CN109215547A - A kind of method and system of fast storage backlight inspection image - Google Patents
A kind of method and system of fast storage backlight inspection image Download PDFInfo
- Publication number
- CN109215547A CN109215547A CN201811324622.4A CN201811324622A CN109215547A CN 109215547 A CN109215547 A CN 109215547A CN 201811324622 A CN201811324622 A CN 201811324622A CN 109215547 A CN109215547 A CN 109215547A
- Authority
- CN
- China
- Prior art keywords
- image
- pretreatment
- benchmark
- foreground
- pixel
- 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
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000007689 inspection Methods 0.000 title claims abstract description 19
- 230000006835 compression Effects 0.000 claims abstract description 33
- 238000007906 compression Methods 0.000 claims abstract description 33
- 238000004321 preservation Methods 0.000 claims abstract description 9
- 230000002950 deficient Effects 0.000 claims abstract description 8
- 238000006073 displacement reaction Methods 0.000 claims description 4
- 230000006870 function Effects 0.000 abstract description 9
- 238000001514 detection method Methods 0.000 abstract description 7
- 238000013135 deep learning Methods 0.000 abstract description 3
- 230000007547 defect Effects 0.000 description 9
- 238000007726 management method Methods 0.000 description 6
- 238000000151 deposition Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003760 hair shine Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G3/00—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes
- G09G3/006—Electronic inspection or testing of displays and display drivers, e.g. of LED or LCD displays
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of method and systems of fast storage backlight inspection image.It includes compression of images storage operation, process are as follows: step 1, one width of acquisition does not have defective panel image as benchmark image, and carries out compression preservation；Step 2, actual panel image is acquired as pretreatment image, and difference shifting function is carried out to pretreatment image by benchmark image and obtains foreground image, compression preservation is carried out to foreground image；Step 3, step 2 is repeated until having acquired all panel images.The present invention is directed to the feature of BLU panel detection image, while greatly reducing image self-information amount, form the relevant image sequence in front and back, both the compression ratio of image had been improved, reduce compression time, sample sequence is had collected for LSTM deep learning again, human cost is greatly reduced, improves efficiency.
Description
Technical field
The invention belongs to the automation defect detecting technique fields of BLU, are based especially on the BLU defects detection field of AOI,
More particularly to a kind of method and system of fast storage backlight inspection image.
Background technique
Light source of the BLU (backlight assembly) as LCD display, with the development of display technology, yield is increasing,
Defects detection required precision is also higher and higher, and the resolution ratio of the corresponding industrial camera used is also higher and higher, in addition, certain defects
Only can just it be taken under side view environment.Assuming that 15.6 cun of BLU of detection, are meeting the premise that minimum detection precision is φ 0.05mm
Under, at least need 1 29M to face camera and a 20M side view camera.Assuming that 2 Pattern of every piece of BLU needs are detected, one
Producing line produces 5000 BLU daily, then needing the space 2*5000* (29+20)=478.5GB to be stored altogether.Generally
It is required that storage time it is long, generally more than half a year, it is assumed that depositing figure requirement is half a year, then the memory space needed in total is
478.5*30*6=84.1TB.
The purpose for depositing figure is easy for the tracking and analysis of problem, and since the defect of BLU is likely to minor defect, therefore not
The storage modes such as lossy compression can be carried out to image, require consideration for how that carrying out permanent effective image is stored in management and drops
Low memory space.
Existing technical solution, for image storage space is reduced, there are mainly two types of thinkings:
Original image storage: as carried out original image storage using bmp format, to image carry out it is not any damage or lossless compression, directly
Connect storage original image.Such method consuming memory space is larger, is not preferable solution.
Lossless compression storage: original image is compressed using png format or tiff format etc., is then stored.Such side
Although method can reduce a part of memory space, in the biggish situation of output image memory, performance loss is more serious, pressure
Contracting efficiency is relatively low.
Summary of the invention
The object of the invention is in order to solve deficiency existing for above-mentioned background technique, provide a kind of high efficiency, high compression
The method and system of the fast storage backlight inspection image of rate.
The technical solution adopted by the present invention is that: image management method in a kind of inspection of backlight, including compression of images storage operation,
Process are as follows:
Step 1, one width of acquisition does not have defective panel image as benchmark image, and carries out compression preservation；
Step 2, actual panel image is acquired as pretreatment image, and it is poor to carry out by benchmark image to pretreatment image
Value shifting function obtains foreground image, carries out compression preservation to foreground image；
Step 3, step 2 is repeated until having acquired all panel images.
Further, the difference shifting function process are as follows: subtract the gray value of pixel each in pretreatment image
The gray value of corresponding each pixel obtains error image on benchmark image, then carries out 128 bit shift behaviour to error image
Obtain foreground image.
It further, further include image restoring operation, process are as follows: the foreground image and benchmark image for needing to restore are read,
The pretreatment image that contrary operation is needed is carried out to foreground image according to benchmark image.
Further, the process of the contrary operation are as follows: foreground image is subjected to 128 backward shift bit manipulations, after operation
Image in the gray value of each pixel obtain pretreatment image plus the gray value of each pixel in benchmark image.
Image management system in a kind of inspection of backlight, including
Image acquisition units do not have defective panel image and actual panel image for acquiring, are respectively defined as base
Benchmark image and pretreatment image are sent to image processing unit by quasi- image and pretreatment image；
Image processing unit obtains foreground image for carrying out difference shifting function to pretreatment image, to foreground image
It is compressed and is sent to image storage unit；
Image storage unit, for storing compressed benchmark image and foreground image.
Further, the difference shifting function process are as follows: subtract the gray value of pixel each in pretreatment image
The gray value of corresponding each pixel obtains error image on benchmark image, then carries out 128 bit shift behaviour to error image
Obtain foreground image.
Further, described image processing unit is also used to read the foreground image for needing to restore from image storage unit
And benchmark image；For carrying out the pretreatment image that contrary operation is needed to foreground image according to benchmark image.
Further, the process of the contrary operation are as follows: foreground image is subjected to 128 backward shift bit manipulations, after operation
The gray value of each pixel obtains pretreatment image plus the gray value of each pixel in benchmark image in image.
Further, described image storage unit includes the first storage unit and the second storage unit, and described first deposits
Storage unit is for storing compressed benchmark image, and second storage unit is for storing compressed foreground image.
The present invention is directed to the feature of BLU panel detection image, while greatly reducing image self-information amount, forms
The relevant image sequence in front and back had both improved the compression ratio of image, reduced compression time, and have collected sample for LSTM deep learning
This sequence greatly reduces human cost, improves efficiency.
Detailed description of the invention
Fig. 1 is the flow diagram of image storing operation of the invention.
Fig. 2 is the flow diagram of image restoring of the present invention operation.
Specific embodiment
The following further describes the present invention in detail with reference to the accompanying drawings and specific embodiments, convenient for this hair is well understood
It is bright, but they limiting the invention.
Since BLU shines uniformly, under the same Pattern, the same camera shooting, the panel image of shooting, same to position
Pixel grey scale otherness is little, and only the panel image of existing defects just can there are a degree of differences with normal panel image
It is different.
Based under the same Pattern of same camera, the very small feature of BLU panel image otherness, the invention proposes
A kind of efficient image management method.The basic implementation method of this method is as follows: firstly, shooting one does not have defective BLU panel
Image regards benchmark, saves without operating directly compression；Then, all images of subsequent shooting all subtract the benchmark image,
Background information is eliminated in obtained image, only remaining part foreground information relevant to defect and noise.In order to eliminate result
Negative value information in image, result figure is carried out 128 displacement bit manipulations by us.Because image information greatly reduces, so compression
Speed can greatly promote.
The restoring operation of image is then the inverse operation of the above process, only benchmark image and foreground image need to be read, by prospect
Image carries out 128 backward shift bit manipulations, then adds benchmark image.
Detailed compression storage operation is as shown in Figure 1, process are as follows:
Step 1, one width of acquisition does not have defective panel image as benchmark image, and carries out compression preservation；
Step 2, actual panel image is acquired as pretreatment image, and it is poor to carry out by benchmark image to pretreatment image
Value shifting function obtains foreground image, carries out compression preservation to foreground image；
Step 3, step 2 is repeated until having acquired all panel images.
In above scheme, the difference shifting function process are as follows: subtract the gray value of pixel each in pretreatment image
It goes the gray value of corresponding each pixel on benchmark image to obtain error image, 128 bit shifts then is carried out to error image
Operation (gray value of pixel each in error image is added 128) obtains foreground image.
It is as follows using this method compression BLU picture effect:
Effect analysis: the first picture is directly to carry out lossless compression preservation to zero defect image, is equivalent to using existing
Lossless compression method in scheme, compression ratio 53.6%, subsequent using method of the invention, individual compression ratio of picture exists
80% or so, the total compression ratio of all pictures is stepped up, approach 80%.
The present invention is directed to the feature of BLU panel detection image, while greatly reducing image self-information amount, forms
The relevant image sequence in front and back had both improved the compression ratio of image, reduced compression time, and have collected sample for LSTM deep learning
This sequence greatly reduces human cost, improves efficiency.Method of the invention both can achieve 80% or more compression ratio
(normal general 50%) of lossless compression, and the processing time of only common lossless compression 60% or so.
As shown in Fig. 2, the process of the image restoring operation in backlight inspection of the invention in image management method are as follows: reading needs
The foreground image and benchmark image to be restored carry out the pretreatment that contrary operation is needed to foreground image according to benchmark image
Image.The process of the contrary operation are as follows: foreground image is subjected to 128 backward shift bit manipulations (by pixel each in foreground image
Gray value all subtract 128), to the gray value of each pixel in the image after operation plus each pixel in benchmark image
Gray value obtain pretreatment image.
To realize image management method in above-mentioned backlight inspection, the present invention also provides a kind of fast storage backlight inspection images
System, including
Image acquisition units do not have defective panel image and actual panel image for acquiring, are respectively defined as base
Benchmark image and pretreatment image are sent to image processing unit by quasi- image and pretreatment image；
Image processing unit obtains foreground image for carrying out difference shifting function to pretreatment image, to foreground image
It is compressed and is sent to image storage unit；
Image storage unit, for storing compressed benchmark image and foreground image.
In above scheme, difference shifting function process are as follows: the gray value of pixel each in pretreatment image is subtracted into base
The gray value of corresponding each pixel obtains error image on quasi- image, then carries out 128 displacement bit manipulations to error image
Obtain foreground image.
In above scheme, image processing unit be also used to read from image storage unit need the foreground image that restores and
Benchmark image；For carrying out the pretreatment image that contrary operation is needed to foreground image according to benchmark image.
In above scheme, the process of contrary operation are as follows: foreground image is subjected to 128 backward shift bit manipulations, to the figure after operation
The gray value of each pixel obtains pretreatment image plus the gray value of each pixel in benchmark image as in.
In above scheme, image storage unit includes the first storage unit and the second storage unit, and the first storage unit can
To be general PC memory, the conventional hard disk of the first storage unit selection, memory capacity is big, and first storage unit is for depositing
Compressed benchmark image is stored up, second storage unit is for storing compressed foreground image.
The content being not described in detail in this specification belongs to the prior art well known to professional and technical personnel in the field.
Claims (9)
1. a kind of method of fast storage backlight inspection image, it is characterised in that: it stores and operates including compression of images, process are as follows:
Step 1, one width of acquisition does not have defective panel image as benchmark image, and carries out compression preservation；
Step 2, actual panel image is acquired as pretreatment image, and difference shifting is carried out to pretreatment image by benchmark image
Bit manipulation obtains foreground image, carries out compression preservation to foreground image；
Step 3, step 2 is repeated until having acquired all panel images.
2. the method for fast storage backlight inspection image according to claim 1, it is characterised in that: the difference shifting function
Process are as follows: the gray value of pixel each in pretreatment image is subtracted to the gray value of corresponding each pixel on benchmark image
Error image is obtained, 128 displacement bit manipulations then are carried out to error image and obtain foreground image.
3. the method for fast storage backlight inspection image according to claim 1, it is characterised in that: further include image restoring behaviour
Make, process are as follows: read the foreground image and benchmark image for needing to restore, contrary operation is carried out to foreground image according to benchmark image
The pretreatment image needed.
4. the method for fast storage backlight inspection image according to claim 3, it is characterised in that: the mistake of the contrary operation
Journey are as follows: foreground image is subjected to 128 backward shift bit manipulations, reference map is added to the gray value of each pixel in the image after operation
The gray value of each pixel obtains pretreatment image as in.
5. a kind of system of fast storage backlight inspection image, it is characterised in that: including
Image acquisition units do not have defective panel image and actual panel image for acquiring, are respectively defined as reference map
Benchmark image and pretreatment image are sent to image processing unit by picture and pretreatment image；
Image processing unit obtains foreground image for carrying out difference shifting function to pretreatment image, carries out to foreground image
It compresses and is sent to image storage unit；
Image storage unit, for storing compressed benchmark image and foreground image.
6. the system of fast storage backlight inspection image according to claim 5, it is characterised in that: the difference shifting function
Process are as follows: the gray value of pixel each in pretreatment image is subtracted to the gray value of corresponding each pixel on benchmark image
Error image is obtained, 128 displacement bit manipulations then are carried out to error image and obtain foreground image.
7. the system of fast storage backlight inspection image according to claim 5, it is characterised in that: described image processing unit
It is also used to read the foreground image and benchmark image for needing to restore from image storage unit；For according to benchmark image to prospect
Image carries out the pretreatment image that contrary operation is needed.
8. the system of fast storage backlight inspection image according to claim 7, it is characterised in that: the mistake of the contrary operation
Journey are as follows: foreground image is subjected to 128 backward shift bit manipulations, reference map is added to the gray value of each pixel in the image after operation
The gray value of each pixel obtains pretreatment image as in.
9. the system of fast storage backlight inspection image according to claim 5, it is characterised in that: described image storage unit
Including the first storage unit and the second storage unit, first storage unit is described for storing compressed benchmark image
Second storage unit is for storing compressed foreground image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811324622.4A CN109215547A (en) | 2018-11-08 | 2018-11-08 | A kind of method and system of fast storage backlight inspection image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811324622.4A CN109215547A (en) | 2018-11-08 | 2018-11-08 | A kind of method and system of fast storage backlight inspection image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109215547A true CN109215547A (en) | 2019-01-15 |
Family
ID=64995108
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811324622.4A Pending CN109215547A (en) | 2018-11-08 | 2018-11-08 | A kind of method and system of fast storage backlight inspection image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109215547A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102033073A (en) * | 2010-12-08 | 2011-04-27 | 北大方正集团有限公司 | Method for determining performance of automatic optical detection equipment and standard film |
| CN107333124A (en) * | 2017-06-30 | 2017-11-07 | 武汉精测电子技术股份有限公司 | Picture compression method and defects of display panel detecting system in AOI detecting systems |
-
2018
- 2018-11-08 CN CN201811324622.4A patent/CN109215547A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102033073A (en) * | 2010-12-08 | 2011-04-27 | 北大方正集团有限公司 | Method for determining performance of automatic optical detection equipment and standard film |
| CN107333124A (en) * | 2017-06-30 | 2017-11-07 | 武汉精测电子技术股份有限公司 | Picture compression method and defects of display panel detecting system in AOI detecting systems |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111598869B (en) | Method, equipment and storage medium for detecting Mura of display screen | |
| CN114494259A (en) | Cloth defect detection method based on artificial intelligence | |
| CN111612882B (en) | Image processing method, image processing device, computer storage medium and electronic equipment | |
| CN110736748A (en) | Immunohistochemical nuclear plasma staining section diagnosis method and system | |
| US11430130B2 (en) | Image processing method and computer-readable recording medium having recorded thereon image processing program | |
| CN108171689B (en) | Appraisal method, device and storage medium for retaking display screen image | |
| CN111753108B (en) | Presentation generation method, device, equipment and medium | |
| WO2024093287A1 (en) | Instrument identification method based on improved u2 network | |
| CN117252872B (en) | Mobile phone screen defect visual detection method and system | |
| CN113888477B (en) | Network model training method, metal surface defect detection method and electronic equipment | |
| WO2020087434A1 (en) | Method and device for evaluating resolution of face image | |
| CN115358943A (en) | Low-light image enhancement method, system, terminal and storage medium | |
| CN109215547A (en) | A kind of method and system of fast storage backlight inspection image | |
| CN105339985A (en) | Data interpolation and classification method for map data visualization | |
| CN115278217A (en) | Image picture detection method and device, electronic equipment and storage medium | |
| CN109584214A (en) | Image management method and system in a kind of inspection of backlight | |
| CN111985503B (en) | Target detection method and device based on improved feature pyramid network structure | |
| CN109166511B (en) | Compression method for compensating gauge of display device | |
| CN111898742A (en) | Method and equipment for monitoring training state of neural network model | |
| CN117333440A (en) | Power transmission and distribution line defect detection method, device, equipment, medium and program product | |
| CN112330611B (en) | FPGA-oriented high-parallelism light spot segmentation method | |
| CN115423716A (en) | Image enhancement method, device and equipment based on multidimensional filtering and storage medium | |
| Zhao et al. | Robust object tracking via sparse representation based on compressive collaborative Haar-like feature space | |
| Bu et al. | Multi-frame image restoration method for Luojia 1-01 night-light remote sensing images | |
| CN118212428B (en) | Landmark extraction method based on bit correlation, terminal equipment and storage medium |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190115 |