CN108932270A - Contrast method is retrieved based on the loquat category germ plasm resource of Bayes and feedback algorithm - Google Patents
Contrast method is retrieved based on the loquat category germ plasm resource of Bayes and feedback algorithm Download PDFInfo
- Publication number
- CN108932270A CN108932270A CN201710390590.7A CN201710390590A CN108932270A CN 108932270 A CN108932270 A CN 108932270A CN 201710390590 A CN201710390590 A CN 201710390590A CN 108932270 A CN108932270 A CN 108932270A
- Authority
- CN
- China
- Prior art keywords
- germplasm
- loquat
- image
- user
- bayes
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Contrast method is retrieved based on the loquat category germ plasm resource of Bayes and feedback algorithm the present invention relates to a kind of, germplasm title is inputted by intelligent movable equipment user end, morphological feature investigation, at least one of image to be checked retrieves object, Cloud Server carries out study and characteristics extraction to retrieval object, after bayesian algorithm and PCA dimensionality reduction, form the retrieval character value for being conducive to efficiently inquire, inquire its Cloud Server resource database, to realize the Rapid identification to target germplasm and obtain relevant information, search result biography is output to intelligent movable equipment, after user feeds back it, system can be optimized and revised.The present invention passes through different retrieval Object Query cloud databases, can meet the needs of profession and layman are to live Rapid identification and understanding loquat germplasm, easy to operate, compensating for conventional retrieval table, to cannot be considered in terms of matching efficiency, matching accuracy unstable, is suitable for promoting.
Description
Technical field
It is especially a kind of based on Bayes and feedback algorithm the present invention relates to the technical field of plant germplasm resource retrieval
Loquat category germ plasm resource retrieves contrast method.
Background technique
Loquat is Chinese name fruit originating from China, and economic value is prominent.Due to complicated growing environment and long cultivation
History, the morphological feature of Eriobotrya germ plasm resource are extremely abundant.In field investigation and production application, stamen feature, fruit
The intuitive morphological feature such as color, fruit shapes, blade back villus, limb sawtooth, leaf and leaf color often identifies that loquat belongs to kind
The important evidence of matter.
Target plant is inquired in content-based image retrieval CBIR (contentbased image retrieval) help
Information, be increasingly taken seriously, it completes to retrieve by color, texture or the shape feature in the image of acquisition.It is existing
Numerous studies are reported in the application in image retrieval, have mobile terminal APP such as " the plant knowledge for plants identification in the market
Not ", " flower is known by Microsoft ", " shape and color ", " Chinese Plants will wechat version " and " the Virginia Tech Tree ID " of foreign countries etc.,
It is to distinguish level with category or inter-species mainly to distinguish ornamental plant, flowers etc., the different sub- of loquat category implants cannot be distinguished
The germplasm of kind, population and varietal level.And patent of invention " the floristics identification about plants identification field of existing announcement
Device " (CN201280030019), " a kind of plants identification method and device that discrimination is high " (CN201410116111), " leaf
Plant species identification method of picture multiple features fusion " (CN201610670754) etc. be intended to according to the morphological features such as blade come
The difference for distinguishing different plant categories or inter-species contributes to the public science popularization application for distinguishing plant, but is not able to achieve a certain object
The subspecies of kind, the population even Division identification of varietal level, are unable to satisfy the scene such as some seedling quotient, learner, breeder quickly
Identification and understand sample germplasm demand.And loquat category germ plasm resource is abundant, the complicated multiplicity of morphological feature, detection device exists
Field also more difficult use, and the professional retrieval table of traditional loquat category germ plasm resource, but exist it is readable poor so that beginner with
And layman etc. efficiently can not retrieve entry according to known feature comparison during field investigation, production application
Rapid identification target germplasm and the details for obtaining germplasm.
Bayes's classification is the rudimentary algorithm of data mining, fast by carrying out with probability theory knowledge to database information
Speed, accurately classification, are further used as the basis of other application.And Relevance Feedback is applied to earliest in text retrieval, after
It is also applied in image retrieval, can effectively improve retrieval performance.
So far it yet there are no by integrating germplasm title, morphological feature investigation, image recognition (leaf, flower, fruit) as retrieval pair
The loquat category germ plasm resource retrieval contradistinction system taken into account identification and retrieve control function and method of elephant, especially by Bayes
The example that algorithm and Relevance Feedback Algorithms improve retrieval the control efficiency and accuracy of system.
Summary of the invention
The purpose of the present invention is to provide a kind of retrieved based on the loquat category germ plasm resource of Bayes and feedback algorithm to compare
Method, to overcome defect existing in the prior art.
To achieve the above object, the technical scheme is that：It is a kind of that kind is belonged to based on the loquat of Bayes and feedback algorithm
Matter resource retrieval contrast method, includes the following steps：
Step S1：The handheld terminal registration or login account that user carries loquat category germ plasm resource searching system by one
Information；
Step S2：User starts the target information to be checked of typing loquat category germplasm to be judged by the handheld terminal；
Step S3：User judges whether the target to be checked is known germplasm；If known germplasm, then it is transferred to step S4, it is no
Then it is transferred to step S5；
Step S4：User inputs germplasm title by germplasm text input module, including：English name, latin name and Chinese
Name；
Step S5：Whether user can acquire the image to be checked of target to be checked by image selection；If image to be checked can acquire,
It is then transferred to step S6, is otherwise transferred to step S7；
Step S6：User is indicated to acquire and upload the figure to be checked of target to be checked by program by Image Acquisition and processing module
As to cloud server；
Step S7：User chooses whether input morphological feature；If selection input morphological feature, is transferred to step S8, otherwise
It is transferred to step S9；
Step S8：If selection input morphological feature, user passes through in morphological feature enquiry module selection retrieval bulleted list
The morphological feature option of offer does not elect to unknown retrieval project, determines retrieval object；
Step S9：Confirmation input is errorless and at least after one retrieval object of input, and the handheld terminal will be retrieved on object
Reach cloud server；
Step S10：After cloud server receives retrieval object, analytical integration is carried out into search instruction, retrieval control loquat
Belong to the database of germ plasm resource, obtains search result；
Step S11：Cloud server analyzes and determines search result；
Step S12：The loquat germplasm to match if it exists, then after cloud server sorts from high to low by matching degree,
Preceding 5 search results are selected, handheld terminal is issued to；
Step S13：If not there is no the germplasm to match, cloud server issues one and does not retrieve matched loquat germplasm
Information, user send shared help information to social platform by the sharing module of seeking help in cloud server, carry out shared ask
It helps.
In an embodiment of the present invention, in the step S1, the following information of typing in the registration process：User name
Title, Demand perference, age and education background.
It in an embodiment of the present invention, further include following steps when acquiring image in the step S6：
Step S61：Guide user to white background and available light by written form；
Step S62：It guides user to acquire background image, is compared with the normal background prestored in system, analyzed and determined
Whether can the selected shooting background of user and light be received, i.e., analyze later image without influence；If shooting background with
Light can not be received, then go to step S63, otherwise, go to step S64；
Step S63：If shooting background can not be received with light, user's correcting background and exposure rate are guided, and go to step
Rapid S61；
Step S64：If shooting background can be received, guidance user acquires image to be checked, and analyzes and determines the acquisition
Whether image can be received；If image can not be received, guidance user acquires image again；If image can be received, guidance is used
Family uploads image.
It in an embodiment of the present invention, further include the database for establishing the loquat category germ plasm resource in accordance with the following steps：
Step S101：Standard loquat germplasm information database is established, according to《Studies on loquat germplasm Description standard and data mark
It is quasi-》Obtain basic information, morphological feature collection and the leaf of loquat standard germplasm, flower, fruit standard drawing image set, and for same
Matter acquires the sample of different year, different growing locations；
Step S102：It extracts and to acquire and input basic information, form shapes feature set and the leaf of germplasm, flower, fruit
The characteristic value that standard picture is concentrated, and it is formatted processing；
Step S103：Clustering is carried out to formatted germ plasm resource data information, germplasm is classified；
Step S104:By PCA Dimension Reduction Analysis, the morphological feature object to cluster result contribution rate greater than 2.0% is obtained,
And be ranked up according to contribution rate size, as the object value preferentially retrieved；
Step S105：According to cluster analysis result and PCA Dimension Reduction Analysis as a result, extracting the form that contribution rate is greater than 2.0%
Feature object establishes searching characteristic vector as preferential retrieval object.
In an embodiment of the present invention, in the step S10, the cloud server further includes that Bayes retrieval is sentenced
Disconnected module, Bayes's retrieval module are retrieved in accordance with the following steps：
Remember a n-dimensional space RnVector x meet Gaussian Profile, i.e. target germplasm x be from the total characteristic value in n character to
Amount, and the probability density function for being retrieved as germplasm x is：
Wherein, x=[x1,....,xn], ε=[ε (x1),...,ε(xn)] and o '=E { (x- ε) (x- ε)T, it respectively indicates
The retrieval characteristics of objects value that target germplasm is inputted, standard germplasm properties and characteristics value and covariance matrix in resource database；
Bayes classifier is established according to above-mentioned probability density function profiles, germplasm x is obtained and belongs to classification wiProbability letter
Number is：
In germplasm information retrieval, using x as searching characteristic vector, and by P (wi| x) maximum target loquat classification wiMake
For search result.
In an embodiment of the present invention, the cloud server further includes an evaluation feedback module, and user passes through the hand
It holds terminal and submits feedback searching as a result, the evaluation feedback module is by the characteristic value of the search result to the evaluation feedback module
Subtract each other with the centre distance value of positive-negative feedback, using the calculated result as debugging reference point, to parameter εi, o ' and P (wi) carry out
Adjustment.
In an embodiment of the present invention, after selection inputs morphological feature, the cloud database carries out data format
With extraction characteristic value, Bayes's searching and determining module is passed data to；After inputting image to be checked, the cloud database into
Row removes dryness processing, and then data formatization and extraction characteristic value, pass data to Bayes's classification judgment module.
In an embodiment of the present invention, the image to be checked includes in the leaf image, floral diagram picture or fruit image of target to be checked
At least one of.
Compared to the prior art, the invention has the advantages that：One kind proposed by the invention be based on Bayes and
The loquat category germ plasm resource of feedback algorithm retrieves contrast method, solve field identification loquat belong in germplasm kind, subspecies, population
To varietal level, the technological deficiencies such as conventional retrieval table readability is poor, field determination rates are low are overcome, by germplasm title,
The some morphological features investigation known and leaf, flower and fruit image carry out retrieval and inquisition loquat resource database, reduce loquat germplasm mirror
Fixed complexity and difficulty helps profession and unprofessional user efficiently to identify germplasm and obtains the relevant information of the germplasm.
Detailed description of the invention
Fig. 1 is to retrieve contrast method based on the loquat category germ plasm resource of bayesian algorithm and Relevance Feedback Algorithms in the present invention
Schematic diagram.
Fig. 2 is the retrieval object composition signal for the loquat category germ plasm resource retrieval contradistinction system that handheld terminal of the present invention carries
Figure.
Fig. 3 is the flow chart of loquat category germ plasm resource retrieval contrast method in the present invention.
Fig. 4 is in the present invention for guiding user correctly to acquire the flow chart of image.
Fig. 5 is the schematic diagram of loquat category germ plasm resource retrieval contradistinction system of the invention.
Specific embodiment
With reference to the accompanying drawing, technical solution of the present invention is specifically described.
The present invention provide it is of the invention it is a kind of contrast method is retrieved based on the loquat germplasm of Bayes and feedback algorithm, such as scheme
Shown in 1, loquat category Germplasm Resources Information database is initially set up, it is special including the basic information of investigation loquat germplasm, form
Collection, leaf, flower and fruit image set, then the information of database is formatted, if the Ganlei that clustering divides germplasm
Not, PCA dimensionality reduction is ranked up the contribution rate of morphological feature in classification, reduces range of search, in this, as retrieval vector characteristics
Value.Specific step is as follows：
Step 101：Standard loquat germplasm information database is established, according to《Studies on loquat germplasm Description standard and data mark
It is quasi-》Collect basic information, morphological feature collection and the leaf of loquat standard germplasm, flower, fruit standard drawing image set；For same
Matter acquires the sample of different year, different growing locations, to enrich the morphological feature and image information of the germplasm.
Step 102：The basic information of acquisition and input germplasm, morphological feature are investigated, in the image (leaf, flower, fruit) of acquisition
Retrieval object in characteristic value, be formatted processing.The image acquired at this includes at least one of leaf, flower or fruit.
Step 103：Clustering is carried out to germ plasm resource data information formatted in database, germplasm is divided
Class.
Step 104：By PCA Dimension Reduction Analysis, excavate to the biggish morphological feature object of cluster result contribution rate, and according to
Contribution rate size is ranked up, as the object value preferentially retrieved.86 characters as investigated 199 parts of loquat germplasm, are shown in Table
1；Q type clustering and PCA analysis are carried out, preceding 28 principal components accumulate contribution rate 74.7%, are shown in Table 2；Such as the 1st principal component tribute
Offering rate is 11.09%, feature vector absolute value it is big be fruit transverse diameter, fruit side diameter, fruit mass, that fruit indulges diameter, pulp is thick
Degree, edible rate, calyx tube depth, carpopodium rugosity, fringe are beared fruit number, and feature vector all 0.613 or more, represents Fruit Economic Traits
The factor；2nd principal component contributor rate is 6.21%, and it is that side shoot length, center branch length, side shoot are thick that feature vector absolute value is biggish
Degree, feature vector all 0.793 or more, represent branch factor etc..
86 characters of studies on loquat germplasm that table 1 is investigated and analysed
Characteristic value, contribution rate and the contribution rate of accumulative total of 2 principal component of table
Step 105：According to cluster analysis result and PCA Dimension Reduction Analysis as a result, extracting the feature that contribution rate is greater than 2.0%,
As preferential retrieval object, searching characteristic vector is established.
Step 106：Data change the searching and determining module for passing bayesian algorithm and relevant feedback in cloud database；We
The specific bayesian algorithm and Relevance Feedback Algorithms that method utilizes are as follows：
It is assumed that n-dimensional space RnVector x meet Gaussian Profile, i.e. target germplasm x be from the total characteristic value in n character to
Amount, then the probability density function for being retrieved as germplasm x is：
Wherein, x=[x1,....,xn], ε=[ε (x1),...,ε(xn)], o '=E { (x- ε) (x- ε)T, respectively indicate mesh
Mark the retrieval characteristics of objects value that germplasm is inputted, standard germplasm properties and characteristics value and covariance matrix in resource database.
Hypothesis based on this distribution establishes Bayes classifier, and available germplasm x belongs to classification wiProbability function
For
Wherein, 3 parameters in Bayes classifier:εi, o ' and P (wi) according to the interaction feedback process of user constantly more
Newly, the recall precision and accuracy of system are continuously improved with this, which has also been proved to.Retrieval knot of the user to output
Fruit carries out evaluation feedback, and after system can evaluate the feedback of user, adjust above 3 parameter values.In evaluation procedure, mesh
The centre distance value for marking the retrieval character value and positive-negative feedback that are inputted of germplasm is subtracted each other, and is made according to obtained end value
For the reference finally debugged.In germplasm information retrieval, for x as searching characteristic vector, the purpose of retrieval is to find P (wi|
X) maximum target loquat classification wi, and the target of user feedback is to find satisfaction：?
I class loquat germplasm.
Step 107~108：It is shown by similarity distance, that is, matching degree as a result, user feeds back it；When not retrieving
When matching, user, which shares, to seek help to social platform, and feeds back to system.
Further, as shown in figure 4, to be retrieved according to this Eriobotrya germ plasm resource shown in an exemplary embodiment
The flow chart of contradistinction system, includes the following steps：
Step 001, user selects this action application program and starts to execute this action application program；
Step 002, user's registration or login account information mainly include user's name, Demand perference, age, religion
Educate the relevant informations such as background；
Step 003, user judges whether known germplasm title；
Step 004, if known germplasm title, germplasm title, including English name, latin name and Chinese name, the step are inputted
In, user terminal can also click directly on confirmation inquiry, and germplasm title is sent to cloud server；
Step 005, user chooses whether that image to be checked can be acquired；
Step 006, if image to be checked can acquire, correct acquisition is indicated by program and uploads image, in the step, user terminal
Confirmation inquiry can be also clicked directly on, the image of input is sent to cloud server；
Further, since during actual acquisition, user's acquisition is difficult to acquire leaf, flower and the image of fruit simultaneously,
Therefore image to be checked includes at least one in the leaf image, floral diagram picture or fruit image of target to be checked.That is, including at least leaf, flower
And any one of image of fruit.
Further, in the present embodiment, in the step S202 in Fig. 1, also provide for guidance user correctly acquire to
The method for examining image, as shown in Figure 3.The application program guides user to white background and available light for most when acquiring image
It is suitable for (step 302).The application program guides user to acquire background image (step 303), with the normal background prestored in system
It is compared, analyzes and determines the selected shooting background of user and the whether acceptable i.e. later image analysis of light without influence (step
It is rapid 304).If shooting background can not be received with light, the application program further guide user how correcting background and exposure
Rate, until can be received.If shooting background can be received, which further guides user to acquire image to be checked
(step 305).User's progress Image Acquisition (step 306) to be checked to sample, the application program analyze and determine that the image whether may be used
Receive (step 307).If image can not be received, which further guides user to acquire image again；If image can
Received, which guides user to upload image.
Step 007, user chooses whether input morphological feature；
It step 008, can to unknown retrieval project if selection input morphological feature, please click search terms purpose option for features
It does not elect；
Step 009, confirmation input is errorless and at least after one retrieval object of input, clicks index button, will retrieve object
It is transferred to cloud server；
Step 010, after cloud server receives retrieval object, analytical integration is carried out into search instruction, retrieval control loquat
Belong to the database of germ plasm resource；
Further, as shown in Fig. 2, user can selectively be inputted by mobile intelligent terminal germplasm title, image to be checked,
At least one of morphological feature investigation result retrieves object, carries out retrieval and inquisition.After selection input morphological feature investigation, need
It carries out data formatization and extracts characteristic value, data are finally changed to the Bayes's searching and determining module for passing cloud database；When defeated
It after entering image to be checked, needs to carry out it to remove dryness processing, then data formatization and extraction characteristic value, data are finally changed pass
To the Bayes's classification judgment module of cloud database；After inputting germplasm title, directly data can be changed and pass cloud database
Bayes's searching and determining module.
Step 011, server analyzes and determines search result；
Step 012, it if the germplasm for thering is loquat to match, sorts from high to low by matching degree, selects preceding 5 retrievals knot
Fruit is transferred to intelligent movable equipment user end；
Step 013, if not there is no the germplasm to match, user can carry out shared seek help to social platform by server.
Further, the retrieval contradistinction system theory of constitution figure of loquat category germ plasm resource as shown in Fig. 5 mainly includes
Intelligent movable equipment user end and cloud retrieval server.Wherein, loquat category germplasm money is carried at intelligent movable equipment user end
Source searching system, including germplasm text input module, morphological feature enquiry module, Image Acquisition and processing module, memory module
And wireless communication apparatus, pass through at least one of input germplasm title, known morphological feature and image to be checked etc. convenient for user
Inquiry instruction, cross reference cloud database.Cloud database server, including loquat category germplasm database, data base administration
With retrieval and inquisition system, shared system of seeking help and wireless communication apparatus, inquiry instruction for being sent according to user, retrieval control
Preset database sorts by matching degree height to client feedback search result.
The above are preferred embodiments of the present invention, all any changes made according to the technical solution of the present invention, and generated function is made
When with range without departing from technical solution of the present invention, all belong to the scope of protection of the present invention.
Claims (8)
1. a kind of retrieve contrast method based on the loquat category germ plasm resource of Bayes and feedback algorithm, which is characterized in that including such as
Lower step：
Step S1：The handheld terminal registration or login account letter that user carries loquat category germ plasm resource searching system by one
Breath；
Step S2：User starts the target information to be checked of typing loquat category germplasm to be judged by the handheld terminal；
Step S3：User judges whether the target to be checked is known germplasm；If known germplasm, then it is transferred to step S4, is otherwise turned
Enter step S5；
Step S4：User inputs germplasm title by germplasm text input module in loquat category germ plasm resource searching system, including：
English name, latin name and Chinese name；
Step S5：Whether user can acquire the image to be checked of target to be checked by image selection；If image to be checked can acquire, turn
Enter step S6, is otherwise transferred to step S7；
Step S6：User is acquired simultaneously by Image Acquisition in loquat category germ plasm resource searching system and processing module by program instruction
The image to be checked of target to be checked is uploaded to cloud server；
Step S7：User chooses whether input morphological feature；If selection input morphological feature, is transferred to step S8, is otherwise transferred to
Step S9；
Step S8：If selection input morphological feature, user inquire mould by morphological feature in loquat category germ plasm resource searching system
The morphological feature option provided in block selection retrieval bulleted list, does not elect to unknown retrieval project, determines retrieval object；
Step S9：Confirmation input is errorless and at least after one retrieval object of input, and the handheld terminal is uploaded to object is retrieved
With the matched cloud server of the hand-held mobile terminal；
Step S10：After cloud server reception retrieval object, analytical integration is carried out into search instruction, retrieval control loquat category kind
The database of matter resource obtains search result；
Step S11：Cloud server analyzes and determines search result；
Step S12：The loquat germplasm to match if it exists is selected after then cloud server sorts from high to low by matching degree
Preceding 5 search results, are issued to handheld terminal；
Step S13：If not there is no the germplasm to match, cloud server issues one and does not retrieve matched loquat germplasm letter
Breath, user send shared help information to social platform by the sharing module of seeking help in cloud server, carry out shared seek help.
2. according to claim 1 retrieve contrast method based on the loquat category germ plasm resource of Bayes and feedback algorithm,
It is characterized in that, in the step S1, the following information of typing in the registration process：User's name, Demand perference, the age and
Education background.
3. according to claim 1 retrieve contrast method based on the loquat category germ plasm resource of Bayes and feedback algorithm,
It is characterized in that, further includes following steps when acquiring image in the step S6：
Step S61：Guide user to white background and available light by written form；
Step S62：It guides user to acquire background image, is compared with the normal background prestored in system, analyze and determine user
Can selected shooting background and light be received, i.e., whether on later period image analysis to be checked without influence；If shooting background with
Light can not be received, then go to step S63, otherwise, go to step S64；
Step S63：If shooting background can not be received with light, user's correcting background and exposure rate are guided, and go to step
S61；
Step S64：If shooting background can be received, guidance user acquires image to be checked, and analyzes and determines that the institute acquires image
Whether can be received；If image can not be received, guidance user acquires image again；If image can be received, guide on user
Blit picture.
4. according to claim 1 retrieve contrast method based on the loquat category germ plasm resource of Bayes and feedback algorithm,
It is characterized in that, further includes the database for establishing the loquat category germ plasm resource in accordance with the following steps：
Step S101：Standard loquat germplasm information database is established, according to《Studies on loquat germplasm Description standard and data standard》
Obtain basic information, morphological feature collection and the leaf of loquat standard germplasm, flower, fruit standard drawing image set, and for same germplasm
Acquire the sample of different year, different growing locations；
Step S102：Extract acquire and input basic information, morphological feature collection and the leaf of germplasm, flower, fruit standard picture
The characteristic value of concentration, and it is formatted processing；
Step S103：Clustering is carried out to formatted germ plasm resource data information, germplasm is classified；
Step S104：By PCA Dimension Reduction Analysis, the morphological feature object to cluster result contribution rate greater than 2.0%, and root are obtained
It is ranked up according to contribution rate size, as the object value preferentially retrieved；
Step S105：According to cluster analysis result and PCA Dimension Reduction Analysis as a result, extracting form spy of the contribution rate greater than 2.0%
Object is levied, as preferential retrieval object, establishes searching characteristic vector.
5. according to claim 4 retrieve contrast method based on the loquat category germ plasm resource of Bayes and feedback algorithm,
It is characterized in that, in the step S10, the cloud server further includes Bayes's searching and determining module, Bayes inspection
Rope module is retrieved in accordance with the following steps：
Remember a n-dimensional space RnVector x meet Gaussian Profile, i.e. target germplasm x be by the total characteristic value vector in n character, and
The probability density function for being retrieved as germplasm x is：
Wherein, x=[x1,....,xn], ε=[ε (x1),...,ε(xn)] and o '=E { (x- ε) (x- ε)TRespectively indicate target species
The retrieval characteristics of objects value that matter is inputted, standard germplasm properties and characteristics value and covariance matrix in resource database；
Bayes classifier is established according to above-mentioned probability density function profiles, germplasm x is obtained and belongs to classification wiProbability function be：
In loquat category germplasm information retrieval, using x as searching characteristic vector, and by P (wi| x) maximum target loquat classification wi
As search result.
6. according to claim 5 retrieve contrast method based on the loquat category germ plasm resource of Bayes and feedback algorithm,
It is characterized in that, the cloud server further includes an evaluation feedback module, and user is anti-to the evaluation by the handheld terminal
It presents module and submits feedback searching as a result, the evaluation feedback module is by the center of the characteristic value of the search result and positive-negative feedback
Distance value subtracts each other, using the calculated result as debugging reference point, to parameter εi, o ' and P (wi) be adjusted.
7. according to claim 5 retrieve contrast method based on the loquat category germ plasm resource of Bayes and feedback algorithm,
It is characterized in that, after selection inputs morphological feature, the cloud database carries out data formatization and extracts characteristic value, by data
Pass to Bayes's searching and determining module；After inputting image to be checked, the cloud database carries out removing dryness processing, then data
Characteristic value is formatted and extracted, Bayes's classification judgment module is passed data to.
8. according to claim 1 retrieve contrast method based on the loquat category germ plasm resource of Bayes and feedback algorithm,
It is characterized in that, the image to be checked includes at least one in the leaf image, floral diagram picture or fruit image of target to be checked.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710390590.7A CN108932270B (en) | 2017-05-27 | 2017-05-27 | Loquat germplasm resource retrieval contrast method based on Bayes and feedback algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710390590.7A CN108932270B (en) | 2017-05-27 | 2017-05-27 | Loquat germplasm resource retrieval contrast method based on Bayes and feedback algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108932270A true CN108932270A (en) | 2018-12-04 |
| CN108932270B CN108932270B (en) | 2021-11-30 |
Family
ID=64450766
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710390590.7A Active CN108932270B (en) | 2017-05-27 | 2017-05-27 | Loquat germplasm resource retrieval contrast method based on Bayes and feedback algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108932270B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111078732A (en) * | 2019-12-31 | 2020-04-28 | 新疆农业科学院园艺作物研究所 | Method for identifying almond variety |
| WO2020228007A1 (en) * | 2019-05-14 | 2020-11-19 | 南京聚智达信息科技有限公司 | Cloud service information sharing method |
| CN113469530A (en) * | 2021-07-01 | 2021-10-01 | 辽宁省果树科学研究所 | Method and system for managing germplasm resources and investigating field |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101701916A (en) * | 2009-12-01 | 2010-05-05 | 中国农业大学 | Method for quickly identifying and distinguishing variety of corn |
| CN102663396A (en) * | 2012-03-09 | 2012-09-12 | 华中科技大学 | Method for automatically detecting rice milky ripe stage |
| CN103559507A (en) * | 2013-11-18 | 2014-02-05 | 沈阳工业大学 | Method for detecting traffic signs based on combination of color feature and shape feature |
| CN103902996A (en) * | 2014-03-31 | 2014-07-02 | 合肥晶奇电子科技有限公司 | Mobile phone APP designing method for recognizing diversification plants |
| CN104331713A (en) * | 2014-11-04 | 2015-02-04 | 无锡北斗星通信息科技有限公司 | Mobile terminal for detecting leaf varieties |
| CN105472553A (en) * | 2015-11-17 | 2016-04-06 | 李楠 | Plant identification method and system based on mobile terminal |
| CN105760365A (en) * | 2016-03-14 | 2016-07-13 | 云南大学 | Probability latent parameter estimation model of image semantic data based on Bayesian algorithm |
| CN106202994A (en) * | 2016-07-04 | 2016-12-07 | 上海交通大学 | The holographyization authentication method of a kind of living species or tissue and system |
| CN106295661A (en) * | 2016-08-15 | 2017-01-04 | 北京林业大学 | The plant species identification method of leaf image multiple features fusion and device |
| US20170075887A1 (en) * | 2015-09-15 | 2017-03-16 | Canon Kabushiki Kaisha | Method, system and apparatus for generating hash codes |
| CN106599925A (en) * | 2016-12-19 | 2017-04-26 | 广东技术师范学院 | Plant leaf identification system and method based on deep learning |
-
2017
- 2017-05-27 CN CN201710390590.7A patent/CN108932270B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101701916A (en) * | 2009-12-01 | 2010-05-05 | 中国农业大学 | Method for quickly identifying and distinguishing variety of corn |
| CN102663396A (en) * | 2012-03-09 | 2012-09-12 | 华中科技大学 | Method for automatically detecting rice milky ripe stage |
| CN103559507A (en) * | 2013-11-18 | 2014-02-05 | 沈阳工业大学 | Method for detecting traffic signs based on combination of color feature and shape feature |
| CN103902996A (en) * | 2014-03-31 | 2014-07-02 | 合肥晶奇电子科技有限公司 | Mobile phone APP designing method for recognizing diversification plants |
| CN104331713A (en) * | 2014-11-04 | 2015-02-04 | 无锡北斗星通信息科技有限公司 | Mobile terminal for detecting leaf varieties |
| US20170075887A1 (en) * | 2015-09-15 | 2017-03-16 | Canon Kabushiki Kaisha | Method, system and apparatus for generating hash codes |
| CN105472553A (en) * | 2015-11-17 | 2016-04-06 | 李楠 | Plant identification method and system based on mobile terminal |
| CN105760365A (en) * | 2016-03-14 | 2016-07-13 | 云南大学 | Probability latent parameter estimation model of image semantic data based on Bayesian algorithm |
| CN106202994A (en) * | 2016-07-04 | 2016-12-07 | 上海交通大学 | The holographyization authentication method of a kind of living species or tissue and system |
| CN106295661A (en) * | 2016-08-15 | 2017-01-04 | 北京林业大学 | The plant species identification method of leaf image multiple features fusion and device |
| CN106599925A (en) * | 2016-12-19 | 2017-04-26 | 广东技术师范学院 | Plant leaf identification system and method based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| 陈文兵 等: "改进EM算法在图像检索中的应用", 《计算机工程》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020228007A1 (en) * | 2019-05-14 | 2020-11-19 | 南京聚智达信息科技有限公司 | Cloud service information sharing method |
| CN111078732A (en) * | 2019-12-31 | 2020-04-28 | 新疆农业科学院园艺作物研究所 | Method for identifying almond variety |
| CN113469530A (en) * | 2021-07-01 | 2021-10-01 | 辽宁省果树科学研究所 | Method and system for managing germplasm resources and investigating field |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108932270B (en) | 2021-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110826320B (en) | Sensitive data discovery method and system based on text recognition | |
| CN107577688B (en) | Original article influence analysis system based on media information acquisition | |
| CN103824053B (en) | The sex mask method and face gender detection method of a kind of facial image | |
| CN114565826B (en) | Agricultural pest and disease identification and diagnosis method, system and device | |
| CN108932270A (en) | Contrast method is retrieved based on the loquat category germ plasm resource of Bayes and feedback algorithm | |
| CN107145879A (en) | A kind of floristics automatic identifying method and system | |
| CN113657469B (en) | Automatic observation method and system for woody plant waiting period based on image recognition | |
| CN104036235A (en) | Plant species identification method based on leaf HOG features and intelligent terminal platform | |
| Mann et al. | Automatic flower detection and phenology monitoring using time‐lapse cameras and deep learning | |
| CN111627088A (en) | Sample automatic generation method for mathematical test paper image recognition | |
| CN109213853A (en) | A kind of Chinese community's question and answer cross-module state search method based on CCA algorithm | |
| CN109948429A (en) | Image analysis method, device, electronic equipment and computer-readable medium | |
| CN113282834A (en) | Web search intelligent ordering method, system and computer storage medium based on mobile internet data deep mining | |
| CN112115950A (en) | Wine label identification method, wine information management method, device, equipment and storage medium | |
| CN115099297A (en) | Soybean plant phenotype data statistical method based on improved YOLO v5 model | |
| CN110533102A (en) | Single class classification method and classifier based on fuzzy reasoning | |
| CN107943937B (en) | Debtor asset monitoring method and system based on judicial public information analysis | |
| KR101673558B1 (en) | System and method for providing plant information using smart device | |
| CN113032603A (en) | Bird image recognition and retrieval system and using method | |
| CN116258844A (en) | Rapid and accurate identification method for phenotype character of cotton leaf | |
| CN113157847B (en) | Method and device for rapidly checking forest plant survey data | |
| CN109543512A (en) | The evaluation method of picture and text abstract | |
| CN114676336A (en) | Intelligent classification network search system convenient for retrieval and search | |
| CN206236111U (en) | A kind of leaf image plant automatic identification equipment based on interactive voice | |
| CN108052503B (en) | Confidence coefficient calculation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |