CN109559304A - Image quality online evaluation method, apparatus and application for industrial vision detection - Google Patents
Image quality online evaluation method, apparatus and application for industrial vision detection Download PDFInfo
- Publication number
- CN109559304A CN109559304A CN201811407420.6A CN201811407420A CN109559304A CN 109559304 A CN109559304 A CN 109559304A CN 201811407420 A CN201811407420 A CN 201811407420A CN 109559304 A CN109559304 A CN 109559304A
- Authority
- CN
- China
- Prior art keywords
- image quality
- image
- assessed
- realtime graphic
- online evaluation
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of image quality online evaluation method, apparatus and application for industrial vision detection, described image quality online evaluation method extracts the qualitative characteristics of the realtime graphic to be assessed the following steps are included: 1) obtain realtime graphic to be assessed；2) qualitative characteristics of the realtime graphic to be assessed are compared with preset image quality reference data, described image quality reference data is set based on the qualitative characteristics of qualified images；3) judge whether image quality is qualified according to the comparison result that step 2) obtains, when the judgment result is No, generate abnormal alarm.Compared with prior art, the present invention have many advantages, such as reduce because image quality it is unqualified caused by failure rate, promote the accuracy rate of testing result.
Description
Technical field
The present invention relates to technical field of vision detection, exist more particularly, to a kind of image quality for industrial vision detection
Line appraisal procedure, device and application.
Background technique
With the extensive use of machine vision and computer vision technique in terms of industrial products and components vision-based detection,
The realtime graphic that can be acquired to image-taking devices such as industrial cameras carries out online analysis and processing and becomes more and more important.In capture mould
After group collects image, present extensive way be carried out directly to image processing algorithm product or components qualification with it is unqualified
The image procossing of the artificial intelligence technologys such as Quality Detection result judgement, either traditional algorithm or deep learning, is not always the case,
Lack the process that quality evaluation and real-time detection are carried out to acquisition image.
Some non-industrial picture quality offline evaluation methods are currently, there are, there is the appraisal procedure of focusedimage overall situation quality,
Such as Y-PSNR PSNR (Peak Signal to Noise Ratio), structural similarity SSIM (structural
Similarity), mean square error MSE (mean square error), fidelity of information criterion IFC (Information
Fidelity Criterion), visual information fidelity VIF (Visual Information Fidelity).These technologies exist
Non- industrial circle has certain practicability, achieves some satisfactory results, and still, industrial vision detects scene demand not
Together, by industrial camera noise, light-source brightness variation, lens focus adjustment, movement mechanism error and such as clean journey of working environment
Degree etc. be affected, and industrial vision detect the technology be using image processing algorithm compatibility and accuracy as demand premise,
These appraisal procedures are difficult to directly apply.
There is also following defects in industrial vision detection application aspect for these appraisal procedures at present:
First, from the global statistics of image pixel value, the local visual factor of human eye is not considered, therefore to image office
Portion's quality can not be held, and global characteristics are both paid close attention in industrial vision detection, also pay close attention to local region of interest (ROI, Region of
Interest image quality feature).
Second, it is difficult to meet real-time online application, fail to carry out reverse reason rational analysis according to testing result, provide figure
As the underproof analysis of causes of quality, do not have decision making function, it is difficult to quickly orient the underproof reason of image quality, such as produce
Product clean level is abnormal, image-taking device is abnormal, movement mechanism is abnormal etc..
Third, current industrial vision detection field are also required to batch other than being required to meet real time on-line monitoring
Duplication, and the image quality parameter of measurement quantified, facilitate the quick adjustment for fast implementing image quality, guarantee that image procossing is calculated
The compatibility and accuracy of method, while equipment failure rate is reduced, these methods are difficult to directly be applicable in, and need to be examined according to industrial vision
The particular demands for surveying scene, create new industrial vision appearance detection image quality online evaluation method.
Summary of the invention
It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art and provide one kind to be used for industrial vision
The image quality online evaluation method, apparatus of detection and application.
The purpose of the present invention can be achieved through the following technical solutions:
A kind of image quality online evaluation method for industrial vision detection, which comprises the following steps:
1) realtime graphic to be assessed is obtained, the qualitative characteristics of the realtime graphic to be assessed are extracted；
2) qualitative characteristics of the realtime graphic to be assessed are compared with preset image quality reference data,
Described image quality reference data is set based on the qualitative characteristics of qualified images；
3) judge whether image quality is qualified according to the comparison result that step 2) obtains, when the judgment result is No, generate
Abnormal alarm.
Further, the qualitative characteristics include image overall region or ROI region global characteristics and image or the area ROI
Domain local feature.
Further, in the step 3), when the qualitative characteristics of realtime graphic to be assessed are in image quality reference data
When in range, it is determined as qualification, otherwise, it is determined that being unqualified.
Further, in the step 3), while generating abnormal alarm, call history exception database output abnormality former
Cause.
The present invention also provides a kind of image quality online evaluation devices for industrial vision detection, comprising:
Realtime graphic qualitative characteristics obtain module and extract the realtime graphic to be assessed for obtaining realtime graphic to be assessed
Qualitative characteristics；
Comparison module, for by the qualitative characteristics of the realtime graphic to be assessed and preset image quality base value
According to being compared；
As a result output module when the judgment result is No, is produced for judging whether image quality is qualified according to comparison result
Raw abnormal alarm.
Further, the qualitative characteristics include image overall region or ROI region global characteristics and image or the area ROI
Domain local feature.
Further, in the result output module, when the qualitative characteristics of realtime graphic to be assessed are in image quality base
When in quasi- data area, it is determined as qualification, otherwise, it is determined that being unqualified.
Further, the result output module further include:
Abnormal cause output unit is generating abnormal alarm response, for calling history exception database output abnormality former
Cause.
The present invention also provides a kind of industrial products machine vision product quality detecting methods, comprising the following steps:
Obtain input acquisition image；
Image quality analysis is carried out using image quality online evaluation method as described in claim 1, obtains image product
The image of matter qualification；
Image based on described image quality qualification carries out product quality detection.
Compared with prior art, the present invention have with following the utility model has the advantages that
First, present invention firstly provides the method that realtime graphic quality estimating is added when industrial vision detects, Neng Gouyou
Effect is suitable for industrial scene.
Second, the present invention has carried out screening to input picture and has differentiated, guarantees that the image of image quality qualification enters appearance inspection
Method of determining and calculating process flow, improves the compatibility of algorithm, reduce because image quality it is unqualified caused by failure rate, promote testing result
Accuracy rate.
Third, the present invention can carry out backward inference to abnormal cause according to the warning message of image quality negative characteristics,
Quick positioning failure reason carries out decision for operator and engineer, and work effect can be improved in fast and easy maintenance adjustment
Rate.
4th, image quality reference data of the present invention foundation can share, can according to image quality exemplary feature, as
The reference specification of image-taking device adjustment, in batch duplicating equipment, it is ensured that the image quality of each equipment is close, guarantees soft
Part algorithm and parameter can be reduced with batch duplicating and be adjusted the ginseng time, shorten the friendship phase, working efficiency can be improved, save human cost.
Detailed description of the invention
Fig. 1 is the flow diagram of image quality online evaluation of the present invention；
Fig. 2 is the sharpness schematic diagram of calculation flow in image or ROI region local feature；
Fig. 3 is the center-of-mass coordinate schematic diagram of calculation flow in image or ROI region local feature；
Fig. 4 is the Industry Product Appearance testing process schematic diagram based on image quality online evaluation of the present invention；
Fig. 5 is test image schematic diagram in embodiment.
Specific embodiment
The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.The present embodiment is with technical solution of the present invention
Premised on implemented, the detailed implementation method and specific operation process are given, but protection scope of the present invention is not limited to
Following embodiments.
The present invention provides a kind of image quality online evaluation method for industrial vision detection, can be applied to industrial products
In machine vision product quality detection system, image is filtered and is fed back, the pretreatment before realizing product quality detection.
As shown in Figure 1, the image quality online evaluation method the following steps are included:
Step S101 obtains realtime graphic to be assessed, extracts the qualitative characteristics of the realtime graphic to be assessed；
Step S102, by the qualitative characteristics of the realtime graphic to be assessed and preset image quality reference data into
Row compares, and described image quality reference data is set based on the qualitative characteristics of qualified images, and picture number should meet sample as far as possible
This randomness, quantity are not less than 100, and benchmark need to only be established once, can change according to actual needs, adjust reference data；
Step S103 judges whether image quality is qualified according to the comparison result that step S102 is obtained, is in judging result
When no, abnormal alarm is generated.When the qualitative characteristics of realtime graphic to be assessed are within the scope of image quality reference data, determine
For qualification, otherwise, it is determined that being unqualified.
It extracts the mode of the qualitative characteristics of realtime graphic to be assessed and establishes in image quality reference data to qualified images
Qualitative characteristics extracting mode be identical.The qualitative characteristics include image overall region or ROI region global characteristics and
Image or ROI region local feature, wherein image overall region or ROI region global characteristics include average brightness, uniformity
Deng for measuring the overall variation of image overall region or ROI region；Image or ROI region local feature include clear-cut margin
The center-of-mass coordinate of degree and such as local hole of product geometry or profile, for measuring image or ROI region localized variation,
The extracting mode of ROI includes but is not limited to image binaryzation, the modes such as contours extract or colored pixels matching.
The calculation of average brightness m and uniformity std in global characteristics are to count global or ROI region gray scale to put down
Mean value counts global or ROI region gray standard deviation uniformity index as a whole, formula is such as average brightness index
Under:
The center-of-mass coordinate of such as local hole of clear-cut margin degree and product geometry or profile in local feature is calculated and is needed
The sampling line that bounding edge region is set for fringe region detects the sharpness in specified local edge region, constrains crossover track
Length does not intersect with other edges, and the range apart from about 20 pixels of both sides of edges, crossover track quantity are no less than 10, according to
Fringe region length is uniformly distributed, and center-of-mass coordinate is to elect product hole location or local configuration to be advisable.Fig. 2 is sharpness calculating
Process includes the following steps:
Step S201, input sharpness assess gray scale upper and lower limits；
Step S202 retrieves every crossover track pixel-by-pixel；
Step S203 retrieves the gray value of crossover track pixel-by-pixel；
Step S204 judges gray value whether in gray scale upper and lower limits, if so, gray scale is designated as 1, if it is not, then grey
Scale is 0；
Step S205, the sum for counting the gray scale on crossover track is the sharpness of single crossover track；
Step S206, the sharpness mean value for counting all crossover tracks are exported as final sharpness.
Fig. 3 is center-of-mass coordinate calculation process, is included the following steps:
Step S301, fixed area binaryzation；
Step S302 extracts profile；
Step S303 calculates profile mass center.
After the feature for counting normal picture quality qualification, all approximate Gaussian distributed of discovery single features:
The present invention can set reference data range according to available accuracy demand, and 95.449974% area is in average value
Left and right two 2 σ of standard deviation in the range of, 99.730020% area in the range of three 3 σ of standard deviation of average value or so,
99.993666% area is defaulted as three standard deviations 3 of average value or so in the range of four 4 σ of standard deviation of average value or so
In the range of σ,
[Fmin, Fmax] wherein min=u-3 σ, max=u+3 σ
So the present embodiment can establish image quality character references range according to statistical information:
The reference range [Fmin1, Fmax1] of average brightness
The reference range [Fmin2, Fmax2] of the uniformity
The reference range [Fmin3, Fmax3] of sharpness
The reference range [Fmin4, Fmax4] of center-of-mass coordinate x
The reference range [Fmin5, Fmax5] of center-of-mass coordinate y
It is judged as qualified if the image quality feature of real-time detection is all fallen in range, otherwise exports each and do not conform to
The characteristic value and warning message of lattice.
In another embodiment, while generating abnormal alarm, history exception database output abnormality reason is called, it is convenient
Industry spot operator and Maintenance Engineer position rapidly abnormal cause, provide reasonable quick solution, such as:
If average brightness > Fmax1, image is partially bright, and there are large area height is reflective miscellaneous for possible light source exception or product surface
Matter etc.；
If average brightness < Fmin1, image is partially dark, and there are pollution in wide area for possible light source decaying or product surface, or deposit
It is blocking；
If sharpness > Fmax3, Edge region blur, possible focal distance fluctuation is focused inaccurate；
If center-of-mass coordinate x or y are abnormal, exception may be positioned, product is not in place or movement mechanism is abnormal.
As shown in figure 4, it is in the industrial products machine vision product Quality Detection of above-mentioned image quality online evaluation method
Method, comprising the following steps:
Step S401 obtains input acquisition image；
Step S402 carries out image quality analysis using the image quality online evaluation method, obtains image quality
Qualified image；
Step S403, the image based on described image quality qualification carry out include product appearance etc. product quality detection.
In order to verify performance of the invention, the present embodiment is illustrated with certain Industry Product Appearance detection image, and Fig. 5 is to survey
Attempt picture, first setting topography's quality department feature detection zone, sample areas contains circular hole in figure, and circular hole center-of-mass coordinate can
As position coordinates detection information, for positioning positional shift whether occurs, bore edges or left side edge can be used as edge
Secondly sharpness parameter measure region sets global image qualitative characteristics detection zone, obtain ROI region (foreground area), such as
Shown in the grey parts of Fig. 5 (b), the average gray value in ROI region is counted as average brightness, standard deviation is as the uniformity.
100 normal sample images are screened, reference data is established.
Reference range [Fmin1, Fmax1]=[58.1,63.4] of average brightness
Reference range [Fmin2, Fmax2]=[]=[8.1,12.3] of the uniformity
Reference range [Fmin3, Fmax3]=[2.2-4.5] of sharpness
Reference range [Fmin4, Fmax4]=[169.5-173.2] of center-of-mass coordinate x
Reference range [Fmin5, Fmax5]=[1022-1025] of center-of-mass coordinate y
After the characteristics of image reference data confirmation of 1 equipment, copying equipment is adjusted using this benchmark as common reference
It is whole, achieve the effect that (all same type equipment share a set of benchmark to batch duplicating, and benchmark is reproducible, it is only necessary to adjust hardware, completely
Sufficient benchmark requirement, all establishes benchmark without every).It is reference with the benchmark in embodiment, the features such as all brightness need
It is adjusted within the scope of this, guarantees that multiple devices image quality is approximate, it is ensured that software parameter reusable adjusted saves software tune
Join the time.
The preferred embodiment of the present invention has been described in detail above.It should be appreciated that those skilled in the art without
It needs creative work according to the present invention can conceive and makes many modifications and variations.Therefore, all technologies in the art
Personnel are available by logical analysis, reasoning, or a limited experiment on the basis of existing technology under this invention's idea
Technical solution, all should be within the scope of protection determined by the claims.
Claims (9)
1. a kind of image quality online evaluation method for industrial vision detection, which comprises the following steps:
1) realtime graphic to be assessed is obtained, the qualitative characteristics of the realtime graphic to be assessed are extracted；
2) qualitative characteristics of the realtime graphic to be assessed are compared with preset image quality reference data, it is described
Image quality reference data is set based on the qualitative characteristics of qualified images；
3) judge whether image quality is qualified according to the comparison result that step 2) obtains, when the judgment result is No, generate abnormal
Alarm.
2. the image quality online evaluation method according to claim 1 for industrial vision detection, which is characterized in that institute
Stating qualitative characteristics includes image overall region or ROI region global characteristics and image or ROI region local feature.
3. the image quality online evaluation method according to claim 1 for industrial vision detection, which is characterized in that institute
It states in step 3), when the qualitative characteristics of realtime graphic to be assessed are within the scope of image quality reference data, is determined as qualification,
Otherwise, it is determined that being unqualified.
4. the image quality online evaluation method according to claim 1 for industrial vision detection, which is characterized in that institute
It states in step 3), while generating abnormal alarm, calls history exception database output abnormality reason.
5. a kind of image quality online evaluation device for industrial vision detection characterized by comprising
Realtime graphic qualitative characteristics obtain module and extract the product of the realtime graphic to be assessed for obtaining realtime graphic to be assessed
Matter feature；
Comparison module, for by the qualitative characteristics of the realtime graphic to be assessed and preset image quality reference data into
Row compares；
As a result output module when the judgment result is No, generates different for judging whether image quality is qualified according to comparison result
Often alarm.
6. the image quality online evaluation device according to claim 5 for industrial vision detection, which is characterized in that institute
Stating qualitative characteristics includes image overall region or ROI region global characteristics and image or ROI region local feature.
7. the image quality online evaluation device according to claim 5 for industrial vision detection, which is characterized in that institute
It states in result output module, when the qualitative characteristics of realtime graphic to be assessed are within the scope of image quality reference data, determines
For qualification, otherwise, it is determined that being unqualified.
8. the image quality online evaluation device according to claim 5 for industrial vision detection, which is characterized in that institute
State result output module further include:
Abnormal cause output unit is generating abnormal alarm response, for calling history exception database output abnormality reason.
9. a kind of industrial products machine vision product quality detecting method, which comprises the following steps:
Obtain input acquisition image；
Image quality analysis is carried out using image quality online evaluation method as described in claim 1, image quality is obtained and closes
The image of lattice；
Image based on described image quality qualification carries out product quality detection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811407420.6A CN109559304A (en) | 2018-11-23 | 2018-11-23 | Image quality online evaluation method, apparatus and application for industrial vision detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811407420.6A CN109559304A (en) | 2018-11-23 | 2018-11-23 | Image quality online evaluation method, apparatus and application for industrial vision detection |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109559304A true CN109559304A (en) | 2019-04-02 |
Family
ID=65867155
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811407420.6A Pending CN109559304A (en) | 2018-11-23 | 2018-11-23 | Image quality online evaluation method, apparatus and application for industrial vision detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109559304A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109946859A (en) * | 2019-04-09 | 2019-06-28 | 深圳市华星光电半导体显示技术有限公司 | The detection method and device of backlight module |
| CN111273264A (en) * | 2020-02-14 | 2020-06-12 | 广东博智林机器人有限公司 | Method and device for determining optimal position of light source of machine vision system |
| CN112621408A (en) * | 2020-12-16 | 2021-04-09 | 新沂海福尔通用仪表有限公司 | Transparent quartz tube preform polishing system with recovery mechanism and control method thereof |
| CN112907572A (en) * | 2021-03-24 | 2021-06-04 | 国家石油天然气管网集团有限公司华南分公司 | Motor control accuracy assessment method and system |
| CN113065559A (en) * | 2021-06-03 | 2021-07-02 | 城云科技（中国）有限公司 | Image comparison method and device, electronic equipment and storage medium |
| CN113553999A (en) * | 2021-09-17 | 2021-10-26 | 江苏新恒基特种装备股份有限公司 | Method and system for judging molding abnormity based on image recognition |
| CN113762427A (en) * | 2021-11-10 | 2021-12-07 | 聚时科技(江苏)有限公司 | Feeding abnormity detection method in industrial automation detection scene |
| TWI779808B (en) * | 2021-08-30 | 2022-10-01 | 宏碁股份有限公司 | Image processing method |
| TWI793653B (en) * | 2020-10-29 | 2023-02-21 | 日商樂天集團股份有限公司 | Information processing device and information processing method |
| CN117124560A (en) * | 2023-09-28 | 2023-11-28 | 佛山赛和薄膜科技有限公司 | Preparation method of MLCC polyester base film for 5G communication |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2225726A2 (en) * | 2008-01-04 | 2010-09-08 | Thomson Licensing | Method for assessing image quality |
| US20150317779A1 (en) * | 2014-05-02 | 2015-11-05 | Ricoh Company, Limited | Image processing system, image processing method, and computer program product |
| CN106204524A (en) * | 2016-06-23 | 2016-12-07 | 凌云光技术集团有限责任公司 | A kind of method and device of evaluation image quality |
| CN107886500A (en) * | 2017-10-13 | 2018-04-06 | 北京邮电大学 | A kind of production monitoring method and system based on machine vision and machine learning |
-
2018
- 2018-11-23 CN CN201811407420.6A patent/CN109559304A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2225726A2 (en) * | 2008-01-04 | 2010-09-08 | Thomson Licensing | Method for assessing image quality |
| US20150317779A1 (en) * | 2014-05-02 | 2015-11-05 | Ricoh Company, Limited | Image processing system, image processing method, and computer program product |
| CN106204524A (en) * | 2016-06-23 | 2016-12-07 | 凌云光技术集团有限责任公司 | A kind of method and device of evaluation image quality |
| CN107886500A (en) * | 2017-10-13 | 2018-04-06 | 北京邮电大学 | A kind of production monitoring method and system based on machine vision and machine learning |
Non-Patent Citations (2)
| Title |
|---|
| 张政等: "简论一种工业CT图像质量的快速评价方法", 《 2006年西南地区第九届NDT学术年会暨2006年全国射线检测新技术研讨会论文集》 * |
| 王靖宇等: "工业机器人视觉图像的质量评价方法", 《机械科学与技术》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109946859A (en) * | 2019-04-09 | 2019-06-28 | 深圳市华星光电半导体显示技术有限公司 | The detection method and device of backlight module |
| CN109946859B (en) * | 2019-04-09 | 2022-02-01 | 深圳市华星光电半导体显示技术有限公司 | Backlight module detection method and device |
| CN111273264A (en) * | 2020-02-14 | 2020-06-12 | 广东博智林机器人有限公司 | Method and device for determining optimal position of light source of machine vision system |
| TWI793653B (en) * | 2020-10-29 | 2023-02-21 | 日商樂天集團股份有限公司 | Information processing device and information processing method |
| CN112621408A (en) * | 2020-12-16 | 2021-04-09 | 新沂海福尔通用仪表有限公司 | Transparent quartz tube preform polishing system with recovery mechanism and control method thereof |
| CN112907572A (en) * | 2021-03-24 | 2021-06-04 | 国家石油天然气管网集团有限公司华南分公司 | Motor control accuracy assessment method and system |
| CN113065559A (en) * | 2021-06-03 | 2021-07-02 | 城云科技（中国）有限公司 | Image comparison method and device, electronic equipment and storage medium |
| TWI779808B (en) * | 2021-08-30 | 2022-10-01 | 宏碁股份有限公司 | Image processing method |
| CN113553999A (en) * | 2021-09-17 | 2021-10-26 | 江苏新恒基特种装备股份有限公司 | Method and system for judging molding abnormity based on image recognition |
| CN113553999B (en) * | 2021-09-17 | 2021-12-17 | 江苏新恒基特种装备股份有限公司 | Method and system for judging molding abnormity based on image recognition |
| CN113762427A (en) * | 2021-11-10 | 2021-12-07 | 聚时科技(江苏)有限公司 | Feeding abnormity detection method in industrial automation detection scene |
| CN117124560A (en) * | 2023-09-28 | 2023-11-28 | 佛山赛和薄膜科技有限公司 | Preparation method of MLCC polyester base film for 5G communication |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109559304A (en) | Image quality online evaluation method, apparatus and application for industrial vision detection | |
| CN109816678B (en) | Automatic nozzle atomization angle detection system and method based on vision | |
| CN112053318B (en) | Two-dimensional PCB defect real-time automatic detection and classification device based on deep learning | |
| CN102175700B (en) | Method for detecting welding seam segmentation and defects of digital X-ray images | |
| CN109308447A (en) | The method of equipment operating parameter and operating status is automatically extracted in remote monitoriong of electric power | |
| CN109084955A (en) | Display screen quality determining method, device, electronic equipment and storage medium | |
| CN107895362B (en) | Machine vision method for detecting quality of miniature wiring terminal | |
| CN106623493B (en) | A kind of detection method of steel band punching Continuous maching | |
| JP5168215B2 (en) | Appearance inspection device | |
| CN108921840A (en) | Display screen peripheral circuit detection method, device, electronic equipment and storage medium | |
| CN110146019A (en) | A kind of workpiece size on-line checking and automatic compensating method based on industrial camera | |
| CN110186375A (en) | Intelligent high-speed rail white body assemble welding feature detection device and detection method | |
| CN108961239A (en) | Continuous casting billet quality detection method, device, electronic equipment and storage medium | |
| CN110189375A (en) | A kind of images steganalysis method based on monocular vision measurement | |
| CN110096980A (en) | Character machining identifying system | |
| CN102621154A (en) | Method and device for automatically detecting cloth defects on line based on improved differential box multi-fractal algorithm | |
| CN113808087A (en) | Defect management and control method and device for surface of steel plate and computer readable storage medium | |
| CN107607205A (en) | Wire harness color sequences detecting system and method | |
| CN117237747B (en) | Hardware defect classification and identification method based on artificial intelligence | |
| CN113191977A (en) | Image enhancement system for target detection and identification under severe environment condition | |
| CN116840240A (en) | Visual detection system of power supply distributor | |
| CN117035669A (en) | Enterprise safety production management method and system based on artificial intelligence | |
| CN111815705B (en) | Method and device for identifying pollution of filter protection lens of laser tracker and electronic equipment | |
| CN117372321A (en) | Defect detection method, device, equipment and storage medium | |
| TWM619409U (en) | Intelligent processing system |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190402 |
|
| RJ01 | Rejection of invention patent application after publication |