CN104517114A - Method and system for recognizing component feature - Google Patents
Method and system for recognizing component feature Download PDFInfo
- Publication number
- CN104517114A CN104517114A CN201410848530.1A CN201410848530A CN104517114A CN 104517114 A CN104517114 A CN 104517114A CN 201410848530 A CN201410848530 A CN 201410848530A CN 104517114 A CN104517114 A CN 104517114A
- Authority
- CN
- China
- Prior art keywords
- picture
- feature extraction
- database
- algorithm
- extraction algorithm
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
Abstract
The invention discloses a method for recognizing a component feature. The method comprises the following steps: acquiring a component picture to be recognized; marking the component picture as one having the component feature to be recognized; selecting a feature extraction algorithm with the highest recognition rate for the component feature from a database, wherein the database records the recognition rates of various feature extraction algorithms for the component feature; extracting an algorithm feature value of the component picture according to the feature extraction algorithm; recognizing whether the component picture has the component feature or not according to the algorithm feature value. Moreover, the invention further discloses a system for recognizing the component feature. Through the adoption of the method and the system provided by the invention, the feature extraction algorithm with the highest recognition rate for the component feature can be selected adaptively, so that effectiveness and accuracy are improved for recognition each time.
Description
Technical field
The present invention relates to field of image recognition, particularly relate to a kind of element characteristics recognition methods and system.
Background technology
Automatic visual inspection (AOI, Automated Optical Inspection) be the effective detection method of industrial automation, use machine vision as examination criteria technology, be widely used in LCD/TFT, transistor AND gate PCB industrial process, then may extend to safety system in people's livelihood purposes.Automatic visual inspection is representative gimmick common in industrial process, utilizes optical mode to obtain the surface state of finished product, detects the flaw such as foreign matter or pattern anomalies with image processing, because be contactless inspection, so can at middle engineering supervision semi-manufacture.
When board is in half-finished state, need to detect each element on board, identify whether the element on board is desired element, only have when element each on board is desired element, it is qualified that this board is only.Due to board production line having a lot of board, every block board there are again a lot of elements, too low by the words efficiency manually checked one by one.
Thus, prior art is the combination adopting AOI and feature extraction algorithm, part drawing picture is obtained by AOI, the characteristic of part drawing picture is extracted by feature extraction algorithm, and compare with the characteristic of the part drawing picture expected, if relatively more consistent, then described element is qualified, otherwise element is underproof.But because environmental factor (as illumination etc.) can have influence on identifying, the board identification of different batches adopts the effect of identical feature extraction algorithm different, and discrimination can be affected.
Summary of the invention
The object of the invention is to address the deficiencies of the prior art, propose a kind of element characteristics recognition methods and system, can choose a kind of algorithm the highest to described component recognition rate, strong adaptability, recognition effect is good.
To achieve the above object, the embodiment of the present invention provides a kind of element characteristics recognition methods, comprising:
Obtain an element picture to be identified; Described element picture is demarcated as having element characteristics to be identified;
A kind of feature extraction algorithm the highest to the discrimination of described element characteristics is chosen from database; Described database records multiple different feature extraction algorithm to the discrimination of described element characteristics;
The algorithm characteristics value of described element picture is extracted according to described feature extraction algorithm;
According to the identification of described algorithm characteristics value, whether element picture has described element characteristics.
Further, described element characteristics comprises element names and element dimensions feature;
Described database comprises component recognition database and dimensional characteristics identification database; Described component recognition database describes multiple different feature extraction algorithm to the discrimination of element picture with identity element name; Described element dimensions property data base describes multiple different feature extraction algorithm to the discrimination of element picture with identity element dimensional characteristics;
Describedly from database, choose a kind of feature extraction algorithm the highest to the discrimination of described element characteristics comprise:
Search described component recognition database, judge whether the element names that there is described element picture;
If so, then according to the element names of described element picture, from described component recognition database, the feature extraction algorithm the highest to described element names discrimination is chosen;
If not, then according to the dimensional characteristics of described element picture, from described dimensional characteristics identification database, choose the feature extraction algorithm the highest to described dimensional characteristics discrimination.
Database can be divided into component recognition database and dimensional characteristics identification database, can be the algorithm that the element picture learnt and the element picture do not learnt choose the highest discrimination respectively, can improve the validity of identification further.
Further, also comprise after according to the identification of described algorithm characteristics value, whether element picture has described element characteristics described:
Secondary identification is carried out to described element picture to be identified, determines whether described element picture to be identified has described element characteristics;
Verify whether described feature extraction algorithm identifies accurately according to described secondary identification;
According to the result correction, feature extraction algorithm described in database is to the discrimination of described element picture.
Further, before the element picture that described acquisition one is to be identified, also comprise the step of creation database, described creation database specifically comprises:
Obtain multiple sample components pictures; Described sample components picture has the element names and dimensional characteristics determined;
Multiple different feature extraction algorithm is adopted to extract the algorithm characteristics value of described sample components picture respectively;
Identify whether described sample components picture has described element names respectively according to described algorithm characteristics value;
When the described sample components picture identified has described element names, then described feature extraction algorithm identification success;
When the described sample components picture identified does not have described element names, then described feature extraction algorithm recognition failures;
Calculate described multiple different feature extraction algorithm respectively to the discrimination of described element names, and be documented in described component recognition database;
Identify whether described sample components picture has described dimensional characteristics respectively according to described algorithm characteristics value;
When the described sample components picture identified has described dimensional characteristics, then described feature extraction algorithm identification success;
When the described sample components picture identified does not have described dimensional characteristics, then described feature extraction algorithm recognition failures;
Calculate described multiple different feature extraction algorithm respectively to the discrimination of described dimensional characteristics, and be documented in described dimensional characteristics identification database.
Further, described dimensional characteristics comprises the average exposure degree of component type and element picture.
Correspondingly, the embodiment of the present invention also provides a kind of element characteristics recognition system, and can realize all flow processs of said elements characteristic recognition method, it comprises:
Picture acquisition module, for obtaining an element picture to be identified; Described element picture is demarcated as having element characteristics to be identified;
Algorithm picks module, for choosing a kind of feature extraction algorithm the highest to the discrimination of described element characteristics from database; Described database records multiple different feature extraction algorithm to the discrimination of described element characteristics;
Algorithm characteristics extraction module, for extracting the algorithm characteristics value of described element picture according to described feature extraction algorithm;
Whether identification module, have described element characteristics for element picture according to the identification of described algorithm characteristics value.
Further, described database comprises component recognition database and dimensional characteristics identification database; Described component recognition statistics of database different characteristic extraction algorithm is to the discrimination of element picture with identity element name; Described dimensional characteristics identification database has added up different characteristic extraction algorithm to the discrimination of element picture with same dimensional characteristics;
Described algorithm picks module comprises:
Searching judging unit, for searching described component recognition database, judging whether the element names that there is described element picture;
First algorithm picks unit, for when it is determined that the presence of the element names of described element picture, according to described element names, chooses the feature extraction algorithm the highest to described element names discrimination from described component recognition database;
Second algorithm picks unit, for when judging there is not the element names of described element picture, according to described dimensional characteristics, chooses the feature extraction algorithm the highest to described dimensional characteristics discrimination from described dimensional characteristics identification database.
Further, described element characteristics recognition system also comprises:
Secondary identification module, for carrying out secondary identification to described element picture to be identified, determines whether described element picture to be identified has described element characteristics;
Authentication module, for verifying according to described secondary identification whether described feature extraction algorithm identifies accurately;
Database repairs module, according to the result correction, feature extraction algorithm described in database is to the discrimination of described element picture.
Further, described element characteristics recognition system also comprises for creating described wide area information server creation module; Described database initialize module comprises:
Samples pictures acquiring unit, for obtaining multiple sample components pictures; Described sample components picture has the element names and dimensional characteristics determined;
Sample characteristics extraction unit, for the algorithm characteristics value adopting multiple different feature extraction algorithm to extract described sample components picture respectively;
According to described algorithm characteristics value, element names recognition unit, for identifying whether described sample components picture has described element names respectively; If so, described feature extraction algorithm identification success is then judged; If not, then described feature extraction algorithm recognition failures is judged;
First discrimination computing unit, for calculating described multiple different feature extraction algorithm respectively to the discrimination of described element names, and is documented in component recognition database;
According to described algorithm characteristics value, dimensional characteristics recognition unit, for identifying whether described sample components picture has described dimensional characteristics respectively; If so, described feature extraction algorithm identification success is then judged; If not, then described feature extraction algorithm recognition failures is judged;
Second discrimination computing unit, for calculating described multiple different feature extraction algorithm respectively to the discrimination of described dimensional characteristics, and is documented in described dimensional characteristics identification database.
Further, described dimensional characteristics comprises the average exposure degree of component type and element picture.
Implement the embodiment of the present invention, there is following beneficial effect: the element characteristics recognition methods that embodiments of the invention provide and system, may be used in board component recognition process, choose a kind of feature extraction algorithm of most high-efficiency according to the element characteristics of demarcating in advance, and carry out extraction algorithm eigenwert according to described feature extraction algorithm; Identify whether described element picture possesses described element characteristics and enhance adaptivity, improves recognition effect according to described algorithm characteristics value, for identifying that board element improves effective foundation.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of the first embodiment of element characteristics recognition methods provided by the invention;
Fig. 2 is the idiographic flow schematic diagram of the step S12 in Fig. 1;
Fig. 3 is the schematic flow sheet of the second embodiment of element characteristics recognition methods provided by the invention;
Fig. 4 is the schematic flow sheet of the 3rd embodiment of element characteristics recognition methods provided by the invention;
Fig. 5 is the form schematic diagram of component recognition rate database;
Fig. 6 is the form schematic diagram of dimensional characteristics discrimination database;
Fig. 7 is the element characteristics information table trrellis diagram once identified;
Fig. 8 is the element characteristics information table trrellis diagram of another identification;
Fig. 9 is the structured flowchart of the element characteristics recognition system that the embodiment of the present invention provides;
The structured flowchart of the algorithm picks module in Figure 10 Fig. 9;
Figure 11 is the structured flowchart of the database initialize module in Fig. 9.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.
See Fig. 1, be the schematic flow sheet of the first embodiment of element characteristics recognition methods provided by the invention, the method comprises the following steps:
S11, obtains an element picture to be identified; Described element picture is demarcated as having element characteristics to be identified;
S12, chooses a kind of feature extraction algorithm the highest to the discrimination of described element characteristics from database; Described database records multiple different feature extraction algorithm to the discrimination of described element characteristics;
S13, extracts the algorithm characteristics value of described element picture according to described feature extraction algorithm;
S14, according to the identification of described algorithm characteristics value, whether element picture has described element characteristics.
In S12, described feature extraction algorithm comprises color feature extracted algorithm, Texture Segmentation Algorithm and contour feature extraction algorithm etc., and these feature extraction algorithms are all the algorithms of prior art.
Particularly, described database comprises component recognition database and dimensional characteristics identification database.
Described component recognition database describes multiple different feature extraction algorithm to the discrimination of element picture with identity element name; Described component recognition database can the corresponding relation of recording feature extraction algorithm and element names in table form.
Described element dimensions property data base describes multiple different feature extraction algorithm to the discrimination of element picture with identity element dimensional characteristics; Described dimensional characteristics database can the corresponding relation of recording feature extraction algorithm and dimensional characteristics in table form.In embodiments of the present invention, described dimensional characteristics comprises the exposure of component type and element picture.
Refer to Fig. 2, Fig. 2 is the idiographic flow schematic diagram of the step S12 in Fig. 1.In step s 12, describedly from database, choose a kind of feature extraction algorithm the highest to the discrimination of described element characteristics comprise:
S121, searches described component recognition database, judges whether the element names that there is described element picture;
S122, if so, then according to the element names of described element picture, chooses the feature extraction algorithm the highest to described element names discrimination from described component recognition database;
S123, if not, then according to the dimensional characteristics of described element picture, chooses the feature extraction algorithm the highest to described dimensional characteristics discrimination from described dimensional characteristics identification database.
The element characteristics recognition methods that first embodiment of the invention provides can from the database be pre-created, choose a kind of algorithm the highest to described element picture discrimination according to the element characteristics that element picture to be identified is demarcated, each success ratio identified can be improved.Particularly, database divides into again component recognition database and dimensional characteristics database.Component recognition database is in order to when there being the data of known definite element picture to exist, and directly selects the algorithm of the highest discrimination according to described element picture; Dimensional characteristics database is in order to when the data pre-storage of not this element picture, according to each dimensional characteristics of element picture, can choose the algorithm of the highest discrimination, thus can possess good adaptability.
See Fig. 3, it is the schematic flow sheet of the second embodiment of element characteristics recognition methods provided by the invention, and the method comprises step S11 in the element characteristics recognition methods of first embodiment of the invention, S12, S13 and S14, in addition, also comprise step S15, S16 and S17, specific as follows:
S15, carries out secondary identification to described element picture to be identified, determines whether described element picture to be identified has described element characteristics;
According to described secondary identification, S16, verifies whether described feature extraction algorithm identifies accurately;
S17, according to the result correction, feature extraction algorithm described in database is to the discrimination of described element picture.
The element characteristics recognition methods that second embodiment of the invention provides is the further improvement in the technical scheme of above-mentioned first embodiment, the element characteristics that method for distinguishing determines element picture is known by secondary, thus whether the feature extraction algorithm selected by judging correctly identifies described element picture, and the discrimination of the described database record of amendment further, perfect database, improves the correctness of algorithm picks.
See Fig. 4, it is the schematic flow sheet of the 3rd embodiment of element characteristics recognition methods provided by the invention, the method comprises step S11 in the element characteristics recognition methods of second embodiment of the invention, S12, S13, S14, S15, S16 and S17, in addition, also comprise step S21-S26, specific as follows:
S21, obtains multiple sample components pictures; Described sample components picture has the element names and dimensional characteristics determined;
S22, adopts multiple different feature extraction algorithm to extract the algorithm characteristics value of described sample components picture respectively;
According to described algorithm characteristics value, S23, identifies whether described sample components picture has described element names respectively; When the described sample components picture identified has described element names, then described feature extraction algorithm identification success; When the described sample components picture identified does not have described element names, then described feature extraction algorithm recognition failures;
S24, calculates described multiple different feature extraction algorithm respectively to the discrimination of described element names, and is documented in component recognition database;
According to described algorithm characteristics value, S25, identifies whether described sample components picture has described dimensional characteristics respectively; When the described sample components picture identified has described dimensional characteristics, then described feature extraction algorithm identification success; When the described sample components picture identified does not have described dimensional characteristics, then described feature extraction algorithm recognition failures;
S26, calculates described multiple different feature extraction algorithm respectively to the discrimination of described dimensional characteristics, and is documented in described dimensional characteristics identification database.
The element characteristics recognition methods that third embodiment of the invention provides further illustrates the constructive process of described database, is trained by a large amount of sample components pictures, sets up the database with certain validity.It should be noted that, step S21-S26 is the step before implementation step S11-S17.
As shown in figs. 5 and 6, Fig. 5 is the tabular drawing of component recognition rate database, and Fig. 6 is the tabular drawing of dimensional characteristics discrimination database.Process for the purpose of simplifying the description, only list feature extraction algorithm A, feature extraction algorithm B and feature extraction algorithm C tri-kinds of feature extraction algorithms in algorithm in Fig. 5, in the element names in Fig. 6, specifically list ceramic condenser, 32 pin P I N and voltage stabilizing diode three kinds of concrete elements.Only list feature extraction algorithm A, feature extraction algorithm B and feature extraction algorithm C tri-kinds of feature extraction algorithms in algorithm in Fig. 6, dimensional characteristics comprises component type and average exposure; Component type lists P I N class, capacitance kind and diode-like; Average exposure degree is divided into 0-30,31-70,71-100 tri-intervals.With twice identifying, the course of work of the present invention is described below.
(1) in an identifying, an element picture to be identified is obtained.Assuming that described element picture by the element characteristics of demarcating in advance as shown in Figure 7, Fig. 7 is the element characteristics information table trrellis diagram once identified.
Search the form in Fig. 5, the data of ceramic condenser can be there are in decision element identification database.
Choose the feature extraction algorithm the highest to the discrimination of ceramic condenser.From the form of Fig. 5, the discrimination of feature extraction algorithm A to ceramic condenser is the highest, reaches 80%, thus selected characteristic extraction algorithm A.
The algorithm characteristics value of described element picture is extracted according to described feature extraction algorithm A.
According to the identification of described algorithm characteristics value, whether element picture is the picture of ceramic condenser.
Only identifying described element picture is the picture of ceramic condenser, and it is qualified that described element is only.
(2) in another identifying, the element picture that another is to be identified is obtained.Assuming that described element picture by the element characteristics of demarcating in advance as shown in Figure 8, Fig. 8 is the element characteristics information table trrellis diagram of another identification.
Search the form in Fig. 5, can there are not the data of light emitting diode in decision element identification database.
Search the form in Fig. 6, possess diode-like and average exposure degree is 65 two dimensional characteristics due to light emitting diode simultaneously, then comprehensive two dimensional characteristics choose the feature extraction algorithm of the highest discrimination.For diode-like, the discrimination of feature extraction algorithm A is the highest, reaches 60%; Be 65 for average exposure degree, the discrimination of feature extraction algorithm A is also the highest, thus selected characteristic extraction algorithm A.If (for average exposure degree, what discrimination was the highest is not feature extraction algorithm A but feature extraction algorithm B, the discrimination of diode-like and feature extraction algorithm B identify that average exposure degree interval is the discrimination of 31-70, the algorithm that therefrom selective recognition rate is higher then to need further comparative feature extraction algorithm A to identify).
The algorithm characteristics value of described element picture is extracted according to described feature extraction algorithm A.
According to the identification of described algorithm characteristics value, whether element picture is the picture of light emitting diode.
Only identifying described element picture is the picture of light emitting diode, and it is qualified that described element is only.
Correspondingly, the embodiment of the present invention also provides a kind of element characteristics recognition system, and can realize all flow processs of said elements characteristic recognition method, it comprises:
Picture acquisition module 1, for obtaining an element picture to be identified; Described element picture is demarcated as having element characteristics to be identified;
Algorithm picks module 2, for choosing a kind of feature extraction algorithm the highest to the discrimination of described element characteristics from database; Described database records multiple different feature extraction algorithm to the discrimination of described element characteristics;
Algorithm characteristics extraction module 3, for extracting the algorithm characteristics value of described element picture according to described feature extraction algorithm;
Whether identification module 4, have described element characteristics for element picture according to the identification of described algorithm characteristics value.
Particularly, described database comprises component recognition database and dimensional characteristics identification database; Described component recognition statistics of database different characteristic extraction algorithm is to the discrimination of element picture with identity element name; Described dimensional characteristics identification database has added up different characteristic extraction algorithm to the discrimination of element picture with same dimensional characteristics;
Described algorithm picks module 2 comprises:
Searching judging unit 21, for searching described component recognition database, judging whether the element names that there is described element picture;
First algorithm picks unit 22, for when it is determined that the presence of the element names of described element picture, according to described element names, chooses the feature extraction algorithm the highest to described element names discrimination from described component recognition database;
Second algorithm picks unit 23, for when judging there is not the element names of described element picture, according to described dimensional characteristics, chooses the feature extraction algorithm the highest to described dimensional characteristics discrimination from described dimensional characteristics identification database.
Further, described element characteristics recognition system also comprises:
Secondary identification module 5, for carrying out secondary identification to described element picture to be identified, determines whether described element picture to be identified has described element characteristics;
Authentication module 6, for verifying according to described secondary identification whether described feature extraction algorithm identifies accurately;
Database repairs module 7, according to the result correction, feature extraction algorithm described in database is to the discrimination of described element picture.
Further, described element characteristics recognition system also comprises for creating described wide area information server creation module 8; Described database initialize module 8 comprises:
Samples pictures acquiring unit 81, for obtaining multiple sample components pictures; Described sample components picture has the element names and dimensional characteristics determined;
Sample characteristics extraction unit 82, for the algorithm characteristics value adopting multiple different feature extraction algorithm to extract described sample components picture respectively;
According to described algorithm characteristics value, element names recognition unit 83, for identifying whether described sample components picture has described element names respectively; If so, described feature extraction algorithm identification success is then judged; If not, then described feature extraction algorithm recognition failures is judged;
First discrimination computing unit 84, for calculating described multiple different feature extraction algorithm respectively to the discrimination of described element names, and is documented in component recognition database;
According to described algorithm characteristics value, dimensional characteristics recognition unit 85, for identifying whether described sample components picture has described dimensional characteristics respectively; If so, described feature extraction algorithm identification success is then judged; If not, then described feature extraction algorithm recognition failures is judged;
Second discrimination computing unit 86, for calculating described multiple different feature extraction algorithm respectively to the discrimination of described dimensional characteristics, and is documented in described dimensional characteristics identification database.
The element characteristics recognition system that the embodiment of the present invention provides can realize all flow processs of said elements characteristic recognition method, each module component and method step one_to_one corresponding, thus no longer repeats the principle of the element characteristics recognition system of the present embodiment.
Implement the embodiment of the present invention, there is following beneficial effect: the element characteristics recognition methods that embodiments of the invention provide, may be used in board component recognition process, choose a kind of feature extraction algorithm of most high-efficiency according to the element characteristics of demarcating in advance, and carry out extraction algorithm eigenwert according to described feature extraction algorithm; Identify whether described element picture possesses described element characteristics according to described algorithm characteristics value.Particularly, database can be divided into component recognition database and dimensional characteristics identification database, can be the algorithm that the element picture learnt and the element picture do not learnt choose the highest discrimination respectively, enhance adaptivity, improve recognition effect, for identifying that board element improves effective foundation.Embodiments of the invention are also corresponding provides a kind of element characteristics recognition system, can reach above-mentioned identical effect.
One of ordinary skill in the art will appreciate that all or part of flow process realized in above-described embodiment method, that the hardware that can carry out instruction relevant by computer program has come, described program can be stored in a computer read/write memory medium, this program, when performing, can comprise the flow process of the embodiment as above-mentioned each side method.Wherein, described storage medium can be magnetic disc, CD, read-only store-memory body (Read-Only Memory, ROM) or random store-memory body (Random Access Memory, RAM) etc.
The above is the preferred embodiment of the present invention; it should be pointed out that for those skilled in the art, under the premise without departing from the principles of the invention; can also make some improvements and modifications, these improvements and modifications are also considered as protection scope of the present invention.
Claims (10)
1. an element characteristics recognition methods, is characterized in that, comprising:
Obtain an element picture to be identified; Described element picture is demarcated as having element characteristics to be identified;
A kind of feature extraction algorithm the highest to the discrimination of described element characteristics is chosen from database; Described database records multiple different feature extraction algorithm to the discrimination of described element characteristics;
The algorithm characteristics value of described element picture is extracted according to described feature extraction algorithm;
According to the identification of described algorithm characteristics value, whether element picture has described element characteristics.
2. element characteristics recognition methods as claimed in claim 1, is characterized in that,
Described element characteristics comprises element names and element dimensions feature;
Described database comprises component recognition database and dimensional characteristics identification database; Described component recognition database describes multiple different feature extraction algorithm to the discrimination of element picture with identity element name; Described element dimensions property data base describes multiple different feature extraction algorithm to the discrimination of element picture with identity element dimensional characteristics;
Describedly from database, choose a kind of feature extraction algorithm the highest to the discrimination of described element characteristics comprise:
Search described component recognition database, judge whether the element names that there is described element picture;
If so, then according to the element names of described element picture, from described component recognition database, the feature extraction algorithm the highest to described element names discrimination is chosen;
If not, then according to the dimensional characteristics of described element picture, from described dimensional characteristics identification database, choose the feature extraction algorithm the highest to described dimensional characteristics discrimination.
3. element characteristics recognition methods as claimed in claim 1, is characterized in that, also comprises after according to the identification of described algorithm characteristics value, whether element picture has described element characteristics described:
Secondary identification is carried out to described element picture to be identified, determines whether described element picture to be identified has described element characteristics;
Verify whether described feature extraction algorithm identifies accurately according to described secondary identification;
According to the result correction, feature extraction algorithm described in database is to the discrimination of described element picture.
4. element characteristics recognition methods as claimed in claim 2 or claim 3, it is characterized in that, before the element picture that described acquisition one is to be identified, also comprise the step of creation database, described creation database specifically comprises:
Obtain multiple sample components pictures; Described sample components picture has the element names and dimensional characteristics determined;
Multiple different feature extraction algorithm is adopted to extract the algorithm characteristics value of described sample components picture respectively;
Identify whether described sample components picture has described element names respectively according to described algorithm characteristics value;
When the described sample components picture identified has described element names, then described feature extraction algorithm identification success;
When the described sample components picture identified does not have described element names, then described feature extraction algorithm recognition failures;
Calculate described multiple different feature extraction algorithm respectively to the discrimination of described element names, and be documented in described component recognition database;
Identify whether described sample components picture has described dimensional characteristics respectively according to described algorithm characteristics value;
When the described sample components picture identified has described dimensional characteristics, then described feature extraction algorithm identification success;
When the described sample components picture identified does not have described dimensional characteristics, then described feature extraction algorithm recognition failures;
Calculate described multiple different feature extraction algorithm respectively to the discrimination of described dimensional characteristics, and be documented in described dimensional characteristics identification database.
5. element characteristics recognition methods as claimed in claim 1, it is characterized in that, described dimensional characteristics comprises the average exposure degree of component type and element picture.
6. an element characteristics recognition system, is characterized in that, comprising:
Picture acquisition module, for obtaining an element picture to be identified; Described element picture is demarcated as having element characteristics to be identified;
Algorithm picks module, for choosing a kind of feature extraction algorithm the highest to the discrimination of described element characteristics from database; Described database records multiple different feature extraction algorithm to the discrimination of described element characteristics;
Algorithm characteristics extraction module, for extracting the algorithm characteristics value of described element picture according to described feature extraction algorithm;
Whether identification module, have described element characteristics for element picture according to the identification of described algorithm characteristics value.
7. element characteristics recognition system as claimed in claim 6, is characterized in that,
Described database comprises component recognition database and dimensional characteristics identification database; Described component recognition statistics of database different characteristic extraction algorithm is to the discrimination of element picture with identity element name; Described dimensional characteristics identification database has added up different characteristic extraction algorithm to the discrimination of element picture with same dimensional characteristics;
Described algorithm picks module comprises:
Searching judging unit, for searching described component recognition database, judging whether the element names that there is described element picture;
First algorithm picks unit, for when it is determined that the presence of the element names of described element picture, according to described element names, chooses the feature extraction algorithm the highest to described element names discrimination from described component recognition database;
Second algorithm picks unit, for when judging there is not the element names of described element picture, according to described dimensional characteristics, chooses the feature extraction algorithm the highest to described dimensional characteristics discrimination from described dimensional characteristics identification database.
8. element characteristics recognition system as claimed in claim 6, it is characterized in that, described element characteristics recognition system also comprises:
Secondary identification module, for carrying out secondary identification to described element picture to be identified, determines whether described element picture to be identified has described element characteristics;
Authentication module, for verifying according to described secondary identification whether described feature extraction algorithm identifies accurately;
Database repairs module, according to the result correction, feature extraction algorithm described in database is to the discrimination of described element picture.
9. element characteristics recognition system as claimed in claim 7 or 8, is characterized in that,
Described element characteristics recognition system also comprises for creating described wide area information server creation module; Described database initialize module comprises:
Samples pictures acquiring unit, for obtaining multiple sample components pictures; Described sample components picture has the element names and dimensional characteristics determined;
Sample characteristics extraction unit, for the algorithm characteristics value adopting multiple different feature extraction algorithm to extract described sample components picture respectively;
According to described algorithm characteristics value, element names recognition unit, for identifying whether described sample components picture has described element names respectively; If so, described feature extraction algorithm identification success is then judged; If not, then described feature extraction algorithm recognition failures is judged;
First discrimination computing unit, for calculating described multiple different feature extraction algorithm respectively to the discrimination of described element names, and is documented in component recognition database;
According to described algorithm characteristics value, dimensional characteristics recognition unit, for identifying whether described sample components picture has described dimensional characteristics respectively; If so, described feature extraction algorithm identification success is then judged; If not, then described feature extraction algorithm recognition failures is judged;
Second discrimination computing unit, for calculating described multiple different feature extraction algorithm respectively to the discrimination of described dimensional characteristics, and is documented in described dimensional characteristics identification database.
10. element characteristics recognition system as claimed in claim 6, it is characterized in that, described dimensional characteristics comprises the average exposure degree of component type and element picture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410848530.1A CN104517114B (en) | 2014-12-30 | 2014-12-30 | A kind of element characteristics recognition methods and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410848530.1A CN104517114B (en) | 2014-12-30 | 2014-12-30 | A kind of element characteristics recognition methods and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104517114A true CN104517114A (en) | 2015-04-15 |
| CN104517114B CN104517114B (en) | 2018-01-30 |
Family
ID=52792382
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410848530.1A Active CN104517114B (en) | 2014-12-30 | 2014-12-30 | A kind of element characteristics recognition methods and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104517114B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106021677A (en) * | 2016-05-12 | 2016-10-12 | 北京航空航天大学 | Component surface feature extraction method and device |
| CN108090473A (en) * | 2018-01-12 | 2018-05-29 | 北京陌上花科技有限公司 | The method and device of polyphaser human face identification |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101178773A (en) * | 2007-12-13 | 2008-05-14 | 北京中星微电子有限公司 | Image recognition system and method based on characteristic extracting and categorizer |
| CN101286163A (en) * | 2008-05-15 | 2008-10-15 | 深圳先进技术研究院 | Recognition method based on recognition knowledge base |
| US20080298644A1 (en) * | 2007-05-29 | 2008-12-04 | S1 Corporation | System and method for controlling image quality |
-
2014
- 2014-12-30 CN CN201410848530.1A patent/CN104517114B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080298644A1 (en) * | 2007-05-29 | 2008-12-04 | S1 Corporation | System and method for controlling image quality |
| CN101178773A (en) * | 2007-12-13 | 2008-05-14 | 北京中星微电子有限公司 | Image recognition system and method based on characteristic extracting and categorizer |
| CN101286163A (en) * | 2008-05-15 | 2008-10-15 | 深圳先进技术研究院 | Recognition method based on recognition knowledge base |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106021677A (en) * | 2016-05-12 | 2016-10-12 | 北京航空航天大学 | Component surface feature extraction method and device |
| CN108090473A (en) * | 2018-01-12 | 2018-05-29 | 北京陌上花科技有限公司 | The method and device of polyphaser human face identification |
| CN108090473B (en) * | 2018-01-12 | 2021-04-30 | 北京陌上花科技有限公司 | Method and device for recognizing human face under multiple cameras |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104517114B (en) | 2018-01-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108920580B (en) | Image matching method, device, storage medium and terminal | |
| WO2017032311A1 (en) | Detection method and apparatus | |
| CN101915769B (en) | Automatic optical inspection method for printed circuit board comprising resistance element | |
| CN101576956B (en) | On-line character detection method based on machine vision and system thereof | |
| CN113239930B (en) | Glass paper defect identification method, system, device and storage medium | |
| CN105894036A (en) | Image feature template matching method being applied to detection of mobile phone screen defects | |
| CN108827969A (en) | Metal parts surface defects detection and recognition methods and device | |
| CN107016394B (en) | Cross fiber feature point matching method | |
| CN106501272A (en) | Machine vision scolding tin position detecting system | |
| CN113222913B (en) | Circuit board defect detection positioning method, device and storage medium | |
| CN104459421A (en) | Diode polarity detecting method and system | |
| WO2017107533A1 (en) | Electronic component sample labeling method and device | |
| CN110047063A (en) | A kind of detection method that material is fallen, device, equipment and storage medium | |
| CN104200215A (en) | Method for identifying dust and pocking marks on surface of big-caliber optical element | |
| CN116188475A (en) | Intelligent control method, system and medium for automatic optical detection of appearance defects | |
| CN109086781A (en) | A kind of cabinet lamp state identification method based on deep learning | |
| CN111159150A (en) | Data expansion method and device | |
| CN102799861A (en) | Method for rapidly identifying reading of instrument by using color | |
| CN104517114A (en) | Method and system for recognizing component feature | |
| CN114429445A (en) | PCB defect detection and identification method based on MAIRNet | |
| Ramisa et al. | Evaluation of the sift object recognition method in mobile robots | |
| JP4814116B2 (en) | Mounting board appearance inspection method | |
| Pramerdorfer et al. | PCB recognition using local features for recycling purposes | |
| CN206470210U (en) | Machine vision scolding tin position detecting system | |
| CN114235815A (en) | Method for detecting surface defects of outdoor electrical equipment of converter station based on scene filtering |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |