CN108932270B - Loquat germplasm resource retrieval contrast method based on Bayes and feedback algorithm - Google Patents
Loquat germplasm resource retrieval contrast method based on Bayes and feedback algorithm Download PDFInfo
- Publication number
- CN108932270B CN108932270B CN201710390590.7A CN201710390590A CN108932270B CN 108932270 B CN108932270 B CN 108932270B CN 201710390590 A CN201710390590 A CN 201710390590A CN 108932270 B CN108932270 B CN 108932270B
- Authority
- CN
- China
- Prior art keywords
- retrieval
- germplasm
- image
- user
- loquat
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a retrieval and comparison method of eriobotrya germplasm resources based on Bayes and feedback algorithms, which is characterized in that at least one retrieval object in a germplasm name, morphological characteristic investigation and an image to be detected is input through a user side of a mobile intelligent device, a cloud server learns and extracts characteristic values of the retrieval object, the Bayes algorithm and PCA are utilized to reduce dimension, a retrieval characteristic value beneficial to efficient query is formed, a cloud server resource database is queried, so that the target germplasm is rapidly identified and relevant information is obtained, a retrieval result is transmitted to the mobile intelligent device, and after the user feeds back the retrieval result, the system can be optimally adjusted. According to the method, the cloud database is inquired through different retrieval objects, the requirements of professional and non-professional persons on-site rapid identification and understanding of the loquat germplasm can be met, the operation is simple, the defects that the traditional retrieval table cannot give consideration to the matching efficiency and the matching accuracy is unstable are overcome, and the method is suitable for popularization.
Description
Technical Field
The invention relates to the technical field of plant germplasm resource retrieval, in particular to a loquat germplasm resource retrieval contrast method based on Bayes and feedback algorithms.
Background
The loquat is originated in China, is a famous Chinese fruit and has outstanding economic value. Due to the complex growing environment and long cultivation history, the germplasm resources of the Eriobotrya plants have extremely rich morphological characteristics. In field investigation and production application, the characteristics of stamen, fruit color, fruit shape, leaf back villus, leaf edge sawteeth, leaf shape, leaf color and other visual morphological characteristics are often important bases for identifying the germplasm of Eriobotrya.
Content-based image retrieval cbir (content-based image retrieval) helps to query information of a target plant, and is increasingly emphasized, it accomplishes retrieval by color, texture, or shape features in the captured image. There have been a lot of research reports on the application in image retrieval, and there have been mobile terminals APP for plant identification in the market, such as "plant identification", "microsoft flower", "shape and color", "chinese plant sign wechat edition" and "Virginia Tech Tree ID" abroad, mainly to distinguish garden plants, flowers and the like, with the genus or species as the distinguishing level, and it is impossible to distinguish germplasm of different subspecies, populations and variety levels of plants in the eriobotrya. However, the invention patents of the present published invention relating to the field of plant identification, such as "plant species identification device" (CN201280030019), "plant identification method and device with high identification rate" (CN201410116111), "plant species identification method with multi-feature fusion of leaf images" (CN201610670754) aim at distinguishing the differences between different plant species or species according to morphological features such as leaves, and are popular science applications that help the public to distinguish plants, but all fail to realize distinguishing and identifying the subspecies, populations and even the species level of a certain species, and fail to meet the demands of some seedlings, learners, breeders, and the like for rapidly identifying and understanding sample germplasm on site. The Eriobotrya germplasm resources are rich, the morphological characteristics of the Eriobotrya germplasm resources are complex and various, the detection equipment is difficult to use in the field, and the readability of the traditional professional retrieval table of the Eriobotrya germplasm resources is poor, so that beginners, non-professionals and the like cannot efficiently and quickly identify target germplasm and obtain detailed information of the germplasm according to known characteristic contrast retrieval items in the field investigation, production and application processes.
Bayesian classification is a basic algorithm of data mining, and the database information is rapidly and accurately classified by applying probability theory knowledge and further used as the basis of other applications. The related feedback technology is firstly applied to text retrieval and is also applied to image retrieval afterwards, so that the retrieval performance can be effectively improved.
Until now, there is no loquat germplasm resource retrieval and comparison system and method which integrates germplasm names, morphological feature investigation and image recognition (leaves, flowers and fruits) as retrieval objects and has both identification and retrieval and comparison functions, and particularly, there is no example of improving the retrieval and comparison efficiency and accuracy of the system through a Bayesian algorithm and a related feedback algorithm.
Disclosure of Invention
The invention aims to provide a loquat germplasm resource retrieval and comparison method based on Bayes and feedback algorithms to overcome the defects in the prior art.
In order to achieve the purpose, the technical scheme of the invention is as follows: a loquat germplasm resource retrieval contrast method based on Bayes and feedback algorithms comprises the following steps:
step S1: a user registers or logs in account information through a handheld terminal carrying a Eriobotrya germplasm resource retrieval system;
step S2: a user starts to input target information to be detected of the eriobotrya germplasm to be judged through the handheld terminal;
step S3: judging whether the target to be detected is a known germplasm or not by a user; if the germplasm is known, the step S4 is executed, otherwise, the step S5 is executed;
step S4: the user inputs the germplasm name through the germplasm character input module, and the method comprises the following steps: english name, latin name and chinese name;
step S5: a user selects whether to acquire an image to be detected of the target to be detected or not through the image; if the image to be detected can be collected, the step is switched to step S6, otherwise, the step is switched to step S7;
step S6: a user acquires and uploads an image to be detected of a target to be detected to a cloud server according to program instructions through an image acquisition and processing module;
step S7: selecting whether to input morphological characteristics by a user; if the input form characteristics are selected, the step is shifted to step S8, otherwise, the step is shifted to step S9;
step S8: if the input morphological characteristics are selected, the user selects morphological characteristic options provided in the retrieval item list through the morphological characteristic query module, selects no unknown retrieval item and determines a retrieval object;
step S9: after the input is confirmed to be correct and at least one retrieval object is input, the handheld terminal uploads the retrieval object to a cloud server;
step S10: after receiving the retrieval object, the cloud server analyzes and integrates the retrieval object into a retrieval instruction, retrieves and contrasts with a database of loquat germplasm resources, and obtains a retrieval result;
step S11: the cloud server analyzes and judges the retrieval result;
step S12: if the matched loquat germplasms exist, the cloud server selects the first 5 retrieval results after sorting according to the matching degree from high to low, and sends the retrieval results to the handheld terminal;
step S13: and if the matched germplasm does not exist, the cloud server sends the information of the loquat germplasm which is not searched for and matched, and the user sends the shared help-seeking information to the social platform through the help-seeking sharing module in the cloud server to share the help-seeking information.
In an embodiment of the present invention, in step S1, the following information is entered in the registration process: user name, demand preferences, age, and educational background.
In an embodiment of the present invention, in the step S6, when acquiring the image, the method further includes the following steps:
step S61: guiding the user to a white background and natural light through a text form;
step S62: guiding a user to collect a background image, comparing the background image with a standard background prestored in the system, analyzing and judging whether a shooting background and light selected by the user can be accepted or not, namely whether the later-stage image analysis is not influenced or not; if the shooting background and the light cannot be accepted, go to step S63, otherwise, go to step S64;
step S63: if the shooting background and the light cannot be accepted, guiding the user to correct the background and the exposure rate, and going to step S61;
step S64: if the shooting background can be accepted, guiding a user to collect an image to be detected, and analyzing and judging whether the collected image can be accepted or not; if the image cannot be accepted, guiding the user to acquire the image again; and if the image can be accepted, guiding the user to upload the image.
In an embodiment of the present invention, the method further comprises the following steps of establishing a database of the eriobotrya germplasm resources:
step S101: establishing a standard loquat germplasm information database, acquiring basic information and morphological characteristic sets of standard germplasms of loquats and standard image sets of leaves, flowers and fruits according to loquat germplasm resource description specifications and data standards, and acquiring samples of different years and different growing places for the same germplasm;
step S102: extracting basic information, morphological and shape feature sets and feature values in standard image sets of leaves, flowers and fruits of the collected and input germplasm, and formatting;
step S103: performing clustering analysis on the formatted germplasm resource data information, and classifying germplasm;
step S104, obtaining morphological characteristic objects with the contribution rate of more than 2.0% to the clustering result through PCA dimension reduction analysis, and sequencing the morphological characteristic objects according to the contribution rate to serve as object values of preferential retrieval;
step S105: and according to the clustering analysis result and the PCA dimension reduction analysis result, extracting the morphological feature object with the contribution rate of more than 2.0 percent as a priority retrieval object, and establishing a retrieval feature vector.
In an embodiment of the present invention, in the step S10, the cloud server further includes a bayesian retrieval determining module, and the bayesian retrieval determining module performs retrieval according to the following steps:
let an n-dimensional space RnThe vector x satisfies the gaussian distribution, i.e. the target germplasm x is a total eigenvalue vector on n characters, and the probability density function retrieved as germplasm x is:
wherein x is [ x ]1,....,xn]、ε＝[ε(x1),...,ε(xn)]And o { (x-E)TRespectively representing a retrieval object characteristic value input by the target germplasm, a standard germplasm characteristic value and a covariance matrix in a resource database;
establishing a Bayes classifier according to the probability density function distribution to obtain the quality x belonging to the category wiThe probability function of (a) is:
in the germplasm information retrieval, x is used as a retrieval feature vector, and P (w) is usedi| x) largest target loquat class wiAs a result of the search.
In an embodiment of the present invention, the cloud server further includes an evaluation feedback module, the user submits a feedback search result to the evaluation feedback module through the handheld terminal, the evaluation feedback module subtracts a feature value of the search result from a positive and negative feedback center distance value, the calculation result is used as a debugging reference value, and the parameter epsilon is adjusted according to the subtraction resultiO' and P (w)i) And (6) adjusting.
In one embodiment of the invention, after the input morphological characteristics are selected, the cloud database carries out data formatting and characteristic value extraction, and transmits the data to the Bayesian retrieval judgment module; and after the image to be detected is input, the cloud database is subjected to drying treatment, then data is formatted and characteristic values are extracted, and the data are transmitted to the Bayesian classification judgment module.
In an embodiment of the present invention, the image to be detected includes at least one of a leaf image, a flower image, or a fruit image of the object to be detected.
Compared with the prior art, the invention has the following beneficial effects: the loquat germplasm resource retrieval contrast method based on Bayes and feedback algorithms solves the technical defects of poor readability, low field identification efficiency and the like of field identification of the germplasm in loquat, including species, subspecies and population to species levels, searches and queries a loquat resource database through germplasm names, known morphological characteristic surveys and leaf, flower and fruit images, reduces complexity and difficulty of loquat germplasm identification, and helps professional and non-professional users to efficiently identify germplasm and obtain relevant information of the germplasm.
Drawings
FIG. 1 is a schematic diagram of a Eriobotrya germplasm resource retrieval and comparison method based on a Bayesian algorithm and a related feedback algorithm in the invention.
Fig. 2 is a schematic diagram of a retrieval object composition of the eriobotrya germplasm resource retrieval comparison system carried by the handheld terminal.
FIG. 3 is a flow chart of the Eriobotrya germplasm resource retrieval and comparison method in the present invention.
Fig. 4 is a flow chart for guiding a user to correctly capture an image in the present invention.
FIG. 5 is a schematic diagram of the Eriobotrya germplasm resource retrieval and comparison system of the present invention.
Detailed Description
The technical scheme of the invention is specifically explained below with reference to the accompanying drawings.
The invention provides a loquat germplasm retrieval comparison method based on Bayes and feedback algorithms, as shown in figure 1, firstly establishing a loquat germplasm resource information database, wherein the database comprises basic information, a morphological feature set, a leaf, a flower and a fruit image set for investigating loquat germplasm, then formatting the information of the database, performing cluster analysis on a plurality of categories for dividing the germplasm, and sequencing the contribution rates of the morphological features in the categories by PCA dimension reduction to reduce the retrieval range, thereby taking the contribution rates as retrieval vector feature values. The method comprises the following specific steps:
step 101: establishing a standard loquat germplasm information database, and investigating and collecting basic information, morphological characteristic sets and standard image sets of leaves, flowers and fruits of loquat standard germplasm according to loquat germplasm resource description specifications and data standards; samples of different years and different growing places are collected for the same germplasm so as to enrich the morphological characteristics and image information of the germplasm.
Step 102: basic information of germplasm, morphological feature survey, and feature values in search objects in collected images (leaves, flowers, and fruits) are collected and input, and formatting is performed. Where the acquired image includes at least one of a leaf, flower, or fruit.
Step 103: and performing clustering analysis on the formatted germplasm resource data information in the database, and classifying the germplasm.
Step 104: and mining morphological characteristic objects with large contribution rate to the clustering result through PCA dimension reduction analysis, and sequencing according to the contribution rate to serve as object values of priority retrieval. If 86 personality traits of 199 loquat germplasms are investigated, see table 1; performing Q-type clustering analysis and PCA analysis, wherein the cumulative contribution rate of the first 28 principal components reaches 74.7%, which is shown in Table 2; if the contribution rate of the 1 st main component is 11.09%, the absolute value of the characteristic vector is the transverse diameter of the fruit, the lateral diameter of the fruit, the quality of the single fruit, the longitudinal diameter of the fruit, the thickness of the fruit pulp, the edibility, the depth of a sepal drum, the thickness of a fruit stem and the number of fruit setting of the spike, and the characteristic vector is more than 0.613, which represents the economic character factor of the fruit; the contribution rate of the No. 2 main component is 6.21%, the absolute values of the characteristic vectors are the length of the side branches, the length of the central branch and the thickness of the side branches, the characteristic vectors are all above 0.793 and represent branch tip factors and the like.
TABLE 1 investigation of 86 personality traits of loquat germplasm resources
TABLE 2 eigenvalues, contribution ratios and cumulative contribution ratios of principal components
Step 105: and extracting the features with the contribution rate of more than 2.0 percent according to the clustering analysis result and the PCA dimension reduction analysis result, and establishing a retrieval feature vector by taking the features as a priority retrieval object.
Step 106: the data is transferred to a Bayesian algorithm in a cloud database and a retrieval judgment module of relevant feedback; the specific Bayesian algorithm and the related feedback algorithm utilized by the method are as follows:
assuming an n-dimensional space RnSatisfies the gaussian distribution, i.e. the target germplasm x is the total eigenvalue vector on the n characters, then the probability density function of the germplasm x is searchedThe number is as follows:
wherein x is [ x ]1,....,xn]，ε＝[ε(x1),...,ε(xn)]，o′＝E{(x-ε)(x-ε)TAnd the characteristic values of the retrieval object input by the target germplasm, the characteristic values of the standard germplasm characters in the resource database and the covariance matrix are respectively represented.
Establishing a Bayesian classifier based on the distribution hypothesis, and obtaining the category w of the germplasm xiHas a probability function of
Wherein, the 3 parameters in the Bayes classifier are epsiloniO' and P (w)i) This assumption has also been demonstrated by the continual updating of the interactive feedback process by the user to continually improve the efficiency and accuracy of the system's search. The user evaluates and feeds back the output retrieval result, and the system evaluates the feedback of the user and then adjusts the 3 parameter values. In the evaluation process, the input retrieval characteristic value of the target germplasm is subtracted from the positive and negative feedback center distance values, and the obtained result value is used as the reference of final debugging. In the germplasm information retrieval, x is used as a retrieval feature vector, and the purpose of the retrieval is to find P (w)i| x) largest target loquat class wiAnd the goal of the user feedback is to find a solution that satisfies:
the i-th loquat germplasm of (1).
Step 107-108: displaying the result according to the similar distance, namely the matching degree, and feeding back the result by a user; and when the match is not retrieved, the user shares the help to the social platform and feeds back the help to the system.
Further, as shown in fig. 4, a flowchart of the germplasm resource retrieval and comparison system of the eriobotrya plant according to an exemplary embodiment is shown, which includes the following steps:
001, selecting the mobile application program and starting to execute the mobile application program by a user;
005, selecting whether the image to be detected can be collected by a user;
further, in the actual acquisition process, the user is difficult to acquire images of leaves, flowers and fruits simultaneously, so that the image to be detected comprises at least one of a leaf image, a flower image or a fruit image of the object to be detected. That is, at least any one of the images including leaves, flowers, and fruits.
Further, in the present embodiment, in step S202 in fig. 1, a method for guiding the user to correctly acquire the image to be inspected is also provided, as shown in fig. 3. The application guides the user to the white background and natural lighting that is optimal when capturing the image (step 302). The application program guides the user to collect background images (step 303), compares the background images with standard backgrounds prestored in the system, and analyzes and judges whether the shooting background and light selected by the user are acceptable or not, namely, the later image analysis has no influence (step 304). If the shot background and light are not acceptable, the application further guides the user how to correct the background and exposure until acceptable. If the capture context is acceptable, the application further directs the user to capture an image to be reviewed (step 305). The user acquires an image of the specimen to be inspected (step 306) and the application analyzes the image to determine if the image is acceptable (step 307). If the image cannot be accepted, the application program further guides the user to acquire the image again; if the image is acceptable, the application directs the user to upload the image.
further, as shown in fig. 2, the user may selectively input at least one retrieval object of a germplasm name, an image to be examined, and a morphological feature investigation result through the mobile intelligent terminal to perform retrieval query. After the input morphological feature investigation is selected, data formatting and feature value extraction are required, and finally the data are transferred to a Bayesian retrieval judgment module of a cloud database; after an image to be detected is input, drying processing needs to be carried out on the image, then data formatting and characteristic value extraction are carried out, and finally the data are transferred to a Bayesian classification judgment module of a cloud database; after the germplasm name is input, the data can be directly transferred to a Bayesian retrieval judgment module of the cloud database.
Step 011, the server analyzes and judges the retrieval result;
and step 013, if the matched germplasm does not exist, the user can share and ask for help from the social platform through the server.
Further, as shown in fig. 5, a schematic diagram of a retrieval comparison system for eriobotrya germplasm resources mainly includes a mobile intelligent device user side and a cloud retrieval server. The loquat germplasm resource retrieval system is carried by a user side of the mobile intelligent device and comprises a germplasm character input module, a morphological characteristic query module, an image acquisition and processing module, a storage module and a wireless communication device, so that the user can conveniently input at least one query instruction of a germplasm name, a known morphological characteristic, an image to be detected and the like and search a cloud database in a contrast manner. The cloud database server comprises a Eriobotrya germplasm database, a database management and retrieval query system, a help-seeking sharing system and a wireless communication device, and is used for retrieving and contrasting a preset database according to a query instruction sent by a user and feeding back a retrieval result to a client according to the sequence of the matching degree.
The above are preferred embodiments of the present invention, and all changes made according to the technical scheme of the present invention that produce functional effects do not exceed the scope of the technical scheme of the present invention belong to the protection scope of the present invention.
Claims (6)
1. A loquat germplasm resource retrieval and comparison method based on Bayes and feedback algorithms is characterized by comprising the following steps:
step S1: a user registers or logs in account information through a handheld terminal carrying a Eriobotrya germplasm resource retrieval system;
step S2: a user starts to input target information to be detected of the eriobotrya germplasm to be judged through the handheld terminal;
step S3: judging whether the target to be detected is a known germplasm or not by a user; if the germplasm is known, the step S4 is executed, otherwise, the step S5 is executed;
step S4: the user inputs the germplasm name through a germplasm character input module in the Eriobotrya germplasm resource retrieval system, and the method comprises the following steps: english name, latin name and chinese name;
step S5: a user selects whether to acquire an image to be detected of the target to be detected or not through the image; if the image to be detected can be collected, the step is switched to step S6, otherwise, the step is switched to step S7;
step S6: a user acquires and uploads an image to be detected of a target to be detected to a cloud server according to program instructions through an image acquisition and processing module in the Eriobotrya germplasm resource retrieval system;
step S7: selecting whether to input morphological characteristics by a user; if the input form characteristics are selected, the step is shifted to step S8, otherwise, the step is shifted to step S9;
step S8: if the input morphological characteristics are selected, a user selects morphological characteristic options provided in a retrieval item list through a morphological characteristic query module in the Eriobotrya germplasm resource retrieval system, selects no unknown retrieval item and determines a retrieval object;
step S9: after the input is confirmed to be correct and at least one retrieval object is input, the handheld terminal uploads the retrieval object to a cloud server matched with the handheld terminal;
step S10: after receiving the retrieval object, the cloud server analyzes and integrates the retrieval object into a retrieval instruction, retrieves and contrasts with a database of loquat germplasm resources, and obtains a retrieval result;
step S11: the cloud server analyzes and judges the retrieval result;
step S12: if the matched loquat germplasms exist, the cloud server selects the first 5 retrieval results after sorting according to the matching degree from high to low, and sends the retrieval results to the handheld terminal;
step S13: if the matched germplasm does not exist, the cloud server sends information of the loquat germplasm which is not searched for and matched, and the user sends shared help-seeking information to the social platform through a help-seeking sharing module in the cloud server to share help;
the method also comprises the following steps of establishing a database of the Eriobotrya germplasm resources:
step S101: establishing a standard loquat germplasm information database, acquiring basic information and morphological characteristic sets of standard germplasms of loquats and standard image sets of leaves, flowers and fruits according to loquat germplasm resource description specifications and data standards, and acquiring samples of different years and different growing places for the same germplasm;
step S102: extracting basic information and morphological feature sets of the collected and input germplasm and feature values in standard image sets of leaves, flowers and fruits, and formatting;
step S103: performing clustering analysis on the formatted germplasm resource data information, and classifying germplasm;
step S104: obtaining morphological feature objects with the contribution rate of more than 2.0% to the clustering result through PCA dimension reduction analysis, and sequencing the morphological feature objects according to the contribution rate to serve as object values of preferential retrieval;
step S105: according to the clustering analysis result and the PCA dimension reduction analysis result, extracting morphological feature objects with contribution rate larger than 2.0% as a priority retrieval object, and establishing a retrieval feature vector;
in step S10, the cloud server further includes a bayesian retrieval determining module, where the bayesian retrieval determining module performs retrieval according to the following steps:
let an n-dimensional space RnThe vector x satisfies the gaussian distribution, i.e. the target germplasm x is a total eigenvalue vector on n characters, and the probability density function retrieved as germplasm x is:
wherein x is [ x ]1,....,xn]、ε＝[ε(x1),...,ε(xn)]And o { (x-E)TRespectively representing retrieval object characteristic values input by target germplasm, standard germplasm characteristic values and covariance matrixes in a resource database;
establishing a Bayes classifier according to the probability density function distribution to obtain the quality x belonging to the category wiThe probability function of (a) is:
in the loquat germplasm information retrieval, x is used as a retrieval feature vector, and P (w) is usedi| x) largest target loquat class wiAs a result of the search.
2. The eriobotrya germplasm resource retrieval and comparison method based on the bayesian and feedback algorithm as claimed in claim 1, wherein in step S1, the following information is entered in the registration process: user name, demand preferences, age, and educational background.
3. The eriobotrya germplasm resource retrieval and comparison method based on bayesian and feedback algorithms according to claim 1, wherein in step S6, when acquiring images, the method further comprises the following steps:
step S61: guiding the user to a white background and natural light through a text form;
step S62: guiding a user to collect a background image, comparing the background image with a standard background prestored in the system, analyzing and judging whether a shooting background and light selected by the user can be accepted or not, namely whether the analysis of the later-stage image to be detected is not influenced or not; if the shooting background and the light cannot be accepted, go to step S63, otherwise, go to step S64;
step S63: if the shooting background and the light cannot be accepted, guiding the user to correct the background and the exposure rate, and going to step S61;
step S64: if the shooting background can be accepted, guiding a user to collect an image to be detected, and analyzing and judging whether the collected image can be accepted or not; if the image cannot be accepted, guiding the user to acquire the image again; and if the image can be accepted, guiding the user to upload the image.
4. The eriobotrya germplasm resource retrieval and comparison method based on Bayes and feedback algorithms as claimed in claim 1, wherein the cloud server further comprises an evaluation feedback module, a user submits a feedback retrieval result to the evaluation feedback module through the handheld terminal, the evaluation feedback module subtracts a characteristic value of the retrieval result from a positive and negative feedback central distance value, a calculation result is used as a debugging reference value, and a parameter epsilon is calculatediO' and P (w)i) And (6) adjusting.
5. The eriobotrya germplasm resource retrieval and comparison method based on the Bayes and feedback algorithm as claimed in claim 1, wherein after the input morphological characteristics are selected, a cloud database carries out data formatting and characteristic value extraction, and data are transmitted to a Bayes retrieval judgment module; and after the image to be detected is input, the cloud database is subjected to drying treatment, then data is formatted and characteristic values are extracted, and the data are transmitted to the Bayesian classification judgment module.
6. The Eriobotrya germplasm resource retrieval and comparison method based on Bayes and feedback algorithms as claimed in claim 1, wherein said image to be detected comprises at least one of leaf image, flower image or fruit image of target to be detected.
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 CN108932270A (en) | 2018-12-04 |
| CN108932270B true 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) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110138775A (en) * | 2019-05-14 | 2019-08-16 | 南京聚智达信息科技有限公司 | A kind of 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 |
Citations (10)
| 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 |
| 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 |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10885098B2 (en) * | 2015-09-15 | 2021-01-05 | Canon Kabushiki Kaisha | Method, system and apparatus for generating hash codes |
-
2017
- 2017-05-27 CN CN201710390590.7A patent/CN108932270B/en active Active
Patent Citations (10)
| 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 |
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108932270A (en) | 2018-12-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107577688B (en) | Original article influence analysis system based on media information acquisition | |
| Jaware et al. | Crop disease detection using image segmentation | |
| CN108932270B (en) | Loquat germplasm resource retrieval contrast method based on Bayes and feedback algorithm | |
| CN106844614A (en) | A kind of floor plan functional area system for rapidly identifying | |
| CN109446985B (en) | Multi-angle plant identification method based on vector neural network | |
| CN110929760A (en) | Garbage classification software based on computer vision | |
| CN109472280B (en) | Method for updating species recognition model library, storage medium and electronic equipment | |
| WO2018077111A1 (en) | Plant type recognition method, recognition apparatus and server | |
| CN108563791A (en) | A kind of construction quality complains the method and system of text classification | |
| CN104036235A (en) | Plant species identification method based on leaf HOG features and intelligent terminal platform | |
| CN110610193A (en) | Method and device for processing labeled data | |
| Mann et al. | Automatic flower detection and phenology monitoring using time‐lapse cameras and deep learning | |
| CN113761259A (en) | Image processing method and device and computer equipment | |
| CN102812357A (en) | Methods And System For Recognizing Wood Species | |
| CN104713835A (en) | Online numerical recognition method for colors of tobacco leaves | |
| CN109213853A (en) | A kind of Chinese community's question and answer cross-module state search method based on CCA algorithm | |
| CN110751035A (en) | Seed corn production identification method and device | |
| WO2023065663A1 (en) | Video editing method and apparatus, and electronic device and storage medium | |
| Zhang et al. | Deep learning based rapid diagnosis system for identifying tomato nutrition disorders | |
| 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 | |
| CN105243390B (en) | Insect image identification detection method and classification of insect method | |
| Mendigoria et al. | Seed architectural phenes prediction and variety classification of dry beans (phaseolus vulgaris) using machine learning algorithms | |
| CN103824083B (en) | Web online species recognition method based on blade complete and partial two-value characteristics | |
| CN111105041B (en) | Machine learning method and device for intelligent data collision |
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 |