CN108268446A - A kind of processing method and processing device of defect information - Google Patents
A kind of processing method and processing device of defect information Download PDFInfo
- Publication number
- CN108268446A CN108268446A CN201810039720.7A CN201810039720A CN108268446A CN 108268446 A CN108268446 A CN 108268446A CN 201810039720 A CN201810039720 A CN 201810039720A CN 108268446 A CN108268446 A CN 108268446A
- Authority
- CN
- China
- Prior art keywords
- defect
- descriptor
- dictionary
- standardization
- information
- 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
- 230000007547 defect Effects 0.000 title claims abstract description 322
- 238000003672 processing method Methods 0.000 title claims abstract description 17
- 238000000034 method Methods 0.000 claims abstract description 12
- 239000012141 concentrate Substances 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 claims description 2
- 238000007619 statistical method Methods 0.000 abstract description 5
- 230000033228 biological regulation Effects 0.000 abstract description 4
- 230000005540 biological transmission Effects 0.000 description 7
- 230000007797 corrosion Effects 0.000 description 7
- 238000005260 corrosion Methods 0.000 description 7
- 230000005611 electricity Effects 0.000 description 6
- 239000003795 chemical substances by application Substances 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 239000008186 active pharmaceutical agent Substances 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 1
- 230000003628 erosive effect Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- JEIPFZHSYJVQDO-UHFFFAOYSA-N iron(III) oxide Inorganic materials O=[Fe]O[Fe]=O JEIPFZHSYJVQDO-UHFFFAOYSA-N 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
- G06F40/247—Thesauruses; Synonyms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/31—Indexing; Data structures therefor; Storage structures
- G06F16/313—Selection or weighting of terms for indexing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Computational Linguistics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Public Health (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a kind of processing method and processing device of defect information, wherein this method includes：The artificial description defect information of each image in image set is read, and artificial description defect information is decomposed into multiple keywords；It obtains with describing the relevant keyword of benchmark word, and the relevant keyword of benchmark word will be described with every class and separately constitutes all kinds of dictionaries, be utilized respectively the defects of corresponding descriptor and replace the keyword included in dictionary；Default put in order that benchmark word is described according to every class is ranked up for corresponding dictionary, and defect description information is standardized to obtain.Above-mentioned technical proposal disclosed in the present application can obtain standardization defect description information, for statistical analysis in order to the automatic input of defect information and to defect information so as to carry out regulation and standardization to manually describing defect information.
Description
Technical field
The present invention relates to field of computer technology, more specifically to a kind of processing method and processing device of defect information.
Background technology
Transmission line of electricity is one of important component of electric system, wherein, detect transmission line of electricity the defects of be power transmission line
Road becomes more meticulous the main target of inspection.
In the prior art, mostly using being recorded and information the defects of analyze transmission line of electricity by the way of artificial, that is to say, that
After electrical power line inspector check of drawings finds defect, the relevant information of typing defect in folder name, filename or file.But by
In lacking unified description standard so that electrical power line inspector record transmission line of electricity the defects of during information just relatively arbitrarily, this just leads
Computer is caused to be difficult to reading and analyzing defect information automatically, and is not easy to carry out defect information unitized management.
Lack unified standard in conclusion the prior art has artificial description defect information and computer is caused to be difficult to certainly
Dynamic the problem of reading with analyzing defect information.
Invention content
In view of this, the object of the present invention is to provide a kind of processing method and processing device of defect information, to solve existing skill
There is manually description defect information and lack unified standard and computer is caused to be difficult to read automatically and analyzing defect information in art
Problem.
To achieve these goals, the present invention provides following technical solution：
A kind of processing method of defect information, including：
The artificial description defect information of each image in image set is read, and the artificial description defect information is decomposed into
Multiple keywords；
It obtains with describing the relevant keyword of benchmark word, and will be with describing the relevant keyword of benchmark word described in every class
All kinds of dictionaries are separately constituted, the defects of corresponding descriptor is utilized respectively and replaces the keyword included in the dictionary；
Default put in order according to benchmark word is described described in every class is ranked up for the corresponding dictionary, to be marked
Standardization defect description information.
Preferably, it is utilized respectively the defects of corresponding descriptor and replaces the keyword included in the dictionary, including：
It is utilized respectively in the defects of standardization defect dictionary corresponding with dictionary descriptor replacement dictionary and wraps
The keyword contained, the standardization defect dictionary are the standardization defect word formed using the synonym of the keyword
Allusion quotation.
Preferably, it is ranked up according to default put in order that benchmark word is described described in every class for the corresponding dictionary
Later, it further includes：
If there are it is identical the defects of descriptor, retain any one in the defects of identical descriptor.
Preferably, it is ranked up according to default put in order that benchmark word is described described in every class for the corresponding dictionary
Later, it further includes：
If the defects of benchmark word is related descriptor is described to described there are nothing in the dictionary, in the standardization defect
Description information corresponding position fills null value.
Preferably, it after obtaining standardization defect description information, further includes：
Corresponding data tree is built using the standardization defect description information of acquisition.
Preferably, it is described to build corresponding data tree using the standardization defect description information obtained, including：
The value of root node is set as null value, and counting corresponding with the root node is set as 0；
It is concentrated from described image and obtains an image and the standardization defect description information corresponding with described image, and will
Described image makees the first defect descriptor included in the corresponding standardization defect description information as present image
For current defect descriptor；
Using the root node as current node, and will corresponding with the current node count is incremented；
Judge whether the current defect descriptor is null value；
If the current defect descriptor is null value, using the present image as the leaf knot of the current node
Point, and return to described concentrated from described image of execution and obtain an image and the standardization defect description corresponding with described image
The step of information, until described image concentrates all images to complete structure；
If the current defect descriptor is not null value, judge to whether there is and institute in the child node of the current node
State the defects of current defect descriptor is identical descriptor；
If in the presence of, by the child node it is corresponding count is incremented, using the child node as current node, and will described in
The next defect descriptor included in standardization defect description information is performed as current defect descriptor described in the judgement
The step of whether current defect descriptor is null value, until last defect included in the standardization defect description information is retouched
Predicate completes structure, and it is described using the present image as the step of the leafy node of the current node to return to execution；
If being not present, increase by one comprising the current defect descriptor for the current node and corresponding be counted as 1
Child node, using the child node as current node, and perform and described will be included in the standardization defect description information
The step of next defect descriptor is as current defect descriptor.
Preferably, it before corresponding data tree is built using the standardization defect description information obtained, further includes：
Delete in the standardization defect description information segmental defect descriptor that includes and/or to the standardization defect
The defects of being included in description information descriptor reorders.
A kind of processing unit of defect information, including：
Read module is used for：Read the artificial description defect information of each image in image set, and by the artificial description
Defect information is decomposed into multiple keywords；
Acquisition module is used for：It obtains with describing the relevant keyword of benchmark word, and will be with describing benchmark word phase described in every class
The keyword closed separately constitutes all kinds of dictionaries, is utilized respectively the defects of corresponding descriptor and replaces the institute included in the dictionary
State keyword；
Sorting module is used for：According to benchmark word described described in every class it is default put in order for the corresponding dictionary into
Row sequence standardizes defect description information to obtain.
Preferably, the acquisition module includes：
Replacement unit is used for：The defects of standardization defect dictionary corresponding with dictionary descriptor is utilized respectively to replace
The keyword included in the dictionary is changed, the standardization defect dictionary is to be formed using the synonym of the keyword
Standardize defect dictionary.
Preferably, it further includes：
Reservation module is used for：It is putting in order as the corresponding dictionary according to describing the default of benchmark word described in every class
After being ranked up, if there are it is identical the defects of descriptor, retain any one in the defects of identical descriptor.
The present invention provides a kind of processing method and processing device of defect information, wherein this method includes：It reads in image set
The artificial description defect information of each image, and artificial description defect information is decomposed into multiple keywords；It obtains and description base
The quasi- relevant keyword of word, and the relevant keyword of benchmark word will be described with every class and separately constitutes all kinds of dictionaries, is utilized respectively pair
The defects of answering descriptor replaces the keyword included in dictionary；It is corresponding that default put in order of benchmark word is described according to every class
Dictionary is ranked up, and defect description information is standardized to obtain.
Above-mentioned technical proposal disclosed in the present application, reads the artificial description defect information of each image, and will manually describe
Defect information is decomposed into multiple keywords, then, obtains with describing the relevant keyword of benchmark word, will describe benchmark word with every class
Relevant keyword separately constitutes all kinds of dictionaries, and replaces the keyword in corresponding dictionary using defect descriptor, to reach rule
Generalized describes, and after replacing it, default put in order that benchmark word can be described according to every class is ranked up for corresponding dictionary, in this way
Standardization defect description information is obtained with, so as to carry out regulation and standardization to manually describing defect information, in order to lack
Fall into the automatic input of information and for statistical analysis to defect information.
Description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, 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 only this
The embodiment of invention, for those of ordinary skill in the art, without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is a kind of flow chart of the processing method of defect information provided in an embodiment of the present invention；
Fig. 2 is a kind of structure diagram of the processing unit of defect information provided in an embodiment of the present invention.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present invention, the technical solution in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other without making creative work
Embodiment shall fall within the protection scope of the present invention.
Fig. 1 is referred to, it illustrates a kind of flow charts of the processing method of defect information provided in an embodiment of the present invention, need
It is noted that the executive agent of the technical solution provided in the embodiment of the present invention can be provided for the corresponding embodiment of the present invention
A kind of defect information processing unit, and the device can be set in a computer.Therefore, it is provided in an embodiment of the present invention
The executive agent of above-mentioned technical proposal may be above computer, in the embodiment of the present invention using executive agent as computer into
Row explanation.The above method provided in an embodiment of the present invention can include：
S11：The artificial description defect information of each image in image set is read, and artificial description defect information is decomposed into
Multiple keywords.
After being taken pictures to the defects of transmission line of electricity and forming image set, electrical power line inspector can be to defeated in each image
The line defct that electric line occurs is registered, and obtains manually describing defect information, correspondence image IiArtificial description defect
Information is denoted as Si, i=1,2 ..., N, wherein, the corresponding artificial description defect information of each image can be a special text
Word is stated, or relevant information recorded in file name etc., after artificial description defect information is formed, and computer
Image I in image set can be readiArtificial description defect information Si, and will manually be retouched using participle technique or other modes
State defect information SiIt is decomposed into multiple keywords：Wherein,Point
The keyword obtained later Wei not decomposed.
S12：It obtains with describing the relevant keyword of benchmark word, and the relevant keyword difference of benchmark word will be described with every class
All kinds of dictionaries are formed, the defects of corresponding descriptor is utilized respectively and replaces the keyword included in dictionary.
After the keyword decomposed, computer can be deleted and description benchmark according to the description benchmark word of setting
The unrelated keyword of word, such as：Describe benchmark word often from the time (When), place (Where), personage (Who), object (What),
The aspect of defect (Wrong) five is described namely 5W modes, it is of course also possible to be 4W modes or other modes, delete with
After describing the unrelated keyword of benchmark word, the relevant keyword of benchmark word can be obtained and described：Wherein,Point
Corresponding keyword after Wei not deleting, at this point,And
After the keyword after being deleted, computer can will describe the relevant keyword difference of benchmark word with every class
Form all kinds of dictionaries, it is such as corresponding with foregoing description benchmark word, will be described with time class the relevant crucial phrase of benchmark word into
Dictionary forms time dictionary DA1；The relevant crucial phrase of benchmark word will be described with location category into place dictionary DA2... with such
It pushes away, obtains personage's dictionary DA3, object dictionary DA4, defect dictionary DA5, and there is no repetitions in obtained each dictionary
Keyword.
After corresponding dictionary is obtained, computer can be utilized respectively the defects of corresponding with the keyword included in dictionary
Descriptor replaces keyword included in dictionary, to form the description of standardization to keyword.
S13：Default put in order that benchmark word is described according to every class is ranked up for corresponding dictionary, to be standardized
Defect description information.
After the keyword during corresponding dictionary is replaced using defect descriptor, computer can be according to description benchmark word
Default put in order is ranked up for corresponding dictionary, standardizes defect description information to obtain, such as：Benchmark word is described according to the time
(When), after place (Where), personage (Who), object (What), defect (Wrong) are ranked up, correspondingly, computer
It will can in the order described above be arranged using the dictionary after the corresponding keyword of defect descriptor replacement, so as to be marked
Standardization defect description informationTo carry out regulation and standardization to manually describing defect information,
Consequently facilitating typing and statistical analysis are carried out to line defct information.
Above-mentioned technical proposal disclosed in the present application, reads the artificial description defect information of each image, and will manually describe
Defect information is decomposed into multiple keywords, then, obtains with describing the relevant keyword of benchmark word, will describe benchmark word with every class
Relevant keyword separately constitutes all kinds of dictionaries, and replaces the keyword in corresponding dictionary using defect descriptor, to reach rule
Generalized describes, and after replacing it, default put in order that benchmark word can be described according to every class is ranked up for corresponding dictionary, in this way
Standardization defect description information is obtained with, so as to carry out regulation and standardization to manually describing defect information, in order to lack
Fall into the automatic input of information and for statistical analysis to defect information.
A kind of processing method of defect information provided in an embodiment of the present invention is utilized respectively the defects of corresponding descriptor and replaces
The keyword included in dictionary can include：
It is utilized respectively the defects of standardization defect dictionary corresponding with dictionary descriptor and replaces the key included in dictionary
Word, standardization defect dictionary are the standardization defect dictionary formed using the synonym of keyword.
Be utilized respectively the defects of corresponding descriptor replace dictionary in include keyword when, it is used the defects of descriptor come
Derived from standardization defect dictionary DSk, k=1,2 ..., 5, wherein, standardization defect dictionary DSkIt is to utilize keywordSynonym
It forms, that is to say, that defect descriptor can be the synonym of all keywords, can correspondingly be denoted as：
Complete the reduce (cut (S after replacing iti)) then become：
It should be noted that the allocation rule of the corresponding synonym of keyword is：Under Same Scene, each keyword pair
The synonym answered is unique, and the synonym of each keyword can be its own, and multiple keywords can correspond to it is same
A synonym.Under different scenes, the synonym corresponding to keyword can be the same or different, and need according to specific feelings
Condition is set, and can not only ensure the standardization described to defect information, standardization in this way, can also be clever under different scenes
The defects of established standards living information.
The processing method of a kind of defect information provided in an embodiment of the present invention, in the default row that benchmark word is described according to every class
After row sequence is ranked up for corresponding dictionary, it can also include：
If there are it is identical the defects of descriptor, retain any one in the defects of identical descriptor.
Benchmark word described according to every class it is default put in order be ranked up for corresponding dictionary after, if there are identical
The defects of descriptor, then yojan can be carried out to it, yojan process can be：Each defect descriptor is calculated in SYNiIn go out
Existing number and the probability of appearance, if the number that occurs of existing defects descriptor is more than 1, retain it is therein any one
Defect descriptor describes benchmark word with corresponding every class in guarantee standardization defect description information and only exists the description of only one defect
Word namely in defect description information is standardized, one kind description benchmark word only corresponds to a defect descriptor, so that standard
Change defect description information and compare specification, standard.It is of course also possible to use other modes carry out yojan to dictionary, to allow standard
Change the corresponding benchmark word that described per class in defect description information and only exist a defect descriptor, these specific implementations are at this
Within the protection domain of invention.
The processing method of a kind of defect information provided in an embodiment of the present invention, in the default row that benchmark word is described according to every class
After row sequence is ranked up for corresponding dictionary, it can also include：
If there are in dictionary without to description benchmark word it is related the defects of descriptor, standardization defect description information correspond to
Fill null value in position.
According to every class describe benchmark word it is default put in order be ranked up for corresponding dictionary when, if there are dictionaries
The defects of middle nothing is related to description benchmark word descriptor can then fill empty on the corresponding position of standardization defect description information
Value, so that the dictionary included in defect information can be ranked up according to putting in order for benchmark word of description, so as to form specification
The description of change.
Such as putting in order for above-mentioned mentioned description benchmark word is corresponded to, when nothing is retouched with place in corresponding place dictionary
State the defects of benchmark word is related descriptor, and in corresponding personage's dictionary without it is related to description of person's benchmark word the defects of describe
It, then can be second in defect description information be standardized during word namely when corresponding place dictionary and personage's dictionary are empty
Null value is filled at the position of defect descriptor and third defect descriptor.
A kind of processing method of defect information provided in an embodiment of the present invention, after obtaining standardization defect description information,
It can also include：
Corresponding data tree is built using the standardization defect description information of acquisition.
After standardization defect description information is obtained, computer can utilize obtained standardization defect description information
Corresponding data tree is built, in order to form systematism and the storage mode of structuring, and just to standardization defect description information
In the lookup from corresponding data tree and analyze the relevant defect information of transmission line of electricity.
A kind of processing method of defect information provided in an embodiment of the present invention utilizes the standardization defect description information of acquisition
Corresponding data tree is built, can be included：
The value of root node is set as null value, and counting corresponding with root node is set as 0；
An image and standardization defect description information corresponding with image are obtained from image set, and using image as current
Image, using the first defect descriptor included in corresponding standardization defect description information as current defect descriptor；
Using root node as current node, and will corresponding with current node count is incremented；
Judge whether current defect descriptor is null value；
If current defect descriptor is null value, using present image as the leafy node of current node, and execution is returned to
A step of image and standardization defect description information corresponding with image are obtained from image set, until all figures in image set
As until completing structure；
If current defect descriptor is not null value, judges to whether there is in the child node of current node and be retouched with current defect
The defects of predicate is identical descriptor；
If in the presence of, by child node it is corresponding count is incremented, using child node as current node, and by standardization defect retouch
Next defect descriptor for being included in information is stated as current defect descriptor, execution judge current defect descriptor whether be
The step of null value, until last defect descriptor included in standardization defect description information completes structure, and return to execution
Using present image as the step of the leafy node of current node；
If being not present, for current node increase by one comprising current defect descriptor and it is corresponding be counted as 1 child node,
Using child node as current node, and perform and will standardize next defect descriptor for being included in defect description information as ought
The step of preceding defect descriptor.
It should be noted that the keyword that includes in dictionary is replaced using defect descriptor, and to defect descriptor into
Row yojan and after standardization defect description information corresponding position filling null value, obtained standardization defect description information
Middle correspondence describes benchmark word per class containing only there are one defect descriptor, and the defect descriptor may be specifically with defect phase
The information of pass, it is also possible to for null value, correspondingly, utilize the detailed process for standardizing the corresponding data tree of defect description information structure
Can be：
The value of root node is set as null value, i.e., the defects of root node corresponds to descriptor be null value, and will be with root node pair
The counting answered is set as 0, at this point, data tree, which only exists, is counted as 0 and root node of the value for sky；
An image I is obtained from image setiAnd standardization defect description information corresponding with image, for the ease of structure
It builds, then can be using the image got as present image, first will included in corresponding standardization defect description information
Defect descriptor is as current defect descriptor；
, then can be using the root node of data tree as current node for the ease of carrying out subsequent processing to node, it and will be with
Current node is corresponding, and count is incremented, the number occurred in order to statistical shortcomings；
At this point, computer may determine that whether current defect descriptor is null value；
If current defect descriptor is null value, using present image as the leafy node of current node, to represent to this
Picture construction is completed, and computer can return to execution and an image I is obtained from image set at this timeiAnd standard corresponding with image
The step of changing defect description information, to be built to next image, until all images are completed to be configured in image set
Only；
If current defect descriptor is not null value, judges to whether there is in the child node of current node and be retouched with current defect
The defects of predicate is identical descriptor；
If in the presence of, can by the child node identical with current defect descriptor it is corresponding count is incremented, with represent currently lack
The described defect information of descriptor is fallen into repeat, at this point it is possible to using the child node as current node, and standardization is lacked
The next defect descriptor included in description information is fallen into as current defect descriptor, execution judges that current defect descriptor is
No the step of being null value, until the last one defect descriptor for including completes structure, standard in standardization defect description information
Change after all defect descriptor that includes completes structure in defect description information, need to return and perform using present image as ought
The step of leafy node of preceding node.
If being not present, can be current node increase by one comprising current defect descriptor and it is corresponding be counted as 1 son
Node using the child node as current node, and performs and will standardize the next defect description included in defect description information
The step of word is as current defect descriptor finally completes structure, in order to from data tree to image all in image set
The relevant information of defect is obtained, and the number and other relevant informations of defect occurred convenient for statistical shortcomings.
In addition, the characteristics of constructed data tree can be known from the data tree of structure, is：(1) n omicronn-leaf child node
Value, represent the node all offsprings child node ... of child node (child node) this respect comprising the value meaning it is scarce
Fall into information；(2) node count represent to be provided simultaneously with the nodal value and the node all ancestors (father node, father node father
Node ...) nodal value the defects of information the defects of number；(3) root node is all in node count table diagram image set to lack to be empty
Sunken number；(4) leafy node represents corresponding image file, and there is no nodal values and node to count, in addition, for only counting
The situation of analyzing defect number, it may not be necessary to represent the leafy node of image file, that is, build when do not need to using image as
Leafy node is represented in data tree；(5) child node of n omicronn-leaf child node can with the mixing of right and wrong leafy node and leafy node,
It should be noted that mentioned nodal value is to be lacked included in above-mentioned mentioned standardization defect description information here
Fall into descriptor.
A kind of processing method of defect information provided in an embodiment of the present invention is describing letter using the standardization defect obtained
Before breath builds corresponding data tree, it can also include：
Delete in standardization defect description information the segmental defect descriptor that includes and/or to standardizing defect description information
In comprising the defects of descriptor reorder.
Before the defects of corresponding tree is built using the standardization defect description information obtained, can according to actual needs and
Yojan is carried out to standardization defect description information, that is to say, that delete the segmental defect included in standardization defect description information
Descriptor, and can also according to actual needs and in drawbacks of the standard description information comprising the defects of descriptor into rearrangement
Sequence, and the two can be carried out all, can also only carry out one of those according to actual needs, and the sequencing carried out
Can also be adjusted, so as to fulfill flexible analyzing defect information, and to standardization defect description information carry out yojan with
And after reordering, can corresponding data tree be built according to the standardization defect description information newly obtained.
The embodiment of the present invention additionally provides a kind of processing unit of defect information, refers to Fig. 2, and it illustrates of the invention real
A kind of structure diagram of the processing unit of defect information of example offer is provided, can be included：
Read module 11, is used for：The artificial description defect information of each image in image set is read, and will manually describe to lack
It is multiple keywords to fall into information decomposition；
Acquisition module 12, is used for：It obtains with describing the relevant keyword of benchmark word, and it is related to describe benchmark word to every class
Keyword separately constitute all kinds of dictionaries, be utilized respectively the defects of corresponding descriptor and replace the keyword that includes in dictionary；
Sorting module 13, is used for：Default put in order that benchmark word is described according to every class is ranked up for corresponding dictionary,
Defect description information is standardized to obtain.
A kind of processing unit of defect information provided in an embodiment of the present invention, acquisition module 12 can include：
Replacement unit is used for：It is utilized respectively the defects of standardization defect dictionary corresponding with dictionary descriptor substitute
The keyword included in allusion quotation, standardization defect dictionary are the standardization defect dictionary formed using the synonym of keyword.
A kind of processing unit of defect information provided in an embodiment of the present invention, can also include：
Reservation module is used for：It is ranked up in default put in order that benchmark word is described according to every class for corresponding dictionary
Later, if there are it is identical the defects of descriptor, retain any one in the defects of identical descriptor.
As a specific embodiment, technical scheme of the present invention can be illustrated, firstly, it is necessary to explanation, this
Defect is described in specific embodiment in a manner of 5W, and describes the default of benchmark word and puts in order as time (When), place
(Where), personage (Who), object (What), defect (Wrong), correspondingly, being included in standardization defect description information each
A dictionary is arranged in the order described above.
Image IiThe path of defect file filename.jpg storages is：In July, 2017/2017.7.24/ stockbridge dampers/rust
Erosion/filename1.jpg, and the filename in path has recorded defect present in filename.jpg, then correspondingly reads
The artificial description defect information got is SiIn July ,=2017,2017.7.24, stockbridge damper, corrosion, filename1；Segment it
Afterwards, the cut (S obtainediIn July, 2017)=[, 2017.7.24, stockbridge damper, corrosion, filename1], and at this timeYear July ..., and after deleting the keyword unrelated with describing benchmark word, the reduce (cut of gained
(Si))=[in July, 2017,2017.7.24, stockbridge damper, corrosion], at this point,July ... in year；Using lack
The keyword in the corresponding dictionary of descriptor replacement is fallen into, wherein, the corresponding defect descriptor of time dictionary is：It in July, 2017, obtains
To SYNi=[in July, 2017, in July, 2017, stockbridge damper, corrosion], after yojan, obtained standardization defect description information
For：f(IiIn July, 2017)=(,Stockbridge damper, corrosion).
If only comprising this image of filename.jpg in image set, the data tree built is：1) → (2017
July, 1) → filename1.jpg；According toAfter standardization defect description information is carried out yojan and is reordered,
The data tree of structure is：1) → (stockbridge damper, 1) → (in July, 2017,1) → (corrosion, 1) → filename1.jpg, is pressed
Data tree is understood according to different levels：(1) it only sees tree root, represents to include defect at 1 in image set；(2) tree root with
Represent that stockbridge damper has 1 defect in image set；(3) tree root withRepresent that stockbridge damper is sent out in July, 2017 in image set
1 defect is showed；(4) tree root withRepresent that stockbridge damper is found that corrosion defect in July, 2017 in image set
1 time, the type of defect can be obtained using tree the defects of structure, and for statistical analysis etc. to defect information.
The explanation of relevant portion refers to the present invention in a kind of processing unit of defect information provided in an embodiment of the present invention
The detailed description of corresponding part in the processing method of a kind of defect information that embodiment is provided, details are not described herein.In addition, this
The part consistent with corresponding to technical solution realization principle in the prior art be not in the above-mentioned technical proposal that inventive embodiments provide
It is described in detail, in order to avoid excessively repeat.
The foregoing description of the disclosed embodiments enables those skilled in the art to realize or use the present invention.To this
A variety of modifications of a little embodiments will be apparent for a person skilled in the art, and the general principles defined herein can
Without departing from the spirit or scope of the present invention, to realize in other embodiments.Therefore, the present invention will not be limited
The embodiments shown herein is formed on, and is to fit to consistent with the principles and novel features disclosed herein most wide
Range.
Claims (10)
1. a kind of processing method of defect information, which is characterized in that including：
The artificial description defect information of each image in image set is read, and the artificial description defect information is decomposed into multiple
Keyword；
It obtains with describing the relevant keyword of benchmark word, and will distinguish with describing the relevant keyword of benchmark word described in every class
All kinds of dictionaries are formed, the defects of corresponding descriptor is utilized respectively and replaces the keyword included in the dictionary；
Default put in order according to benchmark word is described described in every class is ranked up for the corresponding dictionary, to be standardized
Defect description information.
2. according to the method described in claim 1, it is characterized in that, being utilized respectively the defects of corresponding descriptor replaces the dictionary
In the keyword that includes, including：
It is utilized respectively the defects of standardization defect dictionary corresponding with dictionary descriptor and replaces what is included in the dictionary
The keyword, the standardization defect dictionary are the standardization defect dictionary formed using the synonym of the keyword.
3. according to the method described in claim 2, it is characterized in that, suitable according to the default arrangement that benchmark word is described described in every class
After sequence is ranked up for the corresponding dictionary, further include：
If there are it is identical the defects of descriptor, retain any one in the defects of identical descriptor.
4. according to the method described in claim 3, it is characterized in that, suitable according to the default arrangement that benchmark word is described described in every class
After sequence is ranked up for the corresponding dictionary, further include：
If there are, without the defects of benchmark word is related descriptor is described to described, described in the dictionary in the standardization defect
Information corresponding position fills null value.
5. according to the method described in claim 4, it is characterized in that, after obtaining standardization defect description information, further include：
Corresponding data tree is built using the standardization defect description information of acquisition.
6. according to the method described in claim 5, it is characterized in that, described utilize the standardization defect description information obtained
Corresponding data tree is built, including：
The value of root node is set as null value, and counting corresponding with the root node is set as 0；
It is concentrated from described image and obtains an image and the standardization defect description information corresponding with described image, and by described in
Image is as present image, using the first defect descriptor included in the corresponding standardization defect description information as ought
Preceding defect descriptor；
Using the root node as current node, and will corresponding with the current node count is incremented；
Judge whether the current defect descriptor is null value；
If the current defect descriptor is null value, using the present image as the leafy node of the current node, and
It returns to concentrate from described image described in performing and obtains an image and the standardization defect description information corresponding with described image
The step of, until described image concentrates all images to complete structure；
If the current defect descriptor is not null value, judges to whether there is in the child node of the current node and work as with described
The defects of preceding defect descriptor is identical descriptor；
If in the presence of, by the child node it is corresponding count is incremented, using the child node as current node, and by the standard
Change the next defect descriptor included in defect description information as current defect descriptor, it is described current to perform the judgement
The step of whether defect descriptor is null value, until last defect descriptor included in the standardization defect description information
Structure is completed, and it is described using the present image as the step of the leafy node of the current node to return to execution；
If being not present, for the current node increase by one comprising the current defect descriptor and it is corresponding be counted as 1 son
Node, using the child node as current node, and perform it is described will it is described standardization defect description information in include it is next
The step of a defect descriptor is as current defect descriptor.
7. according to the method described in claim 5, it is characterized in that, utilizing the standardization defect description information structure obtained
It builds before corresponding data tree, further includes：
It deletes the segmental defect descriptor included in the standardization defect description information and/or the standardization defect is described
The defects of being included in information descriptor reorders.
8. a kind of processing unit of defect information, which is characterized in that including：
Read module is used for：Read the artificial description defect information of each image in image set, and by the artificial description defect
Information decomposition is multiple keywords；
Acquisition module is used for：Obtain with the description relevant keyword of benchmark word, and will with to describe benchmark word described in every class relevant
The keyword separately constitutes all kinds of dictionaries, is utilized respectively the defects of corresponding descriptor and replaces the pass included in the dictionary
Keyword；
Sorting module is used for：Default put in order according to benchmark word is described described in every class is arranged for the corresponding dictionary
Sequence standardizes defect description information to obtain.
9. device according to claim 8, which is characterized in that the acquisition module includes：
Replacement unit is used for：It is utilized respectively the defects of standardization defect dictionary corresponding with dictionary descriptor and replaces institute
The keyword included in predicate allusion quotation, the standardization defect dictionary are the standard formed using the synonym of the keyword
Change defect dictionary.
10. device according to claim 9, which is characterized in that further include：
Reservation module is used for：It is carried out according to default put in order that benchmark word is described described in every class for the corresponding dictionary
After sequence, if there are it is identical the defects of descriptor, retain any one in the defects of identical descriptor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810039720.7A CN108268446B (en) | 2018-01-16 | 2018-01-16 | Defect information processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810039720.7A CN108268446B (en) | 2018-01-16 | 2018-01-16 | Defect information processing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108268446A true CN108268446A (en) | 2018-07-10 |
| CN108268446B CN108268446B (en) | 2021-06-01 |
Family
ID=62775643
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810039720.7A Active CN108268446B (en) | 2018-01-16 | 2018-01-16 | Defect information processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108268446B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070055522A1 (en) * | 2005-08-26 | 2007-03-08 | Sbc Knowledge Ventures, L.P. | Self-learning multi-source speech data reconstruction |
| CN101685455B (en) * | 2008-09-28 | 2012-02-01 | 华为技术有限公司 | Method and system of data retrieval |
| CN105426531A (en) * | 2015-12-18 | 2016-03-23 | 河南思维自动化设备股份有限公司 | Automatic parsing method of dispatching command |
| CN105677833A (en) * | 2016-01-06 | 2016-06-15 | 云南电网有限责任公司电力科学研究院 | Method for extracting circuit breaker fault characteristic information on basis of text mining technology |
-
2018
- 2018-01-16 CN CN201810039720.7A patent/CN108268446B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070055522A1 (en) * | 2005-08-26 | 2007-03-08 | Sbc Knowledge Ventures, L.P. | Self-learning multi-source speech data reconstruction |
| CN101685455B (en) * | 2008-09-28 | 2012-02-01 | 华为技术有限公司 | Method and system of data retrieval |
| CN105426531A (en) * | 2015-12-18 | 2016-03-23 | 河南思维自动化设备股份有限公司 | Automatic parsing method of dispatching command |
| CN105677833A (en) * | 2016-01-06 | 2016-06-15 | 云南电网有限责任公司电力科学研究院 | Method for extracting circuit breaker fault characteristic information on basis of text mining technology |
Non-Patent Citations (1)
| Title |
|---|
| 曹靖 等: "基于语义框架的电网缺陷文本挖掘技术及其应用", 《电网技术》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108268446B (en) | 2021-06-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110020437B (en) | Emotion analysis and visualization method combining video and barrage | |
| CN103838724B (en) | Image search method and device | |
| CN114896373B (en) | Image-text mutual inspection model training method and device, image-text mutual inspection method and equipment | |
| CN109816265B (en) | Knowledge characteristic mastery degree evaluation method, question recommendation method and electronic equipment | |
| CN110502664A (en) | Video tab indexes base establishing method, video tab generation method and device | |
| CN104298732B (en) | The personalized text sequence of network-oriented user a kind of and recommendation method | |
| CN111290956A (en) | Brain graph-based testing method and device, electronic equipment and storage medium | |
| CN109002483A (en) | Document management method, device, computer equipment and storage medium | |
| CN107608951A (en) | Report form generation method and system | |
| CN111680506A (en) | External key mapping method and device of database table, electronic equipment and storage medium | |
| CN110221960A (en) | Generate method and device, the computer equipment, storage medium of test case | |
| CN112667866A (en) | Test paper generation method and device, electronic equipment and storage medium | |
| JP2003131931A (en) | Method and device for scheduling, generating and processing document comprising blocks of information | |
| CN110688409A (en) | Learning state mining method, learning state recommendation method and electronic equipment | |
| CN111611781B (en) | Data labeling method, question answering device and electronic equipment | |
| CN104573113B (en) | A kind of information processing method and server | |
| CN108268446A (en) | A kind of processing method and processing device of defect information | |
| CN109509517A (en) | A kind of medical test Index for examination modified method automatically | |
| CN108491209A (en) | The extracting method and device of common code in a kind of html pages | |
| CN108921193A (en) | Picture input method, server and computer storage medium | |
| Liu et al. | A database for perceptual evaluation of image aesthetics | |
| CN112052157A (en) | Test message construction method, device and system | |
| CN107679174A (en) | Construction method, device and the server of Knowledge Organization System | |
| CN106227661A (en) | Data processing method and device | |
| CN105472405A (en) | Reminder generation 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 |