquisition of machine vision photographic device, by analysis and
Processing obtains spraying accounting, spray distribution and spraying error for medicament, realizes the real time monitoring of medicament spraying effect；Three-dimensional swashs
Photoscanner can quickly establish complicated three-dimensional visualization model to the point information of acquisition, save time of measuring, improve inspection
Degree of testing the speed；Further, since spraying spray region in target and there are apparent color difference in the region that do not spray, and machine vision images
The high resolution of device, even if the distribution for spraying medicament can be identified from the visible images of acquisition the weather for having mist,
This method is small to the dependence of environment.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of unmanned plane of the present invention；
Fig. 2 (a) is the schematic diagram that three-dimensional laser scanner of the present invention acquires goal tree comb point information to be sprayed；
Fig. 2 (b) is that the medicament of toning processing of the present invention is sprayed on the schematic diagram of target to be sprayed；
Fig. 2 (c) is the schematic diagram that machine vision photographic device of the present invention acquisition sprays target visible light image；
Fig. 2 (d) is the schematic diagram that the toning medicament color of the present invention sprayed in target is faded completely.
Drawing reference numeral explanation: 1 unmanned plane；2 sprayer apparatus；3 machine vision photographic devices；4 three-dimensional laser scanners；5 to
Spray target；6 spray target.
Specific embodiment
For a better understanding of the present invention, below with reference to the embodiment content that the present invention is furture elucidated, but the present invention is not
It is limited only to the following examples.
Embodiment 1
A method of detection plant protection drone is to fruit tree spraying effect, and as described in Fig. 1,2, the unmanned plane 1 is equipped with machine
Vision photographic device 3 and three-dimensional laser scanner 4, comprising the following steps:
S1: carrying out toning processing to medicament, and the medicament and target 5 to be sprayed after making toning form color difference, and by the medicine after toning
Agent is placed on unmanned plane 1, to obtain the canopy profile of target 5 to be sprayed；
S2: unmanned plane 1 is hovered over above target to be sprayed, autorotational flight one week, passes through the 3 D laser scanning on unmanned plane 1
Instrument 4 acquires the point information on entirely 5 tree crown surface of target to be sprayed, and according to the point information of acquisition, establishes entirely 5 tree of target to be sprayed
The threedimensional model of hat；
S3: the medicament of toning processing is sprayed in target 5 to be sprayed by the sprayer apparatus 2 on unmanned plane 1, obtains being sprayed with medicine
Agent sprays target 6, sprays spray in target 6 region and the region formation color difference that do not spray；
S4: it after spray, by unmanned plane 1 around 360 ° of target flights are sprayed, is adopted by the machine vision photographic device 3 on unmanned plane 1
Collection sprays the visible images of target 6, and the color difference generated in target 5 to be sprayed is sprayed further according to medicament and obtains spray region,
The threedimensional model of 5 tree crown of visible images and target to be sprayed is registrated and is merged, fused image is handled,
Calculate the area of target 5 to be sprayed；Color difference on fused image is distinguished, the area in spray region is calculated,
Accounting is sprayed to obtain；Since medicament is different from the color for spraying target 6, spray region can be differentiated visible according to color
Position in light image.
Carrying out Visualization to spatial information usually has two class methods: first is that the method based on image or photo；Second is that
Method based on geometry obtains point cloud data by three-dimensional laser scanner 4 to establish threedimensional model.Method based on geometry
Working principle are as follows: the laser range finder on three-dimensional laser scanner 4 actively emits laser pulse signal, while receiving by being tested
The signal of target surface reflection carries out ranging, to obtain the three dimensional point cloud on measured target surface, carries out to point cloud data
The processing such as noise remove, multi-veiw registration, data compaction and surface reconstruction obtain the threedimensional model on measured target surface.Wherein noise
Removal refers to the data removed in point cloud data other than measured target surface, that is, removes the point cloud data of background；Multi-veiw registration is
Finger takes multiple scan measured target surface from multi-angle of view, in order to avoid cannot one due to the shape on measured target surface is excessively complicated
It is secondary to measure all data.
Toning processing first is carried out to medicament, the medicament and target 5 to be sprayed after making toning form apparent color difference, convenient for logical
Color is crossed to distinguish the distribution for spraying medicament, after toning, medicament is placed on unmanned plane 1, then by unmanned plane 1 around mesh to be sprayed
360 ° of flights are marked, the point information on entirely 5 tree crown surface of target to be sprayed is acquired by the three-dimensional laser scanner 4 on unmanned plane 1,
The processing such as noise remove, multi-veiw registration, data compaction and surface reconstruction is carried out to the point information of acquisition, obtains entirely mesh to be sprayed
Threedimensional model is divided into several sections by the threedimensional model for marking 5 tree crowns, by simply comparing, can be obtained each section
Extreme value side find out all adjacent extreme value sides then according to the connectivity on extreme value side, can be obtained the profile of threedimensional model
Line obtains the canopy profile of target 5 to be sprayed.
There are apparent color difference with target 5 to be sprayed for the medicament of toning, so spraying spray region in target 6 and not spraying
Medicine region also forms apparent color difference, is handled according to the color difference in spray region and the region that do not spray visible images, is obtained
To spray region；The distribution in target 6 is being sprayed according to spray accounting and the medicament of acquisition, adjusts sprayer apparatus 2 to not spraying
Region is sprayed, and sprays error to reduce.The point information of 4 pairs of three-dimensional laser scanner acquisitions can quickly establish that structure is complicated
Three-dimensional visualization model, save time of measuring, improve detection speed, further, since spray spray region in target 6 with not
Region spray there are apparent color difference, and the high resolution of machine vision photographic device 3, it can be from adopting even if the weather for having mist
The distribution for spraying medicament is identified in the visible images of collection, it is small to the dependence of environment in this way.
Wherein, in the step S1, medicament is carried out at toning by the way that the pigment of energy self-discoloration is added into medicament
Reason.The pigment of energy self-discoloration is added into medicament, stirs evenly, medicament is made to change colour, the medicament of toning is sprayed on wait spray
In target 5, obtain being sprayed with medicament sprays target 6, and the spray region for spraying target 6 will catch color, the color in the region that sprays
After continuing for some time, color will take off completely, so pigment will not be impacted to target 6 is sprayed.
Specifically, in the step S1, the color of the pigment of the self-discoloration is red or blue.The face of self-discoloration
Material can be the pigment being mixed to form by NaClO solution and ordinary ink by 3:7；The color of target 5 to be sprayed is green, so from
The color of the dynamic pigment to fade is preferably that biggish red or blue is differed with green, is spraying mesh in order to distinguish the medicament of toning
Distribution on mark 6.
It wherein, further include the step that initial work processing is carried out to three-dimensional laser scanner 4 before the step S2
Suddenly.The electricity for checking three-dimensional laser scanner 4, establishes the item file of scanning, then initialize to three-dimensional laser scanner 4
Its running parameter is adjusted to required state by work disposal, selects office operation or room according to the handling situations of scanning first
Then outer operation selects resolution ratio, quality and scanning area, then selects chromoscan, three-dimensional laser scanner 4 is adjusted to horizontal position
It sets, the preparation before carrying out scanning keeps the point accuracy of information of scanning higher.
Wherein, the point information that three-dimensional laser scanner 4 acquires in the step S2 includes point three-dimensional coordinate, reflectivity and tree
It is preced with texture information.Point three-dimensional coordinate, reflectivity and the tree crown texture letter of the target to be sprayed 5 of 4 pairs of three-dimensional laser scanner acquisitions
Breath carries out the processing such as noise remove, multi-veiw registration, data compaction and surface reconstruction, obtains the three of entirely 5 tree crown of target to be sprayed
Dimension module, the point information that the scanning of three-dimensional laser scanner 4 obtains is comprehensive, the threedimensional model essence of resulting 5 tree crown of target to be sprayed
Exactness is high.
It wherein, further include that the internal reference of machine vision photographic device 3 and outer is obtained through scaling method before the step S4
Ginseng, and the step of distortion correction is carried out to the visible images that machine vision photographic device 3 acquires according to internal reference and outer ginseng.Pass through
Scaling method obtains the internal references such as focal length, picture centre, the distortion factor of machine vision photographic device 3 and spin matrix and translation square
The outer ginseng such as battle array, demarcates machine vision photographic device 3, then distorted by visible images of the calibration result to acquisition
Correction.Since visible images are generating and the differences such as colour cast, fuzzy, the inclination of geometric distortion, geometry in transmittance process, are easily occurring
The distortion of degree can obtain apparent, more true visible images by carrying out distortion correction to visible images.Calibration
Gridiron pattern scaling method can be selected in method.
Wherein, in the step S4, environmental correction processing first is carried out to it after obtaining visible images, is sprayed further according to medicament
Applying the color difference that generates in target 5 to be sprayed, treated that visible images are handled to environmental correction, obtains spray region.
Unmanned plane 1, because of the interference of environmental factor, causes the visible images of acquisition unintelligible, so need to be to visible in flight course
Light image is pre-processed, and the influence of environmental factor is eliminated.
Specifically, in the step S4, the environmental correction processing includes to visible images using the more of wavelet transformation
Scale Decomposition obtains high frequency section and low frequency part, and is filtered to high frequency section, carries out low frequency part based on color
The correction process of difference, to obtain clearly visible images.The visible light propagated in an atmosphere is easily by cloud and mist, raindrop, solid
Grain etc. absorbs, scattering and reflection, makes to decline from the visible light signal for spraying 6 surface of target and reaching machine vision photographic device 3
Subtract, in addition, atmospheric scattering and the radiation energy of diffusion increase ambient noise to visible light signal, reduces and spray target 6 and ring
The contrast of border background makes the visible images obtained deformation, fuzzy and distortion, and therefore, it is necessary to the visible images to acquisition
Carry out necessary noise smoothing and Background clutter suppression processing.The radio-frequency head that can be obtained by the wavelet transformation to visible images
Divide the median filter process for carrying out local overlapped reference transformation, eliminate general noise and motion artifacts, wavelet transformation is obtained
Low frequency part realizes LFC low-frequency correction, effectively removes environmental colors to visual light imaging using color difference blur correcting method is based on
It influences, in addition, environmental correction processing, which may also include, carries out jitter compensation processing to visible images, i.e., is become using two-dimension fourier
It changes the blurred picture for forming the shake in shooting process to handle, obtained result is normalized and binaryzation becomes
It changes, and combines blurred picture actual conditions, estimate what shake obscured using maximum (MRT) algorithm converted based on Radon
Direction finally obtains clearly visible images.
Wherein, in the step S4, it is seen that the threedimensional model of 5 tree crown of light image and target to be sprayed is registrated and the tool of fusion
Body step is to carry out SIFT feature to the threedimensional model for 5 tree crown of visible images and target to be sprayed for spraying target 6 respectively and mention
It takes, searches key point on different feature spaces, match to obtain 5 tree crown of visible images and target to be sprayed using key point
Threedimensional model match point alignment, to obtain on visible images each spray region and entirely 5 tree crown of target to be sprayed
Threedimensional model rotation and translation matrix parameter, then to the threedimensional model of visible images and 5 tree crown of target to be sprayed into
Row image co-registration carries out binary conversion treatment to fused image according to the canopy profile of target 5 to be sprayed, to be painted to calculate
The area of medicine target 5；Color difference differentiation is carried out to fused image, to calculate the area in spray region, to be sprayed
Accounting.By the processes such as SIFT feature extraction, matching and fusion process, binary conversion treatment, color difference differentiation, mesh to be sprayed is calculated
The area of mark 5 and the area in spray region, the spraying effects information such as spray accounting, spray distribution and spray error to obtain,
Reference data is provided to realize precisely to spray.
The above embodiment of the present invention is only intended to clearly illustrate examples made by the present invention, and is not to of the invention
The restriction of embodiment.For those of ordinary skill in the art, it can also be made on the basis of the above description
Its various forms of variation or variation.There is no necessity and possibility to exhaust all the enbodiments.It is all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should be included in the protection model of the claims in the present invention within mind and principle
Within enclosing.
Claims (9)
1. a kind of detection plant protection drone is to the method for fruit tree spraying effect, which is characterized in that the unmanned plane (1) is equipped with machine
Device vision photographic device (3) and three-dimensional laser scanner (4), comprising the following steps:
S1: carrying out toning processing to medicament, and the medicament and target to be sprayed (5) after making toning form color difference, and will be after toning
Medicament is placed on unmanned plane (1)；
S2: unmanned plane (1) is hovered over above target to be sprayed (5), autorotational flight one week, passes through the three-dimensional on unmanned plane (1)
Laser scanner (4) acquires the point information on entirely target (5) tree crown surface to be sprayed, and according to the point information of acquisition, establishes entire
The threedimensional model of target (5) tree crown to be sprayed, to obtain the canopy profile of target to be sprayed (5)；
S3: the medicament of toning processing is sprayed on target to be sprayed (5) by the sprayer apparatus (2) on unmanned plane (1), is obtained
Be sprayed with medicament sprays target (6), sprays spray on target (6) region and the region formation color difference that do not spray；
S4: it after spray, by unmanned plane (1) around (6) 360 ° of target flights are sprayed, is imaged by the machine vision on unmanned plane (1)
Device (3) acquires the visible images for spraying target (6), and the color difference generated on target to be sprayed (5) is sprayed further according to medicament
Spray region is obtained, the threedimensional model of visible images and target to be sprayed (5) tree crown is registrated and is merged, after fusion
Image handled, calculate the area of target to be sprayed (5)；Color difference on fused image is distinguished, is calculated
The area in spray region out sprays accounting to obtain.
2. method of the detection plant protection drone to fruit tree spraying effect according to claim 1, which is characterized in that the step
In S1, toning processing is carried out to medicament by the way that the pigment of energy self-discoloration is added into medicament.
3. method of the detection plant protection drone to fruit tree spraying effect according to claim 2, which is characterized in that the step
In S1, the color of the pigment of the self-discoloration is red or blue.
4. method of the detection plant protection drone to fruit tree spraying effect according to claim 1, which is characterized in that in the step
Before rapid S2, further include the steps that carrying out initial work processing to three-dimensional laser scanner (4).
5. method of the detection plant protection drone to fruit tree spraying effect according to claim 1, which is characterized in that the step
The point information that three-dimensional laser scanner (4) acquires in S2 includes three-dimensional coordinate, reflectivity and the tree crown texture information of point.
6. method of the detection plant protection drone to fruit tree spraying effect according to claim 1, which is characterized in that in the step
It further include the internal reference that machine vision photographic device (3) are obtained through scaling method and outer ginseng, and according to internal reference and outer ginseng before rapid S4
The step of distortion correction is carried out to the visible images of machine vision photographic device (3) acquisition.
7. method of the detection plant protection drone to fruit tree spraying effect according to claim 1, which is characterized in that the step
In S4, environmental correction processing first is carried out to it after obtaining visible images, sprays further according to medicament and is produced on target to be sprayed (5)
Treated that visible images are handled to environmental correction for raw color difference, obtains spray region.
8. method of the detection plant protection drone to fruit tree spraying effect according to claim 7, which is characterized in that the step
In S4, environmental correction processing includes obtaining high frequency section and low using the multi-resolution decomposition of wavelet transformation to visible images
Frequency part, and high frequency section is filtered, the correction process based on color difference is carried out to low frequency part, to obtain clearly
Visible images.
9. method of the detection plant protection drone to fruit tree spraying effect according to claim 1, which is characterized in that the step
In S4, it is seen that the threedimensional model of light image and target to be sprayed (5) tree crown registration and fusion the specific steps are respectively to spraying
The visible images of target (6) and the threedimensional model of target to be sprayed (5) tree crown carry out SIFT feature extraction, in different features
Key point is spatially searched, the threedimensional model for obtaining visible images and target to be sprayed (5) tree crown is matched using key point
Match point alignment, to obtain on visible images each spray region and the entirely threedimensional model of target (5) tree crown to be sprayed
Rotation and translation matrix parameter, then the threedimensional model progress image of visible images and target to be sprayed (5) tree crown is melted
It closes, binary conversion treatment is carried out to fused image according to the canopy profile of target to be sprayed (5), to calculate mesh to be sprayed
Mark the area of (5)；Color difference differentiation is carried out to fused image, to calculate the area in spray region, is accounted for obtain to spray
Than.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810847304.XA CN109059869B (en) | 2018-07-27 | 2018-07-27 | Method for detecting spraying effect of plant protection unmanned aerial vehicle on fruit trees |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810847304.XA CN109059869B (en) | 2018-07-27 | 2018-07-27 | Method for detecting spraying effect of plant protection unmanned aerial vehicle on fruit trees |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109059869A true CN109059869A (en) | 2018-12-21 |
| CN109059869B CN109059869B (en) | 2020-07-21 |
Family
ID=64835818
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810847304.XA Active CN109059869B (en) | 2018-07-27 | 2018-07-27 | Method for detecting spraying effect of plant protection unmanned aerial vehicle on fruit trees |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109059869B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111008983A (en) * | 2019-12-06 | 2020-04-14 | 国网湖南省电力有限公司 | Crown multi-threshold segmentation method and system applicable to power transmission line tree height monitoring device |
| US20210176912A1 (en) * | 2019-12-16 | 2021-06-17 | Cnh Industrial America Llc | System and method for assessing agricultural operation performance based on image data of processed and unprocessed portions of the field |
| CN113331167A (en) * | 2021-08-02 | 2021-09-03 | 西咸新区千兆智能科技有限公司 | Control system for spraying pesticide in agricultural greenhouse and spraying method thereof |
| CN113822987A (en) * | 2021-09-22 | 2021-12-21 | 杭州趣村游文旅集团有限公司 | Automatic color matching method between adjacent three-dimensional live-action models |
| CN115184372A (en) * | 2022-07-13 | 2022-10-14 | 水利部交通运输部国家能源局南京水利科学研究院 | Intelligent detection device and method for micro-crack fluorescence permeation of inaccessible part of concrete structure |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000280979A (en) * | 1999-03-30 | 2000-10-10 | Kawasaki Heavy Ind Ltd | High speed planing boat |
| WO2011015989A1 (en) * | 2009-08-06 | 2011-02-10 | S.I.A. Società Idee Avioniche S.R.L. | A washing unit for turboprops of aircrafts |
| CN102800083A (en) * | 2012-06-19 | 2012-11-28 | 中国农业大学 | Crop spraying positioning method based on binocular vision gridding partition matching algorithm |
| DE202015000218U1 (en) * | 2015-01-12 | 2016-04-13 | Volker Kruse | Device for cleaning wall and bundle heating surfaces of an incinerator online |
| CN105753509A (en) * | 2016-02-01 | 2016-07-13 | 浙江大学 | Humidity sensing ceramic, and preparation method and application thereof |
| CN106774465A (en) * | 2016-12-14 | 2017-05-31 | 江苏蒲公英无人机有限公司 | A kind of unmanned aerial vehicle (UAV) control method for fruit tree spray |
| CN106814745A (en) * | 2017-03-22 | 2017-06-09 | 常熟理工学院 | UAV Intelligent medicine supplies spray method and system |
| CN106956778A (en) * | 2017-05-23 | 2017-07-18 | 广东容祺智能科技有限公司 | A kind of unmanned plane pesticide spraying method and system |
| JPWO2017081757A1 (en) * | 2015-11-10 | 2018-03-08 | 富士通フロンテック株式会社 | Paper sheet handling apparatus and paper sheet determination method |
| CN108090893A (en) * | 2017-11-23 | 2018-05-29 | 华南农业大学 | Plant protection unmanned plane medicament sprinkling optimization method and device based on thermal infrared identification |
| CN108255194A (en) * | 2018-01-08 | 2018-07-06 | 江苏工程职业技术学院 | A kind of agricultural plant protection unmanned plane precise positioning spraying control method based on UWB |
-
2018
- 2018-07-27 CN CN201810847304.XA patent/CN109059869B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000280979A (en) * | 1999-03-30 | 2000-10-10 | Kawasaki Heavy Ind Ltd | High speed planing boat |
| WO2011015989A1 (en) * | 2009-08-06 | 2011-02-10 | S.I.A. Società Idee Avioniche S.R.L. | A washing unit for turboprops of aircrafts |
| CN102800083A (en) * | 2012-06-19 | 2012-11-28 | 中国农业大学 | Crop spraying positioning method based on binocular vision gridding partition matching algorithm |
| DE202015000218U1 (en) * | 2015-01-12 | 2016-04-13 | Volker Kruse | Device for cleaning wall and bundle heating surfaces of an incinerator online |
| JPWO2017081757A1 (en) * | 2015-11-10 | 2018-03-08 | 富士通フロンテック株式会社 | Paper sheet handling apparatus and paper sheet determination method |
| CN105753509A (en) * | 2016-02-01 | 2016-07-13 | 浙江大学 | Humidity sensing ceramic, and preparation method and application thereof |
| CN106774465A (en) * | 2016-12-14 | 2017-05-31 | 江苏蒲公英无人机有限公司 | A kind of unmanned aerial vehicle (UAV) control method for fruit tree spray |
| CN106814745A (en) * | 2017-03-22 | 2017-06-09 | 常熟理工学院 | UAV Intelligent medicine supplies spray method and system |
| CN106956778A (en) * | 2017-05-23 | 2017-07-18 | 广东容祺智能科技有限公司 | A kind of unmanned plane pesticide spraying method and system |
| CN108090893A (en) * | 2017-11-23 | 2018-05-29 | 华南农业大学 | Plant protection unmanned plane medicament sprinkling optimization method and device based on thermal infrared identification |
| CN108255194A (en) * | 2018-01-08 | 2018-07-06 | 江苏工程职业技术学院 | A kind of agricultural plant protection unmanned plane precise positioning spraying control method based on UWB |
Non-Patent Citations (2)
| Title |
|---|
| SANTIAGO PLANAS DE MARTÍ, AGR. ENG. PHD: "Optimizing pesticide spray application in tree crops", 《AMERICAN SOCIETY OF AGRICULTURAL AND BIOLOGICAL ENGINEERS》 * |
| 张瑞瑞: "航空施药雾滴沉积特性光谱分析检测系统研发与应用", 《农业工程学报》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111008983A (en) * | 2019-12-06 | 2020-04-14 | 国网湖南省电力有限公司 | Crown multi-threshold segmentation method and system applicable to power transmission line tree height monitoring device |
| US20210176912A1 (en) * | 2019-12-16 | 2021-06-17 | Cnh Industrial America Llc | System and method for assessing agricultural operation performance based on image data of processed and unprocessed portions of the field |
| CN113331167A (en) * | 2021-08-02 | 2021-09-03 | 西咸新区千兆智能科技有限公司 | Control system for spraying pesticide in agricultural greenhouse and spraying method thereof |
| CN113331167B (en) * | 2021-08-02 | 2021-11-19 | 南京卡飞软件技术有限公司 | Control system for spraying pesticide in agricultural greenhouse and spraying method thereof |
| CN113822987A (en) * | 2021-09-22 | 2021-12-21 | 杭州趣村游文旅集团有限公司 | Automatic color matching method between adjacent three-dimensional live-action models |
| CN115184372A (en) * | 2022-07-13 | 2022-10-14 | 水利部交通运输部国家能源局南京水利科学研究院 | Intelligent detection device and method for micro-crack fluorescence permeation of inaccessible part of concrete structure |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109059869B (en) | 2020-07-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109059869A (en) | A method of detection plant protection drone is to fruit tree spraying effect | |
| CN107063949B (en) | Method, device and system for measuring droplet deposition distribution | |
| CN112541921B (en) | Urban green land vegetation information data accurate determination method | |
| CA2721891C (en) | Optronic system and method dedicated to identification for formulating three-dimensional images | |
| Haala et al. | Combination of terrestrial laser scanning with high resolution panoramic images for investigations in forest applications and tree species recognition | |
| Bonifazi et al. | Characterisation of flotation froth colour and structure by machine vision | |
| CN110264570A (en) | A kind of autonomous cruising inspection system in forest land based on unmanned plane | |
| CN114485477B (en) | Icing three-dimensional shape online measuring method and device | |
| CN107121192A (en) | A kind of camera lens scan pattern Hyperspectral imager and rotor wing unmanned aerial vehicle | |
| CN107976428A (en) | The lossless real-time detection method and device of aquatic products chemical pollutant | |
| CN109241837A (en) | A kind of plant protection drone precisely sprays control method | |
| CN110363186A (en) | A kind of method for detecting abnormality, device and computer storage medium, electronic equipment | |
| Wen et al. | A new spray deposition pattern measurement system based on spectral analysis of a fluorescent tracer | |
| EP3580725B1 (en) | A method of matching colours | |
| Luo et al. | Portable 4D snapshot hyperspectral imager for fastspectral and surface morphology measurements | |
| CN109285141A (en) | A kind of real-time detection method of plant protection drone spraying effect | |
| EP3237887A1 (en) | Apparatus and method for analysing a surface | |
| CN115546320B (en) | Camera calibration and online ice shape measurement method for measuring ice shape | |
| CN205808362U (en) | A kind of Model Reconstruction device | |
| CN206804512U (en) | Multispectral camera device | |
| CN109544521A (en) | The method for registering of passive millimeter wave image and visible images in a kind of human body safety check | |
| Polewski et al. | Segmentation of single standing dead trees in high-resolution aerial imagery with generative adversarial network-based shape priors | |
| Sun et al. | Design and application of a system for droplet-size measurement in the field based on micro-distance imaging technology | |
| De Cock et al. | Droplet size distribution measurements of ISO nozzles by shadowgraphy method | |
| JP6845386B2 (en) | Coating film analysis device and coating film analysis method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |