CN108776802A - A kind of peanut varieties recognition methods and system - Google Patents
A kind of peanut varieties recognition methods and system Download PDFInfo
- Publication number
- CN108776802A CN108776802A CN201810349090.3A CN201810349090A CN108776802A CN 108776802 A CN108776802 A CN 108776802A CN 201810349090 A CN201810349090 A CN 201810349090A CN 108776802 A CN108776802 A CN 108776802A
- Authority
- CN
- China
- Prior art keywords
- peanut
- image
- described image
- character representation
- feature
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the present invention provides a kind of peanut varieties recognition methods and system, wherein the method provided includes：The image for including peanut is obtained, the feature of peanut part in described image is extracted, the character representation of peanut part in described image is obtained according to the feature of the peanut part；According to the character representation of peanut part in described image, the kind of the peanut in described image is identified.Method provided in an embodiment of the present invention, feature extraction is carried out to the peanut part in image, obtain the character representation of image peanut part, the character representation of peanut part is identified simultaneously, to carry out exact classification to mixing peanut image, for the numerous present situations of peanut varieties, compared to manual identified, classification accuracy is high.
Description
Technical field
The present embodiments relate to field of computer technology more particularly to a kind of peanut varieties recognition methods and systems.
Background technology
With the development of Internet technology and various information technologies, Agricultural Development Model start from original traditional agriculture to
The wisdom agricultural of modernization is changed, and in agricultural production, is frequently encountered a plant different kind comprising there are many,
Such as the type that peanut has been registered at present in agricultural product has reached 500 kinds or more, different peanut varieties have different spies
Property, such as Virginia type peanut type has 68-4,17, Guangdong oil 551, white sand 1016, Hubei Province spent to spend No. 3 etc., the flower with resistance to bacterial wilt
Life includes that Guangdong oil 92, Hubei Province spend No. 5, Shandong spends No. 3, the 28, self-sufficient and strategically located region 3, Guangdong oil 116, Hubei Province is spent to spend the kinds such as No. 4.Different kinds is suitable
With different growing environments, therefore, the assortment identification of peanut has more wide application value in practice, in wisdom
Agriculturally also there is consequence.
In actual life, the plantation of different peanut varieties is easy to confusion, and it is also chaotic to lead to the peanut harvested
, in the prior art, the identification method to peanut varieties is usually that artificial naked eyes identify and classify that this recognition methods depends on
Identification personnel's is professional, and peanut it is various in style due to, the accuracy rate of identification remains unchanged very low.
Invention content
A kind of peanut varieties recognition methods of offer of the embodiment of the present invention and system, to solve the prior art to peanut classification
Identification depend on manual identified, need identification personnel to have certain professional, simultaneously because the original various in style of peanut
Cause, the low problem of the accuracy rate of identification.
The embodiment of the present invention provides a kind of peanut varieties recognition methods, including：It obtains and includes the image of peanut, described in extraction
The feature of peanut part in image obtains the character representation of peanut part in described image according to the feature of the peanut part；
According to the character representation of peanut part in described image, the kind of the peanut in described image is identified.
Wherein, the feature for extracting peanut part in described image, obtains the mark sheet of peanut part in described image
The step of showing specifically includes：
Described image is divided into the identical square window of multiple sizes, extracts the SIFT descriptions of the square window
The constant SIFT descriptors of wherein translation and illumination variation are made the SIFT descriptors for representing peanut object by symbol, will be described
The SIFT descriptors for representing peanut object are put into Density Clustering, the model for establishing cluster feature；
By density clustering algorithm, the SIFT descriptors for representing peanut object are clustered, multiple clusters are built
Center, and code book is built according to the cluster centre；
According to Bag-of-words model models, the BoW expressions for corresponding to the code book in described image are obtained, by institute
State the character representation that BoW is denoted as peanut part in described image.
Wherein, the character representation according to peanut part in described image, to the kind of the peanut in described image into
The step of row identification, specifically includes：
The character representation of peanut part in described image is input in preset SVM classifier, SVM classifier pair is passed through
The character representation of peanut part carries out Classification and Identification in described image.
Wherein, the number of the SVM classifier is k (k-1)/2；Wherein, k is the number for the peanut type for needing to identify
Amount.
Wherein, described specific to the character representation progress Classification and Identification of peanut part in described image by SVM classifier
Including：The character representation of peanut part in described image is tested according to k (k-1)/2 SVM classifier, takes ballot
Form, the highest recognition result that will win the vote is as final recognition result.
Wherein, the method further includes the step trained to the SVM classifier, wherein is trained to the SVM classifier
The step of include：
It includes peanut classification mark to obtain multiple, including the image of peanut is as training sample set, according to the training
Sample set is trained SVM classifier.
Wherein, the character representation according to peanut part in described image, to the kind of the peanut in described image into
The step of row identification, specifically includes：
The character representation of peanut part in described image is input in preset neural network, by neural network to institute
The character representation for stating peanut part in image carries out Classification and Identification.
According to the second aspect of the invention, a kind of peanut varieties identifying system is provided, including：Preprocessing module, for obtaining
It includes peanut image to take, and extracts the feature of peanut part in described image, obtains the character representation of peanut part in described image；
Identification module, for the character representation according to peanut part in described image, to the product of the peanut in described image
Kind is identified.
According to the third aspect of the invention we, a kind of computer readable storage medium is provided, computer program is stored thereon with,
Such as above-mentioned peanut varieties recognition methods is realized when the program is executed by processor.
The embodiment of the present invention also provides a kind of peanut varieties identification equipment, including：
At least one processor；And at least one processor being connected to the processor, wherein：The memory is deposited
The program instruction that can be executed by the processor is contained, the processor calls described program instruction to be able to carry out following action：
The image for including peanut is obtained, the feature of peanut part in described image is extracted, institute is obtained according to the feature of the peanut part
State the character representation of peanut part in image；According to the character representation of peanut part in described image, to the flower in described image
Raw kind is identified.
Peanut varieties recognition methods provided in an embodiment of the present invention carries out feature extraction to the peanut part in image, obtains
The character representation of image peanut part is obtained, while the character representation of peanut part is identified, to mixing peanut image
Exact classification is carried out, for the numerous present situations of peanut varieties, compared to manual identified, classification accuracy is high.
Description of the drawings
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is this hair
Some bright embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is a kind of flow chart for peanut varieties recognition methods that one embodiment of the invention provides；
Fig. 2 is a kind of structure chart for peanut varieties identifying system that another embodiment of the present invention provides；
Fig. 3 is a kind of structure chart for peanut varieties identification equipment that yet another embodiment of the invention provides.
Specific implementation mode
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
The every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
With reference to figure 1, Fig. 1 is a kind of flow chart for peanut varieties recognition methods that one embodiment of the invention provides, this implementation
Example provide method include：
S1, it includes peanut image to obtain, and extracts the feature of peanut part in described image, obtains peanut portion in described image
The character representation divided；
S2, according to the character representation of peanut part in described image, the kind of the peanut in described image is identified.
Specifically, actual comprising in the image acquisition process of peanut, due to the acquisition environment of image, harvester etc.
Reason so that the image effect of acquisition is not good enough, to which the effect of feature extraction can be caused poor, cannot be satisfied accuracy of identification
Demand, therefore when receiving the image comprising peanut, need to pre-process image, extract peanut part in image
Feature, wherein an image may have comprising multiple peanuts, therefore the peanut part of each peanut in image be carried out special
Sign extraction, obtains the character representation of all peanuts in image.In obtaining image after the character representation of peanut part, by right
These character representations are identified, so as to obtain the kind of peanut in original image.
Wherein, the peanut Partial Feature of image indicates, can be the range of the image in image occupied by single peanut,
Can be the contour images of peanut in image, after eliminating edge effect and noise section, by the figure of remaining peanut part
As the character representation as peanut in image, and then the identification to character representation can be carried out.
By the method, feature extraction is carried out to the peanut part in image, obtains the character representation of image peanut part,
The character representation of peanut part is identified simultaneously, to carry out exact classification to mixing peanut image, for peanut varieties
Numerous present situations, compared to manual identified, classification accuracy is high.
On the basis of the above embodiments, the feature for extracting peanut part in described image, obtains in described image
The step of character representation of peanut part, specifically includes：
Described image is divided into the identical square window of multiple sizes, extracts the SIFT descriptions of the square window
The constant SIFT descriptors of wherein translation and illumination variation are made the SIFT descriptors for representing peanut object by symbol, will be described
The SIFT descriptors for representing peanut object are put into Density Clustering, the model for establishing cluster feature；It is calculated by Density Clustering
Method clusters the SIFT descriptors for representing peanut object, builds multiple cluster centres, and according to the cluster centre
Build code book；According to Bag-of-words model models, the BoW expressions for corresponding to the code book in described image are obtained, it will
The BoW is denoted as the character representation of peanut part in described image.
Specifically, the pre-treatment step to image specifically includes：Extract discrete Scale invariant features transform (Discrete
Scale-invariant feature transform, DSIFT) feature, the feature that DSIFT algorithms obtain is gathered by density
Class algorithm carries out dimensionality reduction operation, and builds code book by cluster centre, then by Bag-of-words model models according to structure
The code book built obtains the BoW expressions that image corresponds to code book, to complete the pretreatment to image.
Wherein, DSIFT be according to Scale invariant features transform (Scale-invariant feature transform,
SIFT the pretreatment to image realized on the basis of), SIFT are a kind of descriptions for image processing field.This description tool
There is scale invariability, can detect key point in the picture, is a kind of local feature description's.Feature detection includes mainly following
4 basic steps, 1, scale space extremum extracting：Search for the picture position on all scales.Known by gaussian derivative function
Not potentially for the point of interest of scale and invariable rotary.2, crucial point location：It is quasi- by one on the position of each candidate
Fine model is closed to determine position and scale.The selection gist of key point is in their degree of stability.3, direction determines, is based on
The gradient direction of image local distributes to each key point position one or more direction.It is all subsequent to image data
Operation is converted both relative to the direction of key point, scale and position, to provide the invariance for these transformation.4, it closes
Key point describes：In the neighborhood around each key point, the gradient of image local is measured on selected scale.These gradient quilts
It is transformed into a kind of expression, this deformation and the illumination variation for indicating to allow bigger local shape.
In the present embodiment, DSIFT algorithms are selected, a sliding window are introduced in spatial extrema detection-phase, in window
Non-maxima suppression is carried out to the detection of extreme point so that the distribution of characteristic point is relatively uniform, and arithmetic speed faster, and is kept
The invariance such as scale, rotation, affine.It is added to down-sampled operation before feature extraction, position is added before calculating homography matrix
The step of confidence breath reduction, the introducing K-D trees searches match point during, and the screening in characteristic point and homography matrix
RANSAC algorithms are used in estimation, all reduce the time overhead in image registration each stage.It is handled by grid representation and includes
The image of peanut, after image is divided into the identical window of size, with direction arrow come indicate the image in grid and
Rotation, translation and the constant descriptor of illumination variation are selected to represent peanut object, in specified extracted region SIFT descriptors,
Once obtaining descriptor, just place them into Density Clustering, to establish the model of cluster feature.
The characteristic dimension that DSIFT algorithms obtain is larger, so the characteristic dimension of DSIFT algorithms should be reduced to improve speed
And accuracy.Density clustering algorithm is a kind of unsupervised learning algorithm, because the label of features described above grouping is unknown, institute
With in cluster process, target is by Iteration Classification come processing feature value.Density clustering algorithm usually will be closely coupled
Sample divide one kind into, to obtain one cluster classification, by dividing the closely coupled sample of all each groups into each difference
Classification, then can obtain all cluster category results, it is then logical according to the code book of image of the cluster centre structure comprising peanut
It crosses BoW models and obtains the BoW expressions that image corresponds to code book.
Bag-of-words model (BoW model) appear in the fields NLP and IR earliest, which neglects text
Grammer and word order express passage or a document with one group of unordered word (words).In recent years, BoW models are wide
It is general to be applied in computer vision, it is taken as word with the feature (feature) of the BoW analogies applied to text, image
(Word).To every pictures, each word of the picture is calculated by Density Clustering should belong to " which in codebook
Class " word indicates to obtain the picture corresponding to the BoW of the code book.
By the method, characteristics of image is extracted by DSIFT, then dropped to the DEIFT features extracted with Density Clustering
Dimension, using feature construction BoW models, knows the peanut type in image using grader or neural network to realize
Not, by the way that pretreated characteristics of image is identified, accuracy of identification can effectively be promoted.
On the basis of the above embodiments, the character representation according to peanut part in described image, to described image
In the kind of peanut the step of being identified specifically include：
The character representation of peanut part in described image is input in preset SVM classifier, SVM classifier pair is passed through
The character representation of peanut part carries out Classification and Identification in described image.
Wherein, the number of the SVM classifier is k (k-1)/2；Wherein, k is the number for the peanut type for needing to identify
Amount.
Wherein, described specific to the character representation progress Classification and Identification of peanut part in described image by SVM classifier
Including：The character representation of peanut part in described image is tested according to k (k-1)/2 SVM classifier, takes ballot
Form, the highest recognition result that will win the vote is as final recognition result.
Specifically, be input in the SVM classifier preset by the character representation of peanut part in the image by acquisition,
To which SVM classifier can be identified the type of peanut according to feature.
Further, bad for multi-class problem effect since SVM generally can be used only in two class problems, therefore specific
Implement in converging, k classification is identified thereby using more SVM classifiers in the quantity of the peanut type identified as needed
Peanut just need to design k (k-1)/2 SVM and carry out category identification, such as need now to three different types of peanuts
A, B and C are identified, then 3 SVM of structure is needed to correspond to type A and B, type A and C and type B and C respectively.It is being identified
During, the character representation of peanut part is input in above three SVM classifier, category identification is carried out, then takes
The form of ballot finally obtains one group of result as final recognition result.
On the basis of the above embodiments, the method further includes the step trained to the SVM classifier, wherein right
The step of SVM classifier training includes：It includes peanut classification mark to obtain multiple, including the image of peanut is as training
Sample set is trained according to the training sample set pair SVM classifier.
Specifically, by acquiring multiple images for including peanut, the peanut type in image is labeled, structure training
Sample set is trained by training sample set pair SVM classifier.
On the basis of the above embodiments, the character representation according to peanut part in described image, to described image
In the kind of peanut the step of being identified specifically include：The character representation of peanut part in described image is input to default
Neural network in, by neural network in described image peanut part character representation carry out Classification and Identification.
Specifically, in obtaining image after the character representation of peanut part, can also use neural network to feature into
Row identification, the character representation of acquisition is input in preset neural network, to which the type of peanut be identified.
Equally, before using neural network classification, it is also desirable to which the training sample set for building response comes to preset nerve
Network is trained, to complete the category identification to peanut in the image comprising peanut.
With reference to figure 2, Fig. 2 is a kind of structure chart for peanut varieties identification equipment that further embodiment of this invention provides, described
System includes：Preprocessing module 21 and identification module 22.
Wherein, preprocessing module 21 includes peanut image for obtaining, and extracts the feature of peanut part in described image, obtains
Take the character representation of peanut part in described image.
Identification module 22 is used for the character representation according to peanut part in described image, to the product of the peanut in described image
Kind is identified.
Specifically, actual comprising in the image acquisition process of peanut, due to the acquisition environment of image, harvester etc.
Reason so that the image effect of acquisition is not good enough, to which the effect of feature extraction can be caused poor, cannot be satisfied accuracy of identification
Demand, therefore when receiving the image comprising peanut, need to pre-process image, extract peanut part in image
Feature, wherein an image may have comprising multiple peanuts, therefore the peanut part of each peanut in image be carried out special
Sign extraction, obtains the character representation of all peanuts in image.In obtaining image after the character representation of peanut part, by right
These character representations are identified, so as to obtain the kind of peanut in original image.
Wherein, the peanut Partial Feature of image indicates, can be the range of the image in image occupied by single peanut,
Can be the contour images of peanut in image, after eliminating edge effect and noise section, by the figure of remaining peanut part
As the character representation as peanut in image, and then the identification to character representation can be carried out.
By this system, feature extraction is carried out to the peanut part in image, obtains the character representation of image peanut part,
The character representation of peanut part is identified simultaneously, to carry out exact classification to mixing peanut image, for peanut varieties
Numerous present situations, compared to manual identified, classification accuracy is high.
On the basis of the above embodiments, the preprocessing module 21 is specifically used for, and described image is divided into multiple big
Small identical square window extracts the SIFT descriptors of the square window, and wherein translation and illumination variation is constant
SIFT descriptors are made to represent the SIFT descriptors of peanut object, the SIFT descriptors for representing peanut object are put into close
In degree cluster, the model for establishing cluster feature；
By density clustering algorithm, the SIFT descriptors for representing peanut object are clustered, multiple clusters are built
Center, and code book is built according to the cluster centre；
According to Bag-of-words model models, the BoW expressions for corresponding to the code book in described image are obtained, by institute
State the character representation that BoW is denoted as peanut part in described image.
Specifically, the pre-treatment step to image specifically includes：Extract discrete Scale invariant features transform (Discrete
Scale-invariant feature transform, DSIFT) feature, the feature that DSIFT algorithms obtain is gathered by density
Class algorithm carries out dimensionality reduction operation, and builds code book by cluster centre, then by Bag-of-words model models according to structure
The code book built obtains the BoW expressions that image corresponds to code book, to complete the pretreatment to image.
The image for including peanut is handled by grid representation, after image is divided into the identical window of size, with side
Indicate that the image in grid and rotation, translation and the constant descriptor of illumination variation are selected to represent peanut pair to arrow
As in specified extracted region SIFT descriptors, once obtaining descriptor, just placing them into Density Clustering, to establish cluster
The model of feature then obtains image by BoW models and corresponds to according to the code book of image of the cluster centre structure comprising peanut
The BoW of code book is indicated.
By this system, characteristics of image is extracted by DSIFT, then dropped to the DEIFT features extracted with Density Clustering
Dimension, using feature construction BoW models, knows the peanut type in image using grader or neural network to realize
Not, by the way that pretreated characteristics of image is identified, accuracy of identification can effectively be promoted.
Fig. 3 illustrates a kind of structural schematic diagram of peanut varieties identification equipment, as shown in figure 3, the server may include：
Processor (processor) 310, communication interface (Communications Interface) 320, memory (memory) 330
With bus 340, wherein processor 310, communication interface 320, memory 330 complete mutual communication by bus 340.It is logical
Letter interface 340 can be used for the transmission of the information between server and smart television.Processor 310 can call in memory 330
Logical order, to execute following method：The image for including peanut is obtained, the feature of peanut part in described image, root are extracted
The character representation of peanut part in described image is obtained according to the feature of the peanut part；According to peanut part in described image
The kind of the peanut in described image is identified in character representation.
The embodiment of the present invention also provides a kind of non-transient computer readable storage medium, and the non-transient computer is readable to deposit
Storage media stores computer instruction, and the computer instruction makes the computer execute the side that above-mentioned each method embodiment is provided
Method, such as including：The image for including peanut is obtained, the feature of peanut part in described image is extracted, according to the peanut part
Feature obtain described image in peanut part character representation；According to the character representation of peanut part in described image, to institute
The kind for stating the peanut in image is identified
The apparatus embodiments described above are merely exemplary, wherein the unit illustrated as separating component can
It is physically separated with being or may not be, the component shown as unit may or may not be physics list
Member, you can be located at a place, or may be distributed over multiple network units.It can be selected according to the actual needs
In some or all of module achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying creativeness
Labour in the case of, you can to understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It is realized by the mode of software plus required general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, on
Stating technical solution, substantially the part that contributes to existing technology can be expressed in the form of software products in other words, should
Computer software product can store in a computer-readable storage medium, such as ROM/RAM, magnetic disc, CD, including several fingers
It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation
Method described in certain parts of example or embodiment.
Finally it should be noted that：The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although
Present invention has been described in detail with reference to the aforementioned embodiments, it will be understood by those of ordinary skill in the art that：It still may be used
With technical scheme described in the above embodiments is modified or equivalent replacement of some of the technical features；
And these modifications or replacements, various embodiments of the present invention technical solution that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (10)
1. a kind of peanut varieties recognition methods, which is characterized in that including：
The image for including peanut is obtained, the feature of peanut part in described image is extracted, is obtained according to the feature of the peanut part
Take the character representation of peanut part in described image；
According to the character representation of peanut part in described image, the kind of the peanut in described image is identified.
2. according to the method described in claim 1, it is characterized in that, the feature for extracting peanut part in described image, is obtained
The step of taking the character representation of peanut part in described image specifically includes：
Described image is divided into the identical square window of multiple sizes, extracts the SIFT descriptors of the square window,
The SIFT descriptors that the constant SIFT descriptors work of wherein translation and illumination variation is represented to peanut object, by the representative
The SIFT descriptors of peanut object are put into Density Clustering, the model for establishing cluster feature；
By density clustering algorithms, the SIFT descriptors for representing peanut object are clustered, multiple cluster centres are built,
And code book is built according to the cluster centre；
According to Bag-of-words model models, the BoW expressions for corresponding to the code book in described image are obtained, it will be described
BoW is denoted as the character representation of peanut part in described image.
3. according to the method described in claim 1, it is characterized in that, the mark sheet according to peanut part in described image
The step of showing, the kind of the peanut in described image is identified specifically includes：
The character representation of peanut part in described image is input in preset SVM classifier, by SVM classifier to described
The character representation of peanut part carries out Classification and Identification in image.
4. according to the method described in claim 3, it is characterized in that, the number of the SVM classifier is k (k-1)/2；Its
In, k is the quantity for the peanut type for needing to identify.
5. according to the method described in claim 4, it is characterized in that, it is described by SVM classifier to peanut portion in described image
The character representation divided carries out Classification and Identification and specifically includes：According to k (k-1)/2 SVM classifier to peanut part in described image
Character representation tested, take the form of ballot, the highest recognition result that will win the vote is as final recognition result.
6. according to the method described in claim 3, it is characterized in that, the method further includes being trained to the SVM classifier
Step, wherein to the SVM classifier training step include：
It includes peanut classification mark to obtain multiple, including the image of peanut is as training sample set, according to the training sample
Set pair SVM classifier is trained.
7. according to the method described in claim 1, it is characterized in that, the mark sheet according to peanut part in described image
The step of showing, the kind of the peanut in described image is identified specifically includes：
The character representation of peanut part in described image is input in preset neural network, by neural network to the figure
The character representation of peanut part carries out Classification and Identification as in.
8. a kind of peanut varieties identifying system, which is characterized in that including：
Preprocessing module includes peanut image for obtaining, extracts the feature of peanut part in described image, obtains described image
The character representation of middle peanut part；
Identification module, for according to the character representation of peanut part in described image, to the kind of the peanut in described image into
Row identification.
9. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is held by processor
The step of method as described in any in claim 1 to 7 is realized when row.
10. a kind of peanut varieties identification equipment, which is characterized in that including：
At least one processor；And at least one processor being connected to the processor, wherein：
The memory is stored with the program instruction that can be executed by the processor, and the processor calls described program to instruct energy
Enough methods executed as described in claim 1 to 7 is any.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810349090.3A CN108776802A (en) | 2018-04-18 | 2018-04-18 | A kind of peanut varieties recognition methods and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810349090.3A CN108776802A (en) | 2018-04-18 | 2018-04-18 | A kind of peanut varieties recognition methods and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108776802A true CN108776802A (en) | 2018-11-09 |
Family
ID=64033900
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810349090.3A Pending CN108776802A (en) | 2018-04-18 | 2018-04-18 | A kind of peanut varieties recognition methods and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108776802A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113743161A (en) * | 2020-05-29 | 2021-12-03 | 合肥美亚光电技术股份有限公司 | Peanut kernel number identification method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663401A (en) * | 2012-04-18 | 2012-09-12 | 哈尔滨工程大学 | Image characteristic extracting and describing method |
| CN106845527A (en) * | 2016-12-29 | 2017-06-13 | 南京江南博睿高新技术研究院有限公司 | A kind of vegetable recognition methods |
| CN106934401A (en) * | 2017-03-07 | 2017-07-07 | 上海师范大学 | A kind of image classification method based on improvement bag of words |
-
2018
- 2018-04-18 CN CN201810349090.3A patent/CN108776802A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663401A (en) * | 2012-04-18 | 2012-09-12 | 哈尔滨工程大学 | Image characteristic extracting and describing method |
| CN106845527A (en) * | 2016-12-29 | 2017-06-13 | 南京江南博睿高新技术研究院有限公司 | A kind of vegetable recognition methods |
| CN106934401A (en) * | 2017-03-07 | 2017-07-07 | 上海师范大学 | A kind of image classification method based on improvement bag of words |
Non-Patent Citations (4)
| Title |
|---|
| MURAT OLGUN 等: "Wheat grain classification by using dense SIFT features with SVM classifier", 《COMPUTERS AND ELECTRONICS IN AGRICULTURE》 * |
| 曾祥鑫: "基于图像的空间典型目标分类识别方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 杨晓敏 等: "一种基于词袋模型的图像分类方法", 《太赫兹科学与电子信息学报》 * |
| 邓立苗 等: "基于机器视觉的花生品种识别系统研究", 《农机化研究》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113743161A (en) * | 2020-05-29 | 2021-12-03 | 合肥美亚光电技术股份有限公司 | Peanut kernel number identification method and device |
| CN113743161B (en) * | 2020-05-29 | 2024-04-09 | 合肥美亚光电技术股份有限公司 | Peanut kernel number identification method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Scharr et al. | Leaf segmentation in plant phenotyping: a collation study | |
| CN105488536B (en) | A kind of agricultural pests image-recognizing method based on multiple features depth learning technology | |
| Kumar et al. | Leafsnap: A computer vision system for automatic plant species identification | |
| CN107480620B (en) | Remote sensing image automatic target identification method based on heterogeneous feature fusion | |
| Flores et al. | Application of convolutional neural networks for static hand gestures recognition under different invariant features | |
| Salman et al. | Leaf classification and identification using Canny Edge Detector and SVM classifier | |
| Alharbi et al. | Automatic counting of wheat spikes from wheat growth images | |
| CN106650804B (en) | A kind of face sample cleaning method and system based on deep learning feature | |
| CN113761259A (en) | Image processing method and device and computer equipment | |
| Jana et al. | Intra-class recognition of fruits using color and texture features with neural classifiers | |
| CN110704662A (en) | Image classification method and system | |
| Ambika et al. | Detection of vanilla species by employing image processing approach | |
| Varghese et al. | INFOPLANT: Plant recognition using convolutional neural networks | |
| Schröder et al. | Low-shot learning of plankton categories | |
| Ahmad et al. | Feature extraction of plant leaf using deep learning | |
| Suthakaran et al. | Detection of the affected area and classification of pests using convolutional neural networks from the leaf images | |
| Mukherjee et al. | Morphological feature based maturity level identification of Kalmegh and Tulsi leaves | |
| Said et al. | Comparison of color-based feature extraction methods in banana leaf diseases classification using SVM and K-NN | |
| Montalvo et al. | Identification of plant textures in agricultural images by principal component analysis | |
| El-Henawy et al. | A new muzzle classification model using decision tree classifier | |
| CN108776802A (en) | A kind of peanut varieties recognition methods and system | |
| EP2383680A1 (en) | Classification of objects in harvesting applications | |
| Suryawati et al. | Deep convolutional adversarial network-based feature learning for tea clones identifications | |
| Ahmed et al. | Drone Detection by Neural Network Using GLCM and SURF | |
| Ghazi et al. | Sabanci-Okan system in LifeCLEF 2015 plant identification competition |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181109 |
|
| RJ01 | Rejection of invention patent application after publication |