CN103529041B - Based on circuit board newness degree decision method and the system of characteristics of image - Google Patents
Based on circuit board newness degree decision method and the system of characteristics of image Download PDFInfo
- Publication number
- CN103529041B CN103529041B CN201310533952.5A CN201310533952A CN103529041B CN 103529041 B CN103529041 B CN 103529041B CN 201310533952 A CN201310533952 A CN 201310533952A CN 103529041 B CN103529041 B CN 103529041B
- Authority
- CN
- China
- Prior art keywords
- feature
- circuit board
- image
- new
- old
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a kind of circuit board newness degree decision method based on characteristics of image and system, belong to technical field of circuit board detection.This decision method comprises the following steps: the image of collecting circuit board, extracts the feature in image, and this feature is classified; And contrast with the standard feature in the statistical nature database set up in advance, determine its new and old coefficient; Utilize new and old coefficient and the weight coefficient of described feature, draw new and old coefficient of determination; This new and old coefficient of determination and preset value are compared, draws newness degree result of determination.Utilize decision method of the present invention and system, by the statistic of classification of image feature amount, set up the new and old synthetic determination model of circuit board components fast, can realizing circuit plate newness degree automatic, convenient, quick, accurately check, can be widely used in the testing of circuit board components.
Description
Technical field
The present invention relates to a kind of method that circuit board is detected, particularly relate to a kind of circuit board newness degree decision method based on characteristics of image and system.
Background technology
Along with China's foreign trade is fast-developing, the old electronic product of import all increases by a fairly big margin every year.As a part for industrial foundation equipment, the old electronic product of import different times be cost-saving, speed up the construction, promote the development of China industrial technology and play certain promotion.But the renewal of raising and planning standard day by day relative to product safety, health, environmental requirement, the situation that the mandatory technical manual that the old electronic product of import often exists existing related products safety, health, environmental protection with China is not inconsistent, more there is greater risk and hidden danger in a lot of product.
Discover and seize time in recent years in Check and Examination of Port some illegal enterprises with " with old fill new " or illegal means such as " new and old load in mixture ", escape inspection and quarantine supervision, the old electronic product immigration not meeting safety, health, environmental requirement.Due to the old electronic product of part through renovation or Cheng Xindu very high, simple rely on artificial visual testing efficiency compared with low, fault rate large, and automatically differentiates without effective technology means at present, bring larger difficulty to inspection work, affect assay authority.If too much defective old electronic product immigration, by the security of the lives and property of serious threat to the country and people.
The image that the optical detection of current circuit board mainly relies on the reflected light of analyzing printed circuit board (PCB) or transmitted light to be formed is to detect the quality of printed circuit board (PCB), thus judging that it is new and old, this method not yet can comprehensively judge the new and old of circuit board by many factors at present.
Summary of the invention
Based on this, the object of the invention is to the defect overcoming prior art, provide a kind of circuit board newness degree decision method based on characteristics of image, this decision method from many aspects factor sets out, and considers, and can obtain result of determination more accurately.
For achieving the above object, the present invention takes following technical scheme:
Based on a circuit board newness degree decision method for characteristics of image, comprise the following steps:
The image of collecting circuit board, extract the feature in image, and the feature in image is classified according to the ink color of circuit board, silk-screen layer printed words, defect and electrical node oxidization condition, be divided into ink color feature, silk-screen layer printed words feature, defect characteristic and electrical node oxidation characteristic;
The standard feature stored in above-mentioned ink color feature, silk-screen layer printed words feature, defect characteristic and electrical node oxidation characteristic and the statistical nature database to set up in advance is contrasted, determines the new and old coefficient corresponding with each feature;
Utilize the new and old coefficient of above-mentioned feature, and the predetermined weight coefficient corresponding with this feature, draw new and old coefficient of determination;
This new and old coefficient of determination and preset value are compared, draws newness degree result of determination.
Decision method of the present invention, provide a kind of from iconology angle, to the new and old method judged of circuit board, the present invention is by a large amount of research, after comprehensive many-sided factor is considered, have selected circuit board ink color, circuit board silk-screen layer printed words, circuit board defect, the features such as electrical node oxidation, as evaluation measures, and in order to reflect circuit board truth more accurately, also according to the importance of different characteristic to which imparts different weight coefficients, new and old coefficient eventually through each feature considers with its importance, draw the result of determination of circuit board newness degree.
Wherein in an embodiment, the described statistical nature storehouse set up in advance is set up by the following method:
Gather the image of Different years circuit board in advance, extract the feature in image, as standard feature, set up the statistical nature database of the corresponding different new and old coefficient of various criterion feature.As required, select the circuit board of the Different years (as 0.125,0.25,0.5,1,2,3,5,7,10,10 with first-class) be applicable to, gather its each feature, and the feature in image is classified according to the ink color of circuit board, silk-screen layer printed words, defect and electrical node oxidization condition, be divided into ink color feature, silk-screen layer printed words feature, defect characteristic and electrical node oxidation characteristic, then according to the physical life of circuit board, for each standard feature defines new and old coefficient.With by simulating compared with the characteristics of image that defines, using the characteristics of image of real circuits plate as reference, the newness degree of circuit board can be reflected more accurately.
Wherein in an embodiment, after collecting the image of circuit board, first through image denoising, then extract the feature in image.Adopt conventional Image Denoising Technology, remove various disturbing factor, ensureing to extract the characteristics of image that obtains can the truth of accurate reproduction circuit board.
Wherein in an embodiment, described ink color feature comprises green, red, black, blue subcharacter; Described silk-screen layer printed words feature comprises the subcharacter of wearing and tearing, color; Described defect characteristic comprises the subcharacter of defect, burn into spot; Described electrical node oxidation characteristic comprises the subcharacter of pad, via hole.According to the situation of specific boards, dissimilar feature is segmented, can make a determination from the more comprehensive newness degree of angle to circuit board.
Wherein in an embodiment, the influence degree that described weight coefficient judges circuit board newness degree according to dissimilar feature is established, and is added by the weight coefficient of all types of feature and value is set to 1.By such design, can allow some important features in decision process, play even more important effect, improve the accuracy judged.
Wherein in an embodiment, as some feature do not exist in the circuit board that detects, then corresponding weight coefficient gets 0, and presses weight proportion by its weight allocation in other features.Improve the accuracy judged.
Wherein in an embodiment, after extracting the feature in image, first judge whether to there is significant deficiency, if exist, be then directly judged to be old circuit board or defective circuit board; If do not exist, then carry out subsequent step.If there is significant deficiency, then without the need to carrying out subsequent step, can determination flow be simplified, improving and judging efficiency.Described significant deficiency is: circuit board fractures, components and parts come off.Also can according to different real needs, setting different defects is significant deficiency.
Wherein in an embodiment, the computing method of described new and old coefficient of determination are, are multiplied by the new and old coefficient of all types of feature, then sum up, obtain new and old coefficient of determination with its weight coefficient.The computing method of this new and old coefficient of determination had both considered the impact of circuit board each side on newness degree, considered again the importance that each feature affects newness degree, reflected the newness degree of circuit board comprehensively and objectively.
The present invention also provides a kind of decision-making system of circuit board newness degree, adopt above-mentioned decision method, comprise the image capture module for gathering image, for extracting characteristics of image and carrying out the image analysis module that contrasts and the result determination module for calculating new and old coefficient of determination with statistical nature database Plays feature, described image capture module, image analysis module and result determination module are electrically connected successively.
The decision-making system of circuit board newness degree of the present invention, for above-mentioned decision method provides hardware supported.
Compared with prior art, the present invention has following beneficial effect:
A kind of circuit board newness degree decision method based on characteristics of image of the present invention and system, utilize this decision-making system, by image feature amount statistic of classification, set up the new and old synthetic determination model of circuit board components fast, can realizing circuit plate newness degree automatic, convenient, quick, accurately check, improve detection efficiency, reduce artificial judgement difficulty and error, practical, can be widely used in the testing of circuit board components, particularly be applicable to national each import bank in the detection of import electromechanics.
Accompanying drawing explanation
Fig. 1 is circuit board newness degree decision method process flow diagram in the embodiment of the present invention;
Fig. 2 is the structural representation of circuit board newness degree decision-making system in the embodiment of the present invention;
Fig. 3 is the program flow diagram of circuit board newness degree decision method in the embodiment of the present invention;
Fig. 4 is image subcharacter Classified Statistic Model figure in circuit board newness degree decision method in the embodiment of the present invention.
Embodiment
The present invention is described in detail below in conjunction with the drawings and specific embodiments.
Based on a circuit board newness degree decision method for characteristics of image, as shown in Figure 1, comprise the following steps:
Step 10: the image of collecting circuit board, extract the feature in image, and the feature in image is classified according to the ink color of circuit board, silk-screen layer printed words, defect and electrical node oxidization condition, be divided into ink color feature, silk-screen layer printed words feature, defect characteristic and electrical node oxidation characteristic;
Above-mentioned steps is specially: tested import electromechanical circuits plate device is positioned over image collector and is set up, as shown in Figure 2, CCD(imageing sensor by decision-making system) and camera lens automatically gather the on-the-spot circuit board components image of import bank, and collect in PC by NI image pick-up card, carry out image denoising, image characteristics extraction by graphical analysis determination module.
After the feature extracting image, first judge whether to exist significant deficiency (as circuit board fracture, components and parts come off), if exist, be then directly judged to be old circuit board or defective circuit board; If do not exist, then proceed following step, as shown in Figure 3.
By each feature according to shown in Fig. 4, be automatically classified into the feature that ink color feature c, silk-screen layer printed words feature w, defect characteristic d, electrical node oxidation characteristic h etc. are dissimilar.
Wherein, ink color feature, according to available circuit plate ink colors, is divided into green c
g, red c
r, black c
b1, blue c
b2deng subcharacter; Silk-screen layer printed words feature, according to new and old change, is divided into wearing and tearing w
twith color change w
ctwo seed characteristics; Defect characteristic routinely defect is divided into defect d
w, corrosion d
c, spot d
bdeng subcharacter; Electrical node oxidation characteristic is divided into pad oxide h
pwith via hole oxidation h
otwo seed characteristics; All select corresponding applicable subcharacter according to actual conditions.
Step 20: the standard feature stored in above-mentioned ink color feature, silk-screen layer printed words feature, defect characteristic and electrical node oxidation characteristic and the statistical nature database to set up in advance is contrasted, determines the new and old coefficient corresponding with each feature;
Above-mentioned steps is specially: first by gathering 0.125,0.25,0.5,1,2,3,5,7,10,10 with the subcharacter of first-class time limit circuit board components in advance, as standard feature, set up the statistical nature database of the corresponding different new and old coefficient of various criterion feature, as the contrast standard of the new and old coefficient of subcharacter.
Then each subcharacter in above-mentioned classified each subcharacter and property data base is contrasted, determines the new and old coefficient of each subcharacter of circuit-under-test plate,
Step 30: the new and old coefficient utilizing above-mentioned feature, and the predetermined weight coefficient corresponding with this feature, draw new and old coefficient of determination;
Above-mentioned steps is specially: in advance according to the influence degree statistics that each subcharacter judges circuit board newness degree, determine the weight coefficient of each subcharacter, as shown in Figure 3, comprises ink color α, silk-screen layer printed words wearing and tearing β
1, silk-screen layer printed words color change β
2, defect χ
1, corrosion χ
2, spot χ
3, electrical node pad oxide δ
1, electrical node via hole oxidation δ
2, and meet following relational expression
α+β
1+β
2+χ
1+χ
2+χ
3+δ
1+δ
2=1
Set up following new and old decision model based on each subcharacter weight coefficient of above-mentioned establishment and new and old coefficient, and calculate new and old coefficient of determination R.
R=αc+β
1w
t+β
2w
c+χ
1d
w+χ
2d
c+χ
3d
b+δ
1h
p+δ
2h
o
If some subcharacter does not exist in institute's testing circuit plate, then respective weights coefficient gets 0, and presses weight proportion by its weight allocation in other subitems, ensures the accuracy of result of determination.
Step 40: this new and old coefficient of determination and preset value are compared, draws newness degree result of determination.
Above-mentioned steps is specially: according to above formula, calculates the new and old coefficient of determination R of circuit-under-test plate, and judges import electromechanical circuits plate device newness degree according to R value size, draw judges conclusion as: brand-new, renovate, new and oldly to load in mixture or old circuit board.
The above embodiment only have expressed several embodiment of the present invention, and it describes comparatively concrete and detailed, but therefore can not be interpreted as the restriction to the scope of the claims of the present invention.It should be pointed out that for the person of ordinary skill of the art, without departing from the inventive concept of the premise, can also make some distortion and improvement, these all belong to protection scope of the present invention.Therefore, the protection domain of patent of the present invention should be as the criterion with claims.
Claims (7)
1., based on a circuit board newness degree decision method for characteristics of image, it is characterized in that comprising the following steps:
The image of collecting circuit board, extract the feature in image, and the feature in image is classified according to the ink color of circuit board, silk-screen layer printed words, defect and electrical node oxidization condition, be divided into ink color feature, silk-screen layer printed words feature, defect characteristic and electrical node oxidation characteristic;
The standard feature stored in above-mentioned ink color feature, silk-screen layer printed words feature, defect characteristic and electrical node oxidation characteristic and the statistical nature database to set up in advance is contrasted, determines the new and old coefficient corresponding with each feature;
Utilize the new and old coefficient of above-mentioned feature, and the predetermined weight coefficient corresponding with this feature, draw new and old coefficient of determination;
The influence degree that described weight coefficient judges circuit board newness degree according to dissimilar feature is established, and is added by the weight coefficient of all types of feature and value is set to 1;
The computing method of described new and old coefficient of determination are, are multiplied by the new and old coefficient of all types of feature, then sum up, obtain new and old coefficient of determination with its weight coefficient; This new and old coefficient of determination and preset value are compared, draws newness degree result of determination.
2. the circuit board newness degree decision method based on characteristics of image according to claim 1, is characterized in that, the described statistical nature database set up in advance is set up by the following method:
Gather the image of Different years circuit board in advance, extract the feature in image, as standard feature, set up the statistical nature database of the corresponding different new and old coefficient of various criterion feature.
3. the circuit board newness degree decision method based on characteristics of image according to claim 1, is characterized in that, after collecting the image of circuit board, first through image denoising, then extract the feature in image.
4. the circuit board newness degree decision method based on characteristics of image according to claim 1, it is characterized in that, described ink color feature comprises green, red, black, blue subcharacter; Described silk-screen layer printed words feature comprises the subcharacter of wearing and tearing, color; Described defect characteristic comprises the subcharacter of defect, burn into spot; Described electrical node oxidation characteristic comprises the subcharacter of pad, via hole.
5. the circuit board newness degree decision method based on characteristics of image according to claim 4, it is characterized in that, as some feature do not exist in the circuit board that detects, then corresponding weight coefficient gets 0, and presses weight proportion by its weight allocation in other features.
6. the circuit board newness degree decision method based on characteristics of image according to claim 1, is characterized in that, after extracting the feature in image, first judges whether to there is significant deficiency, if exist, is then directly judged to be old circuit board or defective circuit board; If do not exist, then carry out subsequent step.
7. the decision-making system of a circuit board newness degree, it is characterized in that, adopt the decision method described in any one of claim 1-6, comprise the image capture module for gathering image, for extracting characteristics of image and carrying out the image analysis module that contrasts and the result determination module for calculating new and old coefficient of determination with statistical nature database Plays feature, described image capture module, image analysis module and result determination module are electrically connected successively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310533952.5A CN103529041B (en) | 2013-10-31 | 2013-10-31 | Based on circuit board newness degree decision method and the system of characteristics of image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310533952.5A CN103529041B (en) | 2013-10-31 | 2013-10-31 | Based on circuit board newness degree decision method and the system of characteristics of image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103529041A CN103529041A (en) | 2014-01-22 |
| CN103529041B true CN103529041B (en) | 2015-07-29 |
Family
ID=49931228
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310533952.5A Active CN103529041B (en) | 2013-10-31 | 2013-10-31 | Based on circuit board newness degree decision method and the system of characteristics of image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103529041B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104201130B (en) * | 2014-09-01 | 2017-10-03 | 上海华力微电子有限公司 | A kind of optical detecting method classified for defect |
| TWI642558B (en) * | 2016-03-01 | 2018-12-01 | 簡單綠能股份有限公司 | Product life mark printing method |
| CN109239076B (en) * | 2018-08-29 | 2021-04-06 | 西安理工大学 | Sewing stitch defect detection method based on machine vision |
| CN113791090A (en) * | 2021-09-14 | 2021-12-14 | 慧镕电子***工程股份有限公司 | Rapid verification system and method for welding defects of recovered circuit board |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1238728C (en) * | 2002-01-09 | 2006-01-25 | 英业达股份有限公司 | Circuit board detection method |
| JP4123469B2 (en) * | 2002-04-15 | 2008-07-23 | グローリー株式会社 | Feature extraction method |
| JP2009097899A (en) * | 2007-10-15 | 2009-05-07 | Yokogawa Electric Corp | Dead-weight type ic handler |
| CN101216522A (en) * | 2008-01-16 | 2008-07-09 | 中国电子科技集团公司第四十五研究所 | FPGA based printed circuit board rapid image feature value extraction detection method |
| WO2010029932A1 (en) * | 2008-09-09 | 2010-03-18 | 株式会社メガトレード | Visual examination apparatus |
| TWI405106B (en) * | 2010-04-23 | 2013-08-11 | Wen Jong Wu | Interactive multi touch computer system and control method |
| CN103063677A (en) * | 2012-12-24 | 2013-04-24 | 上海金东唐精机科技有限公司 | Multifunctional printed circuit board (PCB) test system |
-
2013
- 2013-10-31 CN CN201310533952.5A patent/CN103529041B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN103529041A (en) | 2014-01-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103529041B (en) | Based on circuit board newness degree decision method and the system of characteristics of image | |
| WO2017215241A1 (en) | Circuit board element missing detection method and system | |
| CN103886294B (en) | Method and system for recognizing number of movement times of arrester for convertor station based on images | |
| CN106680285A (en) | Infrared image-assisted method of recognizing contamination condition of insulator by visible light image | |
| CN109946303A (en) | Check device and method | |
| CN111242123A (en) | Power equipment fault diagnosis method based on infrared image | |
| CN104897367A (en) | Indicating light optical state automatic test method and system | |
| CN107103600A (en) | A kind of defects of insulator automatic testing method based on machine learning | |
| CN111127448B (en) | Method for detecting air spring fault based on isolated forest | |
| CN106067177A (en) | HDR scene method for detecting and device | |
| CN104897687B (en) | The detecting system and method for probe pin trace position | |
| CN102637258A (en) | Method for creating online surface quality detection system defect library | |
| CN202444572U (en) | Non-reference video monitoring image blurring quality detection device | |
| CN110321944A (en) | A kind of construction method of the deep neural network model based on contact net image quality evaluation | |
| CN102621154A (en) | Method and device for automatically detecting cloth defects on line based on improved differential box multi-fractal algorithm | |
| CN110146520A (en) | A kind of detection device and its detection method of printed wiring board | |
| CN105116301B (en) | A kind of data auxiliary judgment method based on dynamic statistics | |
| CN109309022A (en) | A kind of defect sampling observation method | |
| CN110503092A (en) | The improvement SSD monitor video object detection method adapted to based on field | |
| CN103543400A (en) | Method and system for judging old and new degree of circuit board based on electrical inspection | |
| CN108898187A (en) | A kind of method and device of automatic identification power distribution room indicating equipment image | |
| CN103176092A (en) | Device and method for testing static bracelet | |
| CN116067671B (en) | Method, system and medium for testing vehicle paint quality | |
| CN114324382A (en) | Panel terminal cleanliness detection method and panel terminal cleanliness detection device | |
| CN103344571A (en) | Method and device for evaluating tolerance quality of negotiable securities |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |