CN107886114A - Feature extracting method, recognition methods, device and the computer equipment of plant leaf blade - Google Patents
Feature extracting method, recognition methods, device and the computer equipment of plant leaf blade Download PDFInfo
- Publication number
- CN107886114A CN107886114A CN201711025461.4A CN201711025461A CN107886114A CN 107886114 A CN107886114 A CN 107886114A CN 201711025461 A CN201711025461 A CN 201711025461A CN 107886114 A CN107886114 A CN 107886114A
- Authority
- CN
- China
- Prior art keywords
- pixel
- leaf
- image
- lbp
- blade
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/758—Involving statistics of pixels or of feature values, e.g. histogram matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/48—Extraction of image or video features by mapping characteristic values of the pattern into a parameter space, e.g. Hough transformation
Abstract
The invention discloses a kind of feature extracting method of plant leaf blade, the recognition methods of plant leaf blade, device, computer equipment and storage medium.Wherein feature extracting method includes:Leaf samples image is obtained, and binary conversion treatment is carried out to leaf samples image;Determine the circle shaped neighborhood region of each pixel in the leaf samples image after binary conversion treatment, and multiple rotary circle shaped neighborhood region is to calculate the target local binary patterns LBP values of each pixel;Calculate probability distribution of the target LBP values of each pixel in leaf samples image;LBP characteristic informations using probability distribution as leaf samples image.This method can overcome the error of recognition result when blade is distributed with different shape or different angle in image well, improve the degree of accuracy of feature extraction in blade, can also improve the accuracy of leaf recognition.
Description
Technical field
The present invention relates to technical field of image processing, more particularly to a kind of feature extracting method of plant leaf blade, identification side
Method, device, computer equipment and computer-readable recording medium.
Background technology
The classification of plant and the general local feature for choosing plant of identification, the leaf of such as plant, flower, fruit, stem, paper slip spy
Sign.These organs have respective classification value, but have compared plant others organ, time-to-live of plant leaf blade compared with
It is long, all can be more convenient within the most of the time of 1 year collect, so the identification feature and understanding frequently as plant are planted
Thing referring especially to organ;Simultaneously leaf is an extraordinary index of morphological variation and the differentiation of studying plant species, because
This identification based on blade is a kind of most direct effective and simplest method of plant of identification.
In the identification technology of plant leaf blade it is most important be exactly characteristic information in blade extraction.In correlation technique, generally
It is that LBP (Local Binary are extracted from leaf image by modes such as certain noise reduction, weighting aminated polyepichlorohydrins
Patterns, local binary patterns) feature, and by the LBP features to realize the identification of plant leaf blade.But aforesaid way
Feature extraction can be only carried out to the plant leaf blade under current angular, and by this feature to realize the identification of blade, and when logical
Leaf image obtained from different angle is shot to same type of plant leaf blade is crossed, if utilizing above-mentioned identification
If the image of the same type plant leaf blade under different angle is identified mode, it is possible that different identification
As a result, that is to say, that for same type leaf image under rotational case, can have identification by above-mentioned identification method
As a result there is the problem of error, identification accuracy is low.
The content of the invention
The purpose of the present invention is intended at least solve one of above-mentioned technical problem to a certain extent.
Therefore, first purpose of the present invention is to propose a kind of feature extracting method of plant leaf blade.This method can be with
The error of recognition result when blade is distributed with different shape or different angle in image is overcome well, improves feature in blade
The degree of accuracy of extraction, the accuracy of leaf recognition can also be improved.
Second object of the present invention is to propose a kind of recognition methods of plant leaf blade.
Third object of the present invention is to propose a kind of feature deriving means of plant leaf blade.
Fourth object of the present invention is to propose a kind of identification device of plant leaf blade.
The 5th purpose of the present invention is to propose a kind of computer equipment.
The 6th purpose of the present invention is to propose a kind of non-transitorycomputer readable storage medium.
The 7th purpose of the present invention is to propose another computer equipment.
The 8th purpose of the present invention is to propose another non-transitorycomputer readable storage medium.
To reach above-mentioned purpose, the feature extracting method for the plant leaf blade that first aspect present invention embodiment proposes, including:
Leaf samples image is obtained, and binary conversion treatment is carried out to the leaf samples image;Determine the sample leaf after binary conversion treatment
The circle shaped neighborhood region of each pixel in picture, and circle shaped neighborhood region described in multiple rotary is to calculate the mesh of each pixel
Mark local binary patterns LBP values;Calculate the target LBP values of each pixel probability in the leaf samples image point
Cloth;LBP characteristic informations using the probability distribution as the leaf samples image.
To reach above-mentioned purpose, the recognition methods for the plant leaf blade that second aspect of the present invention embodiment proposes, including:Obtain
Leaf image to be identified;Feature extraction is carried out to the leaf image to be identified according to default feature extracting method, with
The LBP characteristic informations of the leaf image to be identified are obtained, wherein, the default feature extracting method is the present invention first
Feature extracting method described in aspect embodiment;By the LBP characteristic informations of the leaf image to be identified and prestore
The LBP characteristic informations of the leaf samples image are matched;If matching result meets preparatory condition, according to the sample leaf
The LBP characteristic informations of picture and the corresponding relation prestored, determine blade in the leaf image to be identified
Species.
To reach above-mentioned purpose, the feature deriving means for the plant leaf blade that third aspect present invention embodiment proposes, including:
Acquisition module, for obtaining leaf samples image;Binary conversion treatment module, for carrying out binaryzation to the leaf samples image
Processing;First computing module, for determining the circle shaped neighborhood region of each pixel in the leaf samples image after binary conversion treatment, and
Circle shaped neighborhood region described in multiple rotary is to calculate the target LBP values of each pixel;Second computing module, for calculating
State probability distribution of the target LBP values of each pixel in the leaf samples image;First determining module, for by described in
LBP characteristic information of the probability distribution as the leaf samples image.
To reach above-mentioned purpose, the identification device for the plant leaf blade that fourth aspect present invention embodiment proposes, including:Obtain
Module, for obtaining leaf image to be identified;Characteristic extracting module, for being treated according to default feature extracting method to described
The leaf image of identification carries out feature extraction, to obtain the LBP characteristic informations of the leaf image to be identified, wherein, it is described
Default feature extracting method is the feature extracting method described in first aspect present invention embodiment;Matching module, for by institute
The LBP characteristic informations of the LBP characteristic informations and the leaf samples image prestored of stating leaf image to be identified are carried out
Matching;Determining module, for when matching result meets preparatory condition, according to the LBP characteristic informations of the leaf samples image
With the corresponding relation prestored, the species of blade in the leaf image to be identified is determined.
To reach above-mentioned purpose, computer equipment that fifth aspect present invention embodiment proposes, including:Memory, processing
Device and it is stored in the computer program that can be run on the memory and on the processor, journey described in the computing device
During sequence, the feature extracting method of the plant leaf blade described in first aspect present invention embodiment is realized.
To reach above-mentioned purpose, non-transitorycomputer readable storage medium that sixth aspect present invention embodiment proposes,
Computer program is stored thereon with, the plant described in first aspect present invention embodiment is realized when described program is executed by processor
The feature extracting method of blade.
To reach above-mentioned purpose, computer equipment that seventh aspect present invention embodiment proposes, including:Memory, processing
Device and it is stored in the computer program that can be run on the memory and on the processor, journey described in the computing device
During sequence, the recognition methods of the plant leaf blade described in second aspect of the present invention embodiment is realized.
To reach above-mentioned purpose, non-transitorycomputer readable storage medium that eighth aspect present invention embodiment proposes,
Computer program is stored thereon with, the plant described in second aspect of the present invention embodiment is realized when described program is executed by processor
The recognition methods of blade.
Feature extracting method, recognition methods, device, computer equipment and the meter of plant leaf blade according to embodiments of the present invention
Calculation machine readable storage medium storing program for executing, the target of each pixel can be calculated by the circle shaped neighborhood region of each pixel of multiple rotary
LBP values, and probability distribution of the target LBP values of each pixel in leaf samples image is calculated, and then the probability is divided
LBP characteristic information of the cloth as leaf samples image, blade can be overcome in image well with different shape or different angle
The error of recognition result during distribution, improves the degree of accuracy of feature extraction in blade, so, is extracted by this feature extracting mode
Characteristic information in leaf image, and when carrying out leaf recognition according to this feature information, the accuracy of leaf recognition can be improved.
The additional aspect of the present invention and advantage will be set forth in part in the description, and will partly become from the following description
Obtain substantially, or recognized by the practice of the present invention.
Brief description of the drawings
Of the invention above-mentioned and/or additional aspect and advantage will become from the following description of the accompanying drawings of embodiments
Substantially and it is readily appreciated that, wherein:
Fig. 1 is the flow chart of the feature extracting method of plant leaf blade according to an embodiment of the invention;
Fig. 2 is the flow chart of the feature extracting method of plant leaf blade according to embodiments of the present invention;
Fig. 3 is the structural representation of the feature deriving means of plant leaf blade according to an embodiment of the invention;
Fig. 4 is the structural representation according to the feature deriving means of the plant leaf blade of a specific embodiment of the invention;
Fig. 5 is the structural representation according to the feature deriving means of the plant leaf blade of another specific embodiment of the invention;
Fig. 6 is the flow chart of the recognition methods of plant leaf blade according to an embodiment of the invention;
Fig. 7 is the flow chart of the recognition methods of plant leaf blade according to embodiments of the present invention;
Fig. 8 is the structural representation of the identification device of plant leaf blade according to an embodiment of the invention;
Fig. 9 is the structural representation of computer equipment according to an embodiment of the invention;
Figure 10 is the structural representation of computer equipment in accordance with another embodiment of the present invention.
Embodiment
Embodiments of the invention are described below in detail, the example of the embodiment is shown in the drawings, wherein from beginning to end
Same or similar label represents same or similar element or the element with same or like function.Below with reference to attached
The embodiment of figure description is exemplary, it is intended to for explaining the present invention, and is not considered as limiting the invention.
Below with reference to the accompanying drawings feature extracting method, recognition methods, device, the meter of the plant leaf blade of the embodiment of the present invention are described
Calculate machine equipment and computer-readable recording medium.
Fig. 1 is the flow chart of the feature extracting method of plant leaf blade according to an embodiment of the invention.Need what is illustrated
It is that the feature that the feature extracting method of the plant leaf blade of the embodiment of the present invention can be applied to the plant leaf blade of the embodiment of the present invention carries
Device is taken, this feature extraction element can be configured on computer equipment.Wherein, in an embodiment of the present invention, the computer
Equipment can be server or terminal device, have for example, the terminal device can be mobile phone, tablet personal computer, personal digital assistant etc.
The hardware device of various operating systems.
As shown in figure 1, the feature extracting method of the plant leaf blade can include:
S110, leaf samples image is obtained, and binary conversion treatment is carried out to leaf samples image.
Alternatively, the leaf samples image can be that user uploads, for example, user by capture apparatus to different plants
Blade substantial amounts of sample leaf image obtained from being shot, and these sample leaf images are uploaded;
Or the leaf samples image can be user as obtained from scanning for internet.
After leaf samples image is obtained, noise reduction algorithm can be first passed through preliminary treatment is carried out to leaf samples image, with
The redundancy noise in the leaf samples image is removed, and Global thresholding can be used to carry out leaf samples image at binaryzation
Reason, for example, each pixel point value can be 0 or 1 in result, wherein, as a result the element for 1 is blade effective coverage.Its
In, in an embodiment of the present invention, above-mentioned noise reduction algorithm may include but be not limited to neighborhood averaging, median filtering algorithm, low pass
Filtering algorithm etc..
S120, determine the circle shaped neighborhood region of each pixel in the leaf samples image after binary conversion treatment, and multiple rotary
The circle shaped neighborhood region is to calculate the target local binary patterns LBP values of each pixel.
Alternatively, after binary conversion treatment is carried out to leaf samples image, the sample after binary conversion treatment can first be determined
Each pixel in leaf image, and the circle shaped neighborhood region of each pixel is determined, each pixel circle shaped neighborhood region is constantly rotated,
And the target LBP of each pixel is calculated according to the gray value of the pixel in postrotational each pixel circle shaped neighborhood region
Value.
As a kind of example, as shown in Fig. 2 the multiple rotary circle shaped neighborhood region is to calculate the target of each pixel
The implementation process of LBP values may include following steps:
S210, for each pixel, calculate multiple in the gray value and each pixel circle shaped neighborhood region of each pixel
The gray value of pixel.
Alternatively, multiple pixels in each pixel circle shaped neighborhood region, half with the circle shaped neighborhood region of each pixel
Footpath is relevant with user-defined value, for example, when the radius for the circle shaped neighborhood region of each pixel is 1, it gives tacit consent to bag
8 adjacent pixels are included, user can also define the adjacent pixel of upper and lower, left and right four direction therein.
After the circle shaped neighborhood region of each pixel is determined, the gray value of each pixel and each pixel can be calculated
The gray value of multiple pixels in circle shaped neighborhood region.It is appreciated that the gray value of the pixel, the color for referring to black white image midpoint is deep
Degree, scope is typically from 0~255, and white is 255, black 0.
S220, according to the gray value of multiple pixels in the gray value of each pixel and each pixel circle shaped neighborhood region, meter
Calculate the first LBP values of each pixel.
When alternatively, using the radius of the circle shaped neighborhood region for each pixel as 1, its acquiescence include 8 it is adjacent
Exemplified by pixel, using the gray value using center pixel (i.e. described each pixel) as threshold value, by the gray value of 8 pixels around
Compared with it, if the gray value of the pixel of surrounding is less than the gray value of center pixel, the location of pixels is flagged as 0,
If the gray value of the pixel of surrounding is more than or equal to the gray value of center pixel, the location of pixels is marked as 1;Will be labeled
The value (i.e. 0 or 1) of the pixel afterwards multiplied by weight with correspondence position pixel respectively, 8 products and be the LBP of the neighborhood
Value (i.e. the first LBP values of described each pixel).
S230, multiple rotary is carried out to circle shaped neighborhood region, and after each rotation, calculate the current ash of each pixel respectively
The current grayvalue of multiple pixels in angle value and each pixel circle shaped neighborhood region.
S240, according to the current ash of multiple pixels in the current grayvalue of each pixel and each pixel circle shaped neighborhood region
Angle value, calculate multiple 2nd LBP values of each pixel.
S250, the minimum value in the first LBP values and multiple 2nd LBP values is obtained, and using minimum value as each pixel
Target LBP values.
That is, the circle shaped neighborhood region that each pixel can be constantly rotated to each pixel is a series of to obtain
LBP values (the first i.e. above-mentioned LBP values and multiple 2nd LBP values), and chosen from the first LBP values and multiple 2nd LBP values
Target LBP value of the minimum value as each pixel.Thus, by the way that minimum LBP values to be used as to the LBP features of extracted blade
Information, error when blade is distributed with different shape in image can be overcome well, so as to improve the accuracy of identification.
S130, calculate probability distribution of the target LBP values of each pixel in leaf samples image.
Alternatively, after the target LBP values of each pixel are obtained, each target LBP values can be calculated in sample leaf
Probability distribution in picture.For example, one one-dimension array Static [i] of definable, wherein, between i value is 0~255
Integer (i.e. 0≤i≤255), the probability of 256 LBP ranks can be stored in one-dimension array Static [i].That is, sample
Each pixel is respectively provided with respective gray value in this leaf image, and the scope of gray value is [0,255], so, array
Index i in Static [i] can represent the gray value of current pixel point, and value corresponding to it can represent the target of the current pixel point
LBP values.So, an one-dimension array Static [i], 256 LBP in being stored in the array can be established out according to the storage mode
The probability of rank.
Alternatively, after probability distribution of the target LBP values of each pixel in leaf samples image is obtained, may be used also
Generate the LBP probability distribution histograms of leaf samples image.As a kind of example, can be generated by following calculation formula (1)
The LBP probability distribution histograms of the leaf samples image:
FLBPHist [i]=100.0f*Static [i]/total (1)
Wherein, fLBPHist [i] is the LBP values corresponding to ith pixel point in the histogram, and f is floating number, Static
[i] is target LBP values corresponding to ith pixel point, and total is the total number of pixel in leaf samples image.
S140, the LBP characteristic informations using probability distribution as leaf samples image.
Alternatively, after probability distribution of the target LBP values of each pixel in leaf samples image is obtained, can incite somebody to action
LBP characteristic information of the probability distribution as leaf samples image.Or exist in the target LBP values for obtaining each pixel
, can should when also generating the LBP probability distribution histograms of leaf samples image after probability distribution in leaf samples image
LBP characteristic information of the LBP probability distribution histogram as leaf samples image.
In order to realize the identification function of plant leaf blade, the recognition accuracy of blade is improved, further, the one of the present invention
In individual embodiment, the species of leaf samples is may further determine that, and the corresponding pass established between the species of leaf samples and probability distribution
System, and the species, probability distribution and corresponding relation of leaf samples are stored.
That is, it may be determined that leaf samples belong to the plant of any classification, and establish the species of the leaf samples with
Between the LBP characteristic informations (probability distribution of the target LBP values of i.e. described each pixel in leaf samples image)
Corresponding relation, and the corresponding relation, the species of leaf samples, LBP characteristic informations are stored, so as in practical application
In, it is used as reference by the way that the LBP characteristic informations prestored are entered into (i.e. described probability distribution), to leaves of plants to be identified
Picture is identified, to realize the identification function of plant leaf blade.
The feature extracting method of plant leaf blade according to embodiments of the present invention, leaf samples image can be carried out at binaryzation
Reason, and the circle shaped neighborhood region of each pixel in the leaf samples image after binary conversion treatment is determined, and it is circular described in multiple rotary
Neighborhood calculates the target LBP values of each pixel in leaf samples image to calculate the target LBP values of each pixel
Probability distribution, and LBP characteristic informations using probability distribution as leaf samples image.Thus, it is each by multiple rotary
The circle shaped neighborhood region of pixel is to calculate the target LBP values of each pixel, and the target LBP values for calculating each pixel exist
Probability distribution in leaf samples image, and then the LBP characteristic informations using the probability distribution as leaf samples image, can be very
The error of recognition result when blade is distributed with different shape or different angle in image is overcome well, is improved feature in blade and is carried
The degree of accuracy taken, so, the characteristic information in leaf image is extracted by this feature extracting mode, and enter according to this feature information
During row leaf recognition, the accuracy of leaf recognition can be improved.
A kind of embodiment corresponding, of the invention with the feature extracting method of the plant leaf blade of above-mentioned several embodiments offers
A kind of feature deriving means of plant leaf blade are also provided, due to the feature deriving means of plant leaf blade provided in an embodiment of the present invention
It is corresponding with the feature extracting method for the plant leaf blade that above-mentioned several embodiments provide, therefore carried in the feature of afore-mentioned plants blade
The embodiment of method is taken to be also applied for the feature deriving means for the plant leaf blade that the present embodiment provides, in the present embodiment no longer
It is described in detail.Fig. 3 is the structural representation of the feature deriving means of plant leaf blade according to an embodiment of the invention.Such as Fig. 3
Shown, the feature deriving means 300 of the plant leaf blade can include:Acquisition module 310, binary conversion treatment module 320, first are counted
Calculate module 330, the second computing module 340 and the first determining module 350.
Specifically, acquisition module 310 is used to obtain leaf samples image.
Binary conversion treatment module 320 is used to carry out binary conversion treatment to leaf samples image.
First computing module 330 is adjacent for the circle of each pixel in the leaf samples image after determining binary conversion treatment
Domain, and multiple rotary circle shaped neighborhood region is to calculate the target LBP values of each pixel.As a kind of example, as shown in figure 4, should
First computing module 330 can include:First computing unit 331, the second computing unit 332, rotary unit 333 and acquiring unit
334。
Wherein, the first computing unit 331 can be used for be directed to each pixel, calculate each pixel gray value and each
The gray value of multiple pixels in pixel circle shaped neighborhood region.Second computing unit 332 can be used for the gray value according to each pixel
With the gray value of multiple pixels in each pixel circle shaped neighborhood region, the first LBP values of each pixel are calculated.Rotary unit 333
For carrying out multiple rotary to circle shaped neighborhood region.Wherein, in an embodiment of the present invention, first computing unit 331 can be additionally used in
After each rotation, working as multiple pixels in the current grayvalue and each pixel circle shaped neighborhood region of each pixel is calculated respectively
Preceding gray value.Second computing unit 332 can be additionally used in circular adjacent according to the current grayvalue of each pixel and each pixel
The current grayvalue of multiple pixels in domain, calculate multiple 2nd LBP values of each pixel.Acquiring unit 334 can be used for obtaining
Minimum value in first LBP values and multiple 2nd LBP values, and the target LBP values using minimum value as each pixel.
Probability point of the target LBP values that second computing module 340 is used to calculate each pixel in leaf samples image
Cloth.
First determining module 350 is used for the LBP characteristic informations using probability distribution as leaf samples image.
In order to realize the identification function of plant leaf blade, the recognition accuracy of blade is improved, further, the one of the present invention
In individual embodiment, as shown in figure 5, the feature deriving means 300 of the plant leaf blade may also include:Second determining module 360, establish
Module 370 and memory module 380.Wherein, the second determining module 360 can be used for the species for determining leaf samples.Establish module 370
Available for the corresponding relation established between the species of leaf samples and probability distribution.Memory module 380 can be used for leaf samples
Species, probability distribution and corresponding relation stored.
The feature deriving means of plant leaf blade according to embodiments of the present invention, each pixel of multiple rotary can be passed through
Circle shaped neighborhood region calculates the target LBP values of each pixel in leaf samples to calculate the target LBP values of each pixel
Probability distribution in image, and then the LBP characteristic informations using the probability distribution as leaf samples image, can overcome well
The error of recognition result when blade is distributed with different shape or different angle in image, improve the accurate of feature extraction in blade
Degree, so, the characteristic information in leaf image is extracted by this feature extracting mode, and blade knowledge is carried out according to this feature information
When other, the accuracy of leaf recognition can be improved.
The invention also provides a kind of recognition methods of plant leaf blade.Fig. 6 is plant according to an embodiment of the invention
The flow chart of the recognition methods of blade.It should be noted that the recognition methods of the plant leaf blade of the embodiment of the present invention can be applied to
The identification device of the plant leaf blade of the embodiment of the present invention, the identification device can be configured in computer equipment.Wherein, in the present invention
Embodiment in, the computer equipment can be server or terminal device, for example, the terminal device can be mobile phone, tablet personal computer,
Personal digital assistant etc. has the hardware device of various operating systems.
As shown in fig. 6, the recognition methods of the plant leaf blade can include:
S610, obtain leaf image to be identified.
Wherein, in an embodiment of the present invention, the blade to be identified the blade to be identified such as can be regarded as which belong to
A kind of blade of vegetation type.Alternatively, the leaf image to be identified can be that user shoots to obtain or from mutual
Obtained from being searched in networking.
S620, feature extraction is carried out to leaf image to be identified according to default feature extracting method, to obtain waiting knowing
The LBP characteristic informations of other leaf image.Wherein, in an embodiment of the present invention, the default feature extracting method can be
Feature extracting method described in any of the above-described embodiment of the present invention.
For example, can first pass through noise reduction algorithm carries out preliminary treatment to leaf image to be identified, it is to be identified to remove this
Redundancy noise in leaf image, and can use Global thresholding that the leaf image to be identified is carried out into binary conversion treatment.It
Afterwards, it may be determined that each pixel in leaf image to be identified after binary conversion treatment, and determine the circle of each pixel
Neighborhood, each pixel circle shaped neighborhood region is constantly rotated, and according to the pixel in postrotational each pixel circle shaped neighborhood region
Gray value calculates the target LBP values of each pixel.Wherein, the acquisition modes of the target LBP values of each pixel
Reference can be made to the description of above-mentioned embodiment illustrated in fig. 2, will not be repeated here.
After the target LBP values of each pixel in obtaining leaf image to be identified, each pixel can be calculated
Probability distribution of the target LBP values in leaf image to be identified, the probability distribution are the LBP of leaf image to be identified
Characteristic information.
S630, by the LBP features of the LBP characteristic informations of leaf image to be identified and the leaf samples image prestored
Information is matched.
Alternatively, the LBP characteristic informations of the leaf samples image prestored can be obtained, and calculate blade figure to be identified
Similarity between the LBP characteristic informations of the LBP characteristic informations of picture and the leaf samples image prestored so that pass through the phase
The matching feelings between the LBP characteristic informations of leaf image to be identified and the leaf samples image prestored are obtained like degree
Condition.
S640, if matching result meets preparatory condition, according to the LBP characteristic informations of leaf samples image and prestore
Corresponding relation, determine the species of blade in leaf image to be identified.
For example, it is assumed that realize the LBP characteristic informations of leaf image to be identified with depositing in advance by way of similarity
Matching between the leaf samples image of storage, then the LBP characteristic informations of the leaf image to be identified are being obtained with depositing in advance
After Similarity value between the leaf samples image of storage, it can determine whether the Similarity value is more than or equal to predetermined threshold value, if
It is that then can determine that matching result meets preparatory condition, can now obtains the corresponding relation prestored, and it is special according to LBP
Reference ceases the species that corresponding leaf samples are found out with the corresponding relation, and the species is in the leaf image to be identified
The species of blade.
In order to lift Consumer's Experience, alternatively, in one embodiment of the invention, the leaf image to be identified
Number can be multiple, the leaf image to be identified to each can carry out the identification of blade styles respectively.In the present embodiment, pin
To current leaf image to be identified, if the matching result does not meet preparatory condition, recognition failures or identification can determine that
Mistake, and after identification is completed to the multiple leaf image to be identified, identification correct number can be counted and identification is wrong
Quantity (wherein, the wrong situation for also including recognition failures of identification) by mistake, and according to identification correct number and blade figure to be identified
The total number of picture calculates recognition correct rate, and then can be the knowledge that can determine whether the embodiment of the present invention according to the recognition correct rate
The accuracy of other method.
The recognition methods of plant leaf blade according to embodiments of the present invention, the circle of each pixel of multiple rotary can be passed through
Neighborhood calculates the target LBP values of each pixel in leaf samples image to calculate the target LBP values of each pixel
In probability distribution, and then LBP characteristic informations using the probability distribution as leaf samples image can overcome image well
The error of recognition result, improves the degree of accuracy of feature extraction in blade when middle blade is distributed with different shape or different angle,
So, the characteristic information in leaf image is extracted by this feature extracting mode, and leaf recognition is carried out according to this feature information
When, the accuracy of leaf recognition can be improved.
With reference to Fig. 7, the present invention is described further.
For example, as shown in fig. 7, first, can first establish database, leaf samples image can be stored with the database
LBP characteristic informations, the species of the leaf samples and the species of the leaf samples it is corresponding with the LBP characteristic informations
Relation.I.e.:Leaf samples image (S710) can be first read, and binary conversion treatment (S720) is carried out to leaf samples image, afterwards,
Determine the circle shaped neighborhood region of each pixel in the leaf samples image after binary conversion treatment, and circle shaped neighborhood region described in multiple rotary with
The target LBP values (S730) of each pixel are calculated, and calculate the target LBP values of each pixel in leaf samples figure
Probability distribution (S740) as in.Then, the LBP characteristic informations using the probability distribution as leaf samples image.(S750).
Finally, it may be determined that the species of leaf samples, and the corresponding relation established between the species of leaf samples and probability distribution, and by sample
The species, probability distribution and corresponding relation of this blade are stored, to establish the database.
In the application that leaf image to be identified is identified reality, leaf image to be identified can be read
(S760) feature extraction, and according to default feature extracting method to leaf image to be identified is carried out, it is to be identified to obtain
The LBP characteristic informations (S770) of leaf image.Afterwards, the LBP of the leaf samples image prestored in readable data storehouse is special
Reference ceases, and the LBP characteristic informations of the leaf samples image is similar to the LBP characteristic informations progress of leaf image to be identified
Degree matching (S780).If Similarity value is more than or equal to predetermined threshold value, can determine whether to be stored with the leaf to be identified in database
The species of piece, and the blade styles (S790) are provided, if Similarity value is less than predetermined threshold value, it can determine that in database and be not present
The species (S7100) of the blade to be identified.
When leaf image to be identified is multiple, the species of the blade to be identified is stored with database is judged,
And when providing the blade styles, it can also calculate the identification correct number (S7110) for multiple leaf images to be identified and knowledge
Other number of errors, and according to being correct number and the total number of leaf image to be identified calculates recognition correct rate (S7120),
And export the recognition correct rate.
Corresponding with the recognition methods for the plant leaf blade that above-mentioned several embodiments provide, a kind of embodiment of the invention also carries
For a kind of identification device of plant leaf blade, due to identification device and the above-mentioned several realities of plant leaf blade provided in an embodiment of the present invention
It is corresponding to apply the recognition methods of the plant leaf blade of example offer, therefore the embodiment of the recognition methods in afore-mentioned plants blade is also fitted
The identification device of the plant leaf blade provided for the present embodiment, is not described in detail in the present embodiment.Fig. 8 is according to the present invention
The structural representation of the identification device of the plant leaf blade of one embodiment.As shown in figure 8, the identification device 800 of the plant leaf blade
It can include:Acquisition module 810, characteristic extracting module 820, matching module 830 and determining module 840.
Specifically, acquisition module 810 can be used for obtaining leaf image to be identified.
Characteristic extracting module 820 can be used for carrying out feature to leaf image to be identified according to default feature extracting method
Extraction, to obtain the LBP characteristic informations of leaf image to be identified.Wherein, in an embodiment of the present invention, the default spy
Extracting method is levied as the feature extracting method described in any of the above-described individual embodiment of the present invention.
Matching module 830 can be used for the LBP characteristic informations of leaf image to be identified and the leaf samples prestored
The LBP characteristic informations of image are matched.
Determining module 840 can be used for when matching result meets preparatory condition, be believed according to the LBP features of leaf samples image
The corresponding relation for ceasing and prestoring, determine the species of blade in leaf image to be identified.
The identification device of plant leaf blade according to embodiments of the present invention, the circle of each pixel of multiple rotary can be passed through
Neighborhood calculates the target LBP values of each pixel in leaf samples image to calculate the target LBP values of each pixel
In probability distribution, and then LBP characteristic informations using the probability distribution as leaf samples image can overcome image well
The error of recognition result, improves the degree of accuracy of feature extraction in blade when middle blade is distributed with different shape or different angle,
So, the characteristic information in leaf image is extracted by this feature extracting mode, and leaf recognition is carried out according to this feature information
When, the accuracy of leaf recognition can be improved.
In order to realize above-described embodiment, the invention also provides a kind of computer equipment.
Fig. 9 is the structural representation of computer equipment according to an embodiment of the invention.It should be noted that in this hair
In bright embodiment, the computer equipment can be server or terminal device.As shown in figure 9, the computer equipment 900 can wrap
Include:Memory 910, processor 920 and it is stored in the computer program that can be run on memory 910 and on processor 920
930, when processor 920 performs described program 930, realize the feature of the plant leaf blade described in any of the above-described individual embodiment of the present invention
Extracting method.
In order to realize above-described embodiment, the invention also provides a kind of non-transitorycomputer readable storage medium, thereon
Computer program is stored with, the leaves of plants described in any of the above-described individual embodiment of the present invention is realized when described program is executed by processor
The feature extracting method of piece.
In order to realize above-described embodiment, the invention also provides another computer equipment.
Figure 10 is the structural representation of computer equipment in accordance with another embodiment of the present invention.It should be noted that
In embodiments of the invention, the computer equipment can be server or terminal device.As shown in Figure 10, the computer equipment 1000
It can include:Memory 1010, processor 1020 and it is stored in the meter that can be run on memory 1010 and on processor 1020
Calculation machine program 1030, when processor 1020 performs described program 1030, realize the plant described in any of the above-described individual embodiment of the present invention
The recognition methods of thing blade.
In order to realize above-described embodiment, the invention also provides another non-transitorycomputer readable storage medium, its
On be stored with computer program, the plant described in any of the above-described individual embodiment of the present invention is realized when described program is executed by processor
The recognition methods of blade.
In the description of the invention, it is to be understood that term " first ", " second " are only used for describing purpose, and can not
It is interpreted as indicating or implies relative importance or imply the quantity of the technical characteristic indicated by indicating.Thus, define " the
One ", at least one this feature can be expressed or be implicitly included to the feature of " second ".In the description of the invention, " multiple "
It is meant that at least two, such as two, three etc., unless otherwise specifically defined.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or the spy for combining the embodiment or example description
Point is contained at least one embodiment or example of the present invention.In this manual, to the schematic representation of above-mentioned term not
Identical embodiment or example must be directed to.Moreover, specific features, structure, material or the feature of description can be with office
Combined in an appropriate manner in one or more embodiments or example.In addition, in the case of not conflicting, the skill of this area
Art personnel can be tied the different embodiments or example and the feature of different embodiments or example described in this specification
Close and combine.
Any process or method described otherwise above description in flow chart or herein is construed as, and represents to include
Module, fragment or the portion of the code of the executable instruction of one or more the step of being used to realize specific logical function or process
Point, and the scope of the preferred embodiment of the present invention includes other realization, wherein can not press shown or discuss suitable
Sequence, including according to involved function by it is basic simultaneously in the way of or in the opposite order, carry out perform function, this should be of the invention
Embodiment person of ordinary skill in the field understood.
Expression or logic and/or step described otherwise above herein in flow charts, for example, being considered use
In the order list for the executable instruction for realizing logic function, may be embodied in any computer-readable medium, for
Instruction execution system, device or equipment (such as computer based system including the system of processor or other can be held from instruction
The system of row system, device or equipment instruction fetch and execute instruction) use, or combine these instruction execution systems, device or set
It is standby and use.For the purpose of this specification, " computer-readable medium " can any can be included, store, communicate, propagate or pass
Defeated program is for instruction execution system, device or equipment or the dress used with reference to these instruction execution systems, device or equipment
Put.The more specifically example (non-exhaustive list) of computer-readable medium includes following:Electricity with one or more wiring
Connecting portion (electronic installation), portable computer diskette box (magnetic device), random access memory (RAM), read-only storage
(ROM), erasable edit read-only storage (EPROM or flash memory), fiber device, and portable optic disk is read-only deposits
Reservoir (CDROM).In addition, computer-readable medium, which can even is that, to print the paper of described program thereon or other are suitable
Medium, because can then enter edlin, interpretation or if necessary with it for example by carrying out optical scanner to paper or other media
His suitable method is handled electronically to obtain described program, is then stored in computer storage.
It should be appreciated that each several part of the present invention can be realized with hardware, software, firmware or combinations thereof.Above-mentioned
In embodiment, software that multiple steps or method can be performed in memory and by suitable instruction execution system with storage
Or firmware is realized.If, and in another embodiment, can be with well known in the art for example, realized with hardware
Any one of row technology or their combination are realized:With the logic gates for realizing logic function to data-signal
Discrete logic, have suitable combinational logic gate circuit application specific integrated circuit, programmable gate array (PGA), scene
Programmable gate array (FPGA) etc..
Those skilled in the art are appreciated that to realize all or part of step that above-described embodiment method carries
Suddenly it is that by program the hardware of correlation can be instructed to complete, described program can be stored in a kind of computer-readable storage medium
In matter, the program upon execution, including one or a combination set of the step of embodiment of the method.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing module, can also
That unit is individually physically present, can also two or more units be integrated in a module.Above-mentioned integrated mould
Block can both be realized in the form of hardware, can also be realized in the form of software function module.The integrated module is such as
Fruit is realized in the form of software function module and as independent production marketing or in use, can also be stored in a computer
In read/write memory medium.
Storage medium mentioned above can be read-only storage, disk or CD etc..Although have been shown and retouch above
Embodiments of the invention are stated, it is to be understood that above-described embodiment is exemplary, it is impossible to be interpreted as the limit to the present invention
System, one of ordinary skill in the art can be changed to above-described embodiment, change, replace and become within the scope of the invention
Type.
Claims (12)
1. a kind of feature extracting method of plant leaf blade, it is characterised in that comprise the following steps:
Leaf samples image is obtained, and binary conversion treatment is carried out to the leaf samples image;
The circle shaped neighborhood region of each pixel in the leaf samples image after binary conversion treatment is determined, and it is circular adjacent described in multiple rotary
Domain is to calculate the target local binary patterns LBP values of each pixel;
Calculate probability distribution of the target LBP values of each pixel in the leaf samples image;
LBP characteristic informations using the probability distribution as the leaf samples image.
2. the method as described in claim 1, it is characterised in that circle shaped neighborhood region described in the multiple rotary is described each to calculate
The target local binary patterns LBP values of individual pixel, including:
For each pixel, multiple pictures in the gray value and each pixel circle shaped neighborhood region of each pixel are calculated
The gray value of element;
According to the gray value of multiple pixels in the gray value of each pixel and each pixel circle shaped neighborhood region, calculate
First LBP values of each pixel;
Multiple rotary is carried out to the circle shaped neighborhood region, and after each rotation, calculates the current ash of each pixel respectively
The current grayvalue of multiple pixels in angle value and each pixel circle shaped neighborhood region;
According to the current ash of multiple pixels in the current grayvalue of each pixel and each pixel circle shaped neighborhood region
Angle value, calculate multiple 2nd LBP values of each pixel;
The minimum value in the first LBP values and the multiple 2nd LBP values is obtained, and using the minimum value as described each
The target LBP values of pixel.
3. method as claimed in claim 1 or 2, it is characterised in that methods described also includes:
Determine the species of the leaf samples;
The corresponding relation established between the species of the leaf samples and the probability distribution;
The species of the leaf samples, the probability distribution and the corresponding relation are stored.
A kind of 4. recognition methods of plant leaf blade, it is characterised in that including:
Obtain leaf image to be identified;
Feature extraction is carried out to the leaf image to be identified according to default feature extracting method, it is described to be identified to obtain
Leaf image LBP characteristic informations, wherein, the default feature extracting method is such as any one of claims 1 to 3 institute
The feature extracting method stated;
By the LBP features of the LBP characteristic informations of the leaf image to be identified and the leaf samples image prestored
Information is matched;
If matching result meets preparatory condition, according to the LBP characteristic informations of the leaf samples image and the institute prestored
Corresponding relation is stated, determines the species of blade in the leaf image to be identified.
A kind of 5. feature deriving means of plant leaf blade, it is characterised in that including:
Acquisition module, for obtaining leaf samples image;
Binary conversion treatment module, for carrying out binary conversion treatment to the leaf samples image;
First computing module, for determining the circle shaped neighborhood region of each pixel in the leaf samples image after binary conversion treatment, and
Circle shaped neighborhood region described in multiple rotary is to calculate the target LBP values of each pixel;
Second computing module, for calculating probability of the target LBP values of each pixel in the leaf samples image
Distribution;
First determining module, for the LBP characteristic informations using the probability distribution as the leaf samples image.
6. device as claimed in claim 5, it is characterised in that first computing module includes:
First computing unit, for for each pixel, calculating the gray value of each pixel and each pixel
The gray value of multiple pixels in point circle shaped neighborhood region;
Second computing unit, for multiple in the gray value according to each pixel and each pixel circle shaped neighborhood region
The gray value of pixel, calculate the first LBP values of each pixel;
Rotary unit, for carrying out multiple rotary to the circle shaped neighborhood region;
Wherein, first computing unit, it is additionally operable to after each rotation, calculates the current gray level of each pixel respectively
The current grayvalue of multiple pixels in value and each pixel circle shaped neighborhood region;
Second computing unit, it is additionally operable to circular according to the current grayvalue of each pixel and each pixel
The current grayvalue of multiple pixels in neighborhood, calculate multiple 2nd LBP values of each pixel;
Acquiring unit, for obtaining the minimum value in the first LBP values and the multiple 2nd LBP values, and by the minimum
It is worth the target LBP values as each pixel.
7. the device as described in claim 5 or 6, it is characterised in that described device also includes:
Second determining module, for determining the species of the leaf samples;
Module is established, for establishing the corresponding relation between the species of the leaf samples and the probability distribution;
Memory module, for the species of the leaf samples, the probability distribution and the corresponding relation to be stored.
A kind of 8. identification device of plant leaf blade, it is characterised in that including:
Acquisition module, for obtaining leaf image to be identified;
Characteristic extracting module, carried for carrying out feature to the leaf image to be identified according to default feature extracting method
Take, to obtain the LBP characteristic informations of the leaf image to be identified, wherein, the default feature extracting method is as weighed
Profit requires the feature extracting method any one of 1 to 3;
Matching module, for by the LBP characteristic informations of the leaf image to be identified and the leaf samples that prestore
The LBP characteristic informations of image are matched;
Determining module, for when matching result meets preparatory condition, according to the LBP characteristic informations of the leaf samples image and
The corresponding relation prestored, determine the species of blade in the leaf image to be identified.
A kind of 9. computer equipment, it is characterised in that including:Memory, processor and it is stored on the memory and can be
The computer program run on the processor, during the computing device described program, realize as appointed in claims 1 to 3
The feature extracting method of plant leaf blade described in one.
10. a kind of non-transitorycomputer readable storage medium, is stored thereon with computer program, it is characterised in that the journey
The feature extracting method of plant leaf blade as claimed any one in claims 1 to 3 is realized when sequence is executed by processor.
A kind of 11. computer equipment, it is characterised in that including:Memory, processor and it is stored on the memory and can be
The computer program run on the processor, during the computing device described program, realize as claimed in claim 4 plant
The recognition methods of thing blade.
12. a kind of non-transitorycomputer readable storage medium, is stored thereon with computer program, it is characterised in that the journey
The recognition methods of plant leaf blade as claimed in claim 4 is realized when sequence is executed by processor.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711025461.4A CN107886114A (en) | 2017-10-27 | 2017-10-27 | Feature extracting method, recognition methods, device and the computer equipment of plant leaf blade |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711025461.4A CN107886114A (en) | 2017-10-27 | 2017-10-27 | Feature extracting method, recognition methods, device and the computer equipment of plant leaf blade |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107886114A true CN107886114A (en) | 2018-04-06 |
Family
ID=61782572
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711025461.4A Withdrawn CN107886114A (en) | 2017-10-27 | 2017-10-27 | Feature extracting method, recognition methods, device and the computer equipment of plant leaf blade |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107886114A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109117857A (en) * | 2018-08-28 | 2019-01-01 | 苏州芯德锐信息科技有限公司 | A kind of recognition methods of biological attribute, device and equipment |
CN109558883A (en) * | 2018-12-03 | 2019-04-02 | 宁夏智启连山科技有限公司 | Leaf characteristic extracting method and device |
CN109740614A (en) * | 2018-11-20 | 2019-05-10 | 广东智媒云图科技股份有限公司 | A kind of method and device obtaining the blade Background for leaf carving |
CN111753688A (en) * | 2020-06-12 | 2020-10-09 | 广州极飞科技有限公司 | Planting line center line determining method and device, electronic equipment and storage medium |
CN114299299A (en) * | 2021-11-29 | 2022-04-08 | 苏州浪潮智能科技有限公司 | Tree leaf feature extraction method and device, computer equipment and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102542243A (en) * | 2010-12-17 | 2012-07-04 | 北京理工大学 | LBP (Local Binary Pattern) image and block encoding-based iris feature extracting method |
CN106778782A (en) * | 2016-12-15 | 2017-05-31 | 盐城工学院 | Identifying pest and prevention and controls and system |
CN106909873A (en) * | 2016-06-21 | 2017-06-30 | 湖南拓视觉信息技术有限公司 | The method and apparatus of recognition of face |
-
2017
- 2017-10-27 CN CN201711025461.4A patent/CN107886114A/en not_active Withdrawn
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102542243A (en) * | 2010-12-17 | 2012-07-04 | 北京理工大学 | LBP (Local Binary Pattern) image and block encoding-based iris feature extracting method |
CN106909873A (en) * | 2016-06-21 | 2017-06-30 | 湖南拓视觉信息技术有限公司 | The method and apparatus of recognition of face |
CN106778782A (en) * | 2016-12-15 | 2017-05-31 | 盐城工学院 | Identifying pest and prevention and controls and system |
Non-Patent Citations (2)
Title |
---|
宋克臣等: ""局部二值模式方法研究与展望"", 《自动化学报》 * |
徐少平等: ""基于区域最大相似度的快速图像分割方法"", 《光电子.激光》 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109117857A (en) * | 2018-08-28 | 2019-01-01 | 苏州芯德锐信息科技有限公司 | A kind of recognition methods of biological attribute, device and equipment |
CN109117857B (en) * | 2018-08-28 | 2022-08-16 | 苏州芯德锐信息科技有限公司 | Biological attribute identification method, device and equipment |
CN109740614A (en) * | 2018-11-20 | 2019-05-10 | 广东智媒云图科技股份有限公司 | A kind of method and device obtaining the blade Background for leaf carving |
CN109740614B (en) * | 2018-11-20 | 2024-01-23 | 广东智媒云图科技股份有限公司 | Method and device for obtaining blade background image for leaf carving |
CN109558883A (en) * | 2018-12-03 | 2019-04-02 | 宁夏智启连山科技有限公司 | Leaf characteristic extracting method and device |
CN109558883B (en) * | 2018-12-03 | 2023-04-18 | 宁夏智启连山科技有限公司 | Blade feature extraction method and device |
CN111753688A (en) * | 2020-06-12 | 2020-10-09 | 广州极飞科技有限公司 | Planting line center line determining method and device, electronic equipment and storage medium |
CN114299299A (en) * | 2021-11-29 | 2022-04-08 | 苏州浪潮智能科技有限公司 | Tree leaf feature extraction method and device, computer equipment and storage medium |
CN114299299B (en) * | 2021-11-29 | 2024-01-23 | 苏州浪潮智能科技有限公司 | Tree leaf feature extraction method and device, computer equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107886114A (en) | Feature extracting method, recognition methods, device and the computer equipment of plant leaf blade | |
Yuan et al. | Reversion correction and regularized random walk ranking for saliency detection | |
Matthé et al. | Comparison of photo‐matching algorithms commonly used for photographic capture–recapture studies | |
Cintas et al. | Automatic ear detection and feature extraction using geometric morphometrics and convolutional neural networks | |
Yan et al. | Enabling a single deep learning model for accurate gland instance segmentation: A shape-aware adversarial learning framework | |
Sampathkumar et al. | An automated crop and plant disease identification scheme using cognitive fuzzy c-means algorithm | |
CN108337551A (en) | A kind of screen recording method, storage medium and terminal device | |
CN112633297B (en) | Target object identification method and device, storage medium and electronic device | |
CN110009656B (en) | Target object determination method and device, storage medium and electronic device | |
CN111784675A (en) | Method and device for processing article texture information, storage medium and electronic equipment | |
CN109766818A (en) | Pupil center's localization method and system, computer equipment and readable storage medium storing program for executing | |
Zhang et al. | Detection of regions of interest in a high-spatial-resolution remote sensing image based on an adaptive spatial subsampling visual attention model | |
CN108549836A (en) | Reproduction detection method, device, equipment and the readable storage medium storing program for executing of photo | |
CN111882559B (en) | ECG signal acquisition method and device, storage medium and electronic device | |
Mairon et al. | A closer look at context: From coxels to the contextual emergence of object saliency | |
CN115690803A (en) | Digital image recognition method and device, electronic equipment and readable storage medium | |
CN108182444A (en) | The method and device of video quality diagnosis based on scene classification | |
CN108764248B (en) | Image feature point extraction method and device | |
Zambrano-Vizuete et al. | Segmentation of medical image using novel dilated ghost deep learning model | |
CN113378609B (en) | Agent proxy signature identification method and device | |
CN112861743A (en) | Palm vein image anti-counterfeiting method, device and equipment | |
Suárez et al. | Vegetation index estimation from monospectral images | |
Zhang et al. | Multiple morphological constraints-based complex gland segmentation in colorectal cancer pathology image analysis | |
Lin et al. | Alpha matting using robust color sampling and fully connected conditional random fields | |
CN111753723B (en) | Fingerprint identification method and device based on density calibration |
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 | ||
TA01 | Transfer of patent application right | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20190903 Address after: 100192 Dongsheng Science Park, Zhongguancun, 66 Xixiaokou Road, Haidian District, Beijing Applicant after: Green Bay Network Technology Co., Ltd. Address before: 100089 Beijing Haidian District Xixiaokou Road 66 Zhongguancun Dongsheng Science Park B-6 Building B 5 floors Applicant before: Grass count language (Beijing) Technology Co., Ltd. |
|
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20180406 |