CN110020992A - Edge quality Identification Platform - Google Patents
Edge quality Identification Platform Download PDFInfo
- Publication number
- CN110020992A CN110020992A CN201811089033.2A CN201811089033A CN110020992A CN 110020992 A CN110020992 A CN 110020992A CN 201811089033 A CN201811089033 A CN 201811089033A CN 110020992 A CN110020992 A CN 110020992A
- Authority
- CN
- China
- Prior art keywords
- pixel
- value
- image
- glass edge
- connect
- 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
- 239000011521 glass Substances 0.000 claims abstract description 88
- 238000004445 quantitative analysis Methods 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims description 23
- 238000004458 analytical method Methods 0.000 claims description 14
- 239000000284 extract Substances 0.000 claims description 11
- 238000003860 storage Methods 0.000 claims description 10
- 238000011156 evaluation Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 9
- 238000012806 monitoring device Methods 0.000 claims description 9
- 230000005540 biological transmission Effects 0.000 claims description 6
- 230000003068 static effect Effects 0.000 claims description 3
- 238000001514 detection method Methods 0.000 description 6
- 238000000034 method Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000000151 deposition Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- NAPPWIFDUAHTRY-XYDRQXHOSA-N (8r,9s,10r,13s,14s,17r)-17-ethynyl-17-hydroxy-13-methyl-1,2,6,7,8,9,10,11,12,14,15,16-dodecahydrocyclopenta[a]phenanthren-3-one;(8r,9s,13s,14s,17r)-17-ethynyl-13-methyl-7,8,9,11,12,14,15,16-octahydro-6h-cyclopenta[a]phenanthrene-3,17-diol Chemical compound O=C1CC[C@@H]2[C@H]3CC[C@](C)([C@](CC4)(O)C#C)[C@@H]4[C@@H]3CCC2=C1.OC1=CC=C2[C@H]3CC[C@](C)([C@](CC4)(O)C#C)[C@@H]4[C@@H]3CCC2=C1 NAPPWIFDUAHTRY-XYDRQXHOSA-N 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000003795 desorption Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30168—Image quality inspection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of edge quality Identification Platforms, comprising: reverse light-source video camera carries out reverse light-source shooting towards glass edge, to obtain corresponding glass edge image, and exports the glass edge image;CSI output interface is connect with the reverse light-source video camera, for exporting the glass edge image;Data prediction structure is connect with the CSI output interface, for receiving the glass edge image, is executed data prediction to the glass edge image, to obtain corresponding pretreatment image, and is exported the pretreatment image;Quantitative analysis equipment, the quantity for calculating each problem pixel point occupy the ratio of the sum of the pixel of the pretreatment image, and when the ratio is more than preset ratio threshold value, issue unqualified control signal.Through the invention, the accurate recognition to glass edge quality is realized.
Description
Technical field
The present invention relates to glass identification field more particularly to a kind of edge quality Identification Platforms.
Background technique
Glass edge is extracted occupies the very raw status wanted with detection in image procossing, and the superiority and inferiority of algorithm directly affects
The performance developed.The quality of glass edge quality directly determines the quality of glass finished-product, and measures glass producer
One of the key factor of manufacture craft level, therefore, the detection of glass edge are that each glass producer must before glass factory
The standard of one of standby link, detection is also most stringent.
Various edge detection method is proposed.This method has their own characteristics, and is both also all to exist respectively
Limitation and shortcoming.
Summary of the invention
In order to solve to lack the technology to the effective mechanism that problem pixel point is counted in glass image in the prior art
Problem, the present invention provides a kind of edge quality Identification Platforms.
Wherein, the present invention needs to have following four key inventive point:
(1) reverse light-source Image Acquisition and detection are carried out to glass edge, to determine glass acceptable level;
(2) the blue color channels value also based on pixel and the blue color channels value of its surrounding pixel point calculate it and arrive
Each gradient value of surrounding all directions is examined using maximum value in each gradient value as reference gradient value according to reference gradient value
Survey problem pixel point;
(3) based on the noise profile situation to image, the ruler for the sliding window being filtered to described image is determined
It is very little;
(4) when being filtered to pixel, it is average to select to reach in its neighbouring pixel pixel value isolated noise
Multiple brightness values of pixel quantity are occupied as operation pixel value, arithmetic mean of instantaneous value meter is carried out to the multiple operation pixel value
It calculates to obtain the replacement pixel value of pixel processed.
According to an aspect of the present invention, a kind of edge quality Identification Platform is provided, the platform includes:
Reverse light-source video camera carries out reverse light-source shooting towards glass edge, to obtain corresponding glass edge image, and it is defeated
The glass edge image out;CSI output interface is connect with the reverse light-source video camera, is used for the glass edge image
Output;Data prediction structure is connect with the CSI output interface, for receiving the glass edge image, to the glass
Edge image executes data prediction, to obtain corresponding pretreatment image, and exports the pretreatment image;Gradient analysis is set
It is standby, it is connect with the data prediction structure, for receiving the pretreatment image, extracts each in the pretreatment image
The blue color channels value of pixel, execute following processing to each pixel: the Blue based on the pixel is logical
Road value and the blue color channels value of its surrounding pixel point calculate its each gradient value for arriving surrounding all directions, by each gradient
Maximum value is exported as reference gradient value in value;Pixel extract equipment is connect with the gradient analysis equipment, and being used for will be described
Reference gradient value is more than the pixel of predetermined gradient threshold value as problem pixel point, the output pretreatment figure in pretreatment image
Each problem pixel point as in;Quantitative analysis equipment is connect with the pixel extract equipment, for calculating described each ask
The quantity of topic pixel occupies the ratio of the sum of the pixel of the pretreatment image, and is more than preset ratio in the ratio
When threshold value, unqualified control signal is issued;Wherein, the quantitative analysis equipment is also used to be less than preset ratio in the ratio
When threshold value, qualified control signal is issued.
More specifically, in the edge quality Identification Platform, further includes:
Static storage device is connect, for storing respectively with the quantitative analysis equipment and the pixel extract equipment
The predetermined gradient threshold value and the preset ratio threshold value;Wherein, chip is handled using DSP to realize that the gradient analysis is set
It is standby, the DSP processing built-in chip type memory and timer.
More specifically, in the edge quality Identification Platform: the data prediction structure includes noise monitoring device,
It is connect with the CSI output interface, for receiving the glass edge image, obtains each orphan in the glass edge image
Noise is found, and determines the quantity for the pixel that it is occupied in the glass edge image based on each isolated noise;It is described
Data prediction structure includes noise evaluation equipment, is connect with the noise monitoring device, for receiving each isolated noise point
The quantity for the pixel not occupied carries out mean value computation to each quantity, to obtain corresponding number to obtain each quantity
Measure mean value.
More specifically, in the edge quality Identification Platform: the data prediction structure includes window-setting equipment,
It is connect with the noise monitoring device, for receiving the quantity for the pixel that each isolated noise occupies respectively, based on described each
The quantity for the pixel that a isolated noise occupies respectively determines the sliding window being filtered to the glass edge image
Size;The data prediction structure includes MMC storage equipment, is connect with the window-setting equipment, for receiving and depositing
Store up the size of the sliding window.
More specifically, in the edge quality Identification Platform: the data prediction structure includes parameter extraction equipment,
Equipment is stored with the MMC respectively and the noise evaluation equipment is connect, for receiving the number average value and the sliding window
Mouthful, and following operation is executed to each of glass edge image pixel: it will be every in the glass edge image
One pixel obtains the sliding centered on the subject image vegetarian refreshments as subject image vegetarian refreshments in the glass edge image
Each pixel in window carries out each brightness value of each reference image vegetarian refreshments from big as each reference image vegetarian refreshments
Sort to small sequence, using intermediate serial number, reach multiple brightness values of the number average value quantity as operation pixel value, it is right
The multiple operation pixel value carries out arithmetic mean of instantaneous value calculating, to obtain the replacement pixel value of the subject image vegetarian refreshments;The number
Data preprocess structure includes data output apparatus, is connect respectively with the gradient analysis equipment and the parameter extraction equipment, is used
In each replacement pixel value for receiving each pixel in the glass edge image, and based in the glass edge image
Each replacement pixel value of each pixel obtain the corresponding replacement image of the glass edge image, and to the gradient point
Desorption device exports the corresponding pretreatment image of the glass edge image.
More specifically, in the edge quality Identification Platform: the parameter extraction equipment includes data receipt unit, number
According to transmission unit and pixel value processing unit, the data receipt unit is for receiving the number average value and the sliding window
Mouthful, the pixel value processing unit is connect with the data receipt unit, and the data transmission unit and the pixel value are handled
Unit connection.
More specifically, in the edge quality Identification Platform: the pixel value processing unit is used for the glass edge
Each of edge image pixel executes following operation: using each of glass edge image pixel as object
Pixel obtains each pixel in the sliding window centered on the subject image vegetarian refreshments in the glass edge image
As each reference image vegetarian refreshments, each brightness value of each reference image vegetarian refreshments is subjected to sequence from big to small and is sorted, it will
Intermediate serial number, reach multiple brightness values of the number average value quantity as operation pixel value, to the multiple operation pixel
Value carries out arithmetic mean of instantaneous value calculating, to obtain the replacement pixel value of the subject image vegetarian refreshments.
More specifically, in the edge quality Identification Platform: MMC storage equipment also with the noise evaluation equipment
Connection, for receiving and storing the number average value.
Detailed description of the invention
Embodiment of the present invention is described below with reference to attached drawing, in which:
Fig. 1 is the structural schematic diagram according to the edge quality Identification Platform shown in embodiment of the present invention.
Specific embodiment
The embodiment of edge quality Identification Platform of the invention is described in detail below with reference to accompanying drawings.
In glass edge type, common edge type has step change type, ramp type, linear type and four kinds of roof type.
In specific edge detection, the detection mode of center is to be imaged in picture based on glass, oblique in gray scale
The beginning and end on slope, the first derivative equal road You Yigejie are constant at slope, remaining is zero;Second dervative is in slope starting point
A upward pulse is generated, generates down pulse in terminal, remaining is zero.Therefore slope can be determined by first derivative
Type edge, and the zero crossing at edge, the i.e. center at edge can be determined by passing through second dervative.
In order to overcome above-mentioned deficiency, the present invention has built a kind of edge quality Identification Platform, can effectively solve the problem that corresponding
Technical problem.
Fig. 1 is the structural schematic diagram according to the edge quality Identification Platform shown in embodiment of the present invention, the platform packet
It includes:
Reverse light-source video camera carries out reverse light-source shooting towards glass edge 1, to obtain corresponding glass edge image, and
Export the glass edge image;
Top clamping plate 2, lower section clamping plate 3 and fixed frame 4, the top clamping plate are arranged above glass, the lower section folder
Plate is arranged below glass, the top clamping plate and the lower section clamping plate provided for fixing glass glass edge carry out it is reflective
Source shooting, fixed frame is for fixing the top clamping plate and the lower section clamping plate;
CSI output interface is connect with the reverse light-source video camera, for exporting the glass edge image;
Data prediction structure is connect with the CSI output interface, for receiving the glass edge image, to described
Glass edge image executes data prediction, to obtain corresponding pretreatment image, and exports the pretreatment image;
Gradient analysis equipment is connect with the data prediction structure, for receiving the pretreatment image, described in extraction
The blue color channels value of each pixel in pretreatment image executes following processing to each pixel: based on described
The blue color channels value of pixel and the blue color channels value of its surrounding pixel point calculate it to each of surrounding all directions
A gradient value is exported maximum value in each gradient value as reference gradient value;
Pixel extract equipment is connect with the gradient analysis equipment, is used for reference gradient in the pretreatment image
Value is more than the pixel of predetermined gradient threshold value as problem pixel point, exports each problem pixel in the pretreatment image
Point;
Quantitative analysis equipment is connect with the pixel extract equipment, for calculating the number of each problem pixel point
Amount occupies the ratio of the sum of the pixel of the pretreatment image, and when the ratio is more than preset ratio threshold value, issues
Unqualified control signal;
Wherein, the quantitative analysis equipment is also used to when the ratio is less than preset ratio threshold value, issues qualified control
Signal processed.
Then, continue that the specific structure of edge quality Identification Platform of the invention is further detailed.
In the edge quality Identification Platform, further includes:
Static storage device is connect, for storing respectively with the quantitative analysis equipment and the pixel extract equipment
The predetermined gradient threshold value and the preset ratio threshold value;
Wherein, chip is handled using DSP to realize the gradient analysis equipment, the DSP handles built-in chip type memory
And timer.
In the edge quality Identification Platform: the data prediction structure includes noise monitoring device, with the CSI
Output interface connection obtains each isolated noise in the glass edge image for receiving the glass edge image, and
The quantity for the pixel that it is occupied in the glass edge image is determined based on each isolated noise;
The data prediction structure includes noise evaluation equipment, is connect with the noise monitoring device, each for receiving
The quantity for the pixel that a isolated noise occupies respectively carries out mean value computation to each quantity to obtain each quantity, with
Obtain corresponding number average value.
In the edge quality Identification Platform: the data prediction structure includes window-setting equipment, is made an uproar with described
The connection of sound monitoring device is made an uproar for receiving the quantity for the pixel that each isolated noise occupies respectively based on each isolate
The quantity for the pixel that sound occupies respectively determines the size for the sliding window being filtered to the glass edge image;
The data prediction structure includes MMC storage equipment, is connect with the window-setting equipment, for receiving and depositing
Store up the size of the sliding window.
In the edge quality Identification Platform: the data prediction structure includes parameter extraction equipment, respectively with institute
It states MMC storage equipment to connect with the noise evaluation equipment, for receiving the number average value and the sliding window, and to institute
It states each of glass edge image pixel and executes following operation: by each of glass edge image pixel
As subject image vegetarian refreshments, obtained in the glass edge image each in the sliding window centered on the subject image vegetarian refreshments
A pixel carries out sequence from big to small as each reference image vegetarian refreshments, by each brightness value of each reference image vegetarian refreshments
Sequence, using intermediate serial number, reach multiple brightness values of the number average value quantity as operation pixel value, to the multiple fortune
It calculates pixel value and carries out arithmetic mean of instantaneous value calculating, to obtain the replacement pixel value of the subject image vegetarian refreshments;
The data prediction structure includes data output apparatus, is mentioned respectively with the gradient analysis equipment and the parameter
Equipment is taken to connect, for receiving each replacement pixel value of each pixel in the glass edge image, and based on described
Each replacement pixel value of each pixel in glass edge image obtains the corresponding replacement image of the glass edge image,
And the corresponding pretreatment image of the glass edge image is exported to the gradient analysis equipment.
In the edge quality Identification Platform: the parameter extraction equipment includes data receipt unit, data transmission list
Member and pixel value processing unit, the data receipt unit is for receiving the number average value and the sliding window, the picture
Element value processing unit is connect with the data receipt unit, and the data transmission unit is connect with the pixel value processing unit.
In the edge quality Identification Platform: the pixel value processing unit is used for in the glass edge image
Each pixel executes following operation: using each of glass edge image pixel as subject image vegetarian refreshments,
Each pixel in sliding window using centered on the subject image vegetarian refreshments is obtained in the glass edge image as each
Each brightness value of each reference image vegetarian refreshments is carried out sequence from big to small and sorted, by intermediate serial number by reference image vegetarian refreshments
, multiple brightness values that reach the number average value quantity as operation pixel value, the multiple operation pixel value is calculated
Art mean value calculation, to obtain the replacement pixel value of the subject image vegetarian refreshments.
In the edge quality Identification Platform: the MMC storage equipment is also connect with the noise evaluation equipment, is used for
Receive and store the number average value.
In addition, the Harvard structure that the inside of DSP processing chip is separated using program and data, has special hardware multiplication
Pile line operation is widely used in device, provides special DSP instruction, can be used to quickly realize that various Digital Signal Processing are calculated
Method.
According to the requirement of Digital Signal Processing, DSP processing chip generally has following some main features: (1) one
An achievable multiplication and a sub-addition in a instruction cycle.(2) program and data space is separated, can simultaneously access instruction and
Data.(3) there is quick RAM in piece, can usually be accessed simultaneously in two pieces by independent data/address bus.(4) there is low open
It sells or without overhead loop and the hardware supported jumped.(5) quickly interrupt processing and Hardware I/O are supported.(6) have in the monocycle
Multiple hardware address generators of interior operation.(7) multiple operations can be executed parallel.(8) support pile line operation, make fetching,
The operations such as decoding and execution can be with Overlapped Execution.
The data format of chip operation is handled according to DSP to classify.Data handle chip with the DSP that fixed point format works
Referred to as fixed DSP handles chip, such as TMS320C1X/C2X, TMS320C2XX/C5X, TMS320C54X/C62XX system of TI company
Column, the ADSP21XX series of AD company, the DSP16/16A of AT & T Corp., the MC56000 etc. of Motolora company.With floating-point lattice
The Floating-point DSP that is known as of formula work handles chip, such as the TMS320C3X/C4X/C8X of TI company, the ADSP21XXX system of AD company
Column, the DSP32/32C of AT & T Corp., the MC96002 etc. of Motolora company.
Not exclusively, some DSP processing chips use to be made by oneself floating-point format used by different Floating-point DSP processing chips
The floating-point format of justice, such as TMS320C3X, and some DSP processing chip then uses the standard floating-point format of IEEE, such as Motorola
The ZR35325 etc. of the MB86232 and ZORAN company of MC96002, FUJITSU company of company.
Using edge quality Identification Platform of the invention, for lacking in the prior art to problem pixel point in glass image
The technical issues of effective mechanism counted, by carrying out reverse light-source Image Acquisition and detection to glass edge, to determine glass
Glass acceptable level;The blue color channels value of blue color channels value and its surrounding pixel point also based on pixel calculates it and arrives
Each gradient value of surrounding all directions is examined using maximum value in each gradient value as reference gradient value according to reference gradient value
Survey problem pixel point;Based on the noise profile situation to image, the sliding window being filtered to described image is determined
Size;When being filtered pixel, select to reach isolated noise averaged occupation picture in its neighbouring pixel pixel value
Multiple brightness values of vegetarian refreshments quantity carry out arithmetic mean of instantaneous value to the multiple operation pixel value and calculate to obtain as operation pixel value
Obtain the replacement pixel value of pixel processed;To solve above-mentioned technical problem.
It is understood that although the present invention has been disclosed in the preferred embodiments as above, above-described embodiment not to
Limit the present invention.For any person skilled in the art, without departing from the scope of the technical proposal of the invention,
Many possible changes and modifications all are made to technical solution of the present invention using the technology contents of the disclosure above, or are revised as
With the equivalent embodiment of variation.Therefore, anything that does not depart from the technical scheme of the invention are right according to the technical essence of the invention
Any simple modifications, equivalents, and modifications made for any of the above embodiments still fall within the range of technical solution of the present invention protection
It is interior.
Claims (8)
1. a kind of edge quality Identification Platform, which is characterized in that the platform includes:
Reverse light-source video camera carries out reverse light-source shooting towards glass edge, to obtain corresponding glass edge image, and exports institute
State glass edge image;
CSI output interface is connect with the reverse light-source video camera, for exporting the glass edge image;
Data prediction structure is connect with the CSI output interface, for receiving the glass edge image, to the glass
Edge image executes data prediction, to obtain corresponding pretreatment image, and exports the pretreatment image;
Gradient analysis equipment is connect with the data prediction structure, for receiving the pretreatment image, extracts the pre- place
The blue color channels value for managing each pixel in image executes following processing to each pixel: being based on the pixel
The blue color channels value of point and the blue color channels value of its surrounding pixel point calculate its each ladder for arriving surrounding all directions
Angle value is exported maximum value in each gradient value as reference gradient value;
Pixel extract equipment is connect with the gradient analysis equipment, for surpassing reference gradient value in the pretreatment image
The pixel of predetermined gradient threshold value is crossed as problem pixel point, exports each problem pixel point in the pretreatment image;
Quantitative analysis equipment is connect with the pixel extract equipment, and the quantity for calculating each problem pixel point accounts for
According to the ratio of the sum of the pixel of the pretreatment image, and when the ratio is more than preset ratio threshold value, sending does not conform to
Lattice control signal;
Wherein, the quantitative analysis equipment is also used to when the ratio is less than preset ratio threshold value, issues qualified control letter
Number.
2. edge quality Identification Platform as described in claim 1, which is characterized in that further include:
Static storage device is connect with the quantitative analysis equipment and the pixel extract equipment respectively, described for storing
Predetermined gradient threshold value and the preset ratio threshold value;
Wherein, chip is handled using DSP to realize the gradient analysis equipment, the DSP handles built-in chip type memory and determines
When device.
3. edge quality Identification Platform as claimed in claim 2, it is characterised in that:
The data prediction structure includes noise monitoring device, is connect with the CSI output interface, for receiving the glass
Edge image obtains each isolated noise in the glass edge image, and determines it in institute based on each isolated noise
State the quantity of the pixel occupied in glass edge image;
The data prediction structure includes noise evaluation equipment, is connect with the noise monitoring device, for receiving each orphan
The quantity for the pixel that vertical noise occupies respectively carries out mean value computation to each quantity, to obtain to obtain each quantity
Corresponding number average value.
4. edge quality Identification Platform as claimed in claim 3, it is characterised in that:
The data prediction structure includes window-setting equipment, is connect with the noise monitoring device, for receiving each orphan
The quantity for the pixel that vertical noise occupies respectively, based on determining pair of quantity of the pixel that each isolated noise occupies respectively
The size for the sliding window that the glass edge image is filtered;
The data prediction structure includes MMC storage equipment, is connect with the window-setting equipment, for receiving and storing
State the size of sliding window.
5. edge quality Identification Platform as claimed in claim 4, it is characterised in that:
The data prediction structure includes parameter extraction equipment, stores equipment with the MMC respectively and the noise evaluation is set
Standby connection, for receiving the number average value and the sliding window, and to each of glass edge image pixel
Select the following operation of execution: using each of glass edge image pixel as subject image vegetarian refreshments, in the glass edge
Each pixel in sliding window using centered on the subject image vegetarian refreshments is obtained in edge image as each reference image vegetarian refreshments,
Each brightness value of each reference image vegetarian refreshments is subjected to sequence sequence from big to small, by intermediate serial number, reach described
Multiple brightness values of number average value quantity carry out arithmetic mean of instantaneous value meter as operation pixel value, to the multiple operation pixel value
It calculates, to obtain the replacement pixel value of the subject image vegetarian refreshments;
The data prediction structure includes data output apparatus, is set respectively with the gradient analysis equipment and the parameter extraction
Standby connection for receiving each replacement pixel value of each pixel in the glass edge image, and is based on the glass
The corresponding replacement image of each replacement pixel value acquisition glass edge image of each pixel in edge image, and to
The gradient analysis equipment exports the corresponding pretreatment image of the glass edge image.
6. edge quality Identification Platform as claimed in claim 5, it is characterised in that:
The parameter extraction equipment includes data receipt unit, data transmission unit and pixel value processing unit, and the data connect
Unit is received for receiving the number average value and the sliding window, the pixel value processing unit and the data receipt unit
Connection, the data transmission unit are connect with the pixel value processing unit.
7. edge quality Identification Platform as claimed in claim 6, it is characterised in that:
The pixel value processing unit is used to execute each of glass edge image pixel following operation: by institute
Each of glass edge image pixel is stated as subject image vegetarian refreshments, is obtained in the glass edge image with described right
As each pixel in the sliding window centered on pixel is as each reference image vegetarian refreshments, by each reference image vegetarian refreshments
Each brightness value carry out from big to small sequence sequence, by intermediate serial number, reach the number average value quantity it is multiple bright
Angle value carries out arithmetic mean of instantaneous value calculating as operation pixel value, to the multiple operation pixel value, to obtain the object pixel
The replacement pixel value of point.
8. edge quality Identification Platform as claimed in claim 7, it is characterised in that:
The MMC storage equipment is also connect with the noise evaluation equipment, for receiving and storing the number average value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811089033.2A CN110020992A (en) | 2018-09-18 | 2018-09-18 | Edge quality Identification Platform |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811089033.2A CN110020992A (en) | 2018-09-18 | 2018-09-18 | Edge quality Identification Platform |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110020992A true CN110020992A (en) | 2019-07-16 |
Family
ID=67188479
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811089033.2A Pending CN110020992A (en) | 2018-09-18 | 2018-09-18 | Edge quality Identification Platform |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110020992A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1573319A (en) * | 2003-05-31 | 2005-02-02 | 三星康宁精密琉璃株式会社 | Internal flaw position detecting method in depth direction for glass base plate |
CN204203126U (en) * | 2014-11-25 | 2015-03-11 | 焦作飞鸿安全玻璃有限公司 | A kind of glass monitor station |
CN105044124A (en) * | 2015-08-27 | 2015-11-11 | 李明英 | Glass flaw classification device based on gray mean value analysis |
CN105181705A (en) * | 2015-08-28 | 2015-12-23 | 白薇 | Ceramic appearance analysis method based on multiple-filter |
-
2018
- 2018-09-18 CN CN201811089033.2A patent/CN110020992A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1573319A (en) * | 2003-05-31 | 2005-02-02 | 三星康宁精密琉璃株式会社 | Internal flaw position detecting method in depth direction for glass base plate |
CN204203126U (en) * | 2014-11-25 | 2015-03-11 | 焦作飞鸿安全玻璃有限公司 | A kind of glass monitor station |
CN105044124A (en) * | 2015-08-27 | 2015-11-11 | 李明英 | Glass flaw classification device based on gray mean value analysis |
CN105181705A (en) * | 2015-08-28 | 2015-12-23 | 白薇 | Ceramic appearance analysis method based on multiple-filter |
Non-Patent Citations (2)
Title |
---|
王一丁等: "《数字图像处理》", 31 August 2015, 西安电子科技大学出版社 * |
章鲁等: "《医学图像处理与分析》", 31 August 2006, 上海科学技术出版社 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105974211A (en) | Method and device for testing equipment antenna | |
CN110443212B (en) | Positive sample acquisition method, device, equipment and storage medium for target detection | |
CN104573625A (en) | Facial detection | |
CN110033481A (en) | Method and apparatus for carrying out image procossing | |
CN108596109B (en) | Target detection method and device based on neural network and motion vector | |
CN107909027A (en) | It is a kind of that there is the quick human body target detection method for blocking processing | |
CN107071736B (en) | Shared bicycle data acquisition platform | |
CN107808126A (en) | Vehicle retrieval method and device | |
CN103632247A (en) | Meter information identification system and method based on intelligent storage system | |
CN110390229A (en) | A kind of face picture screening technique, device, electronic equipment and storage medium | |
CN107085229A (en) | Collecting method | |
CN110399822A (en) | Action identification method of raising one's hand, device and storage medium based on deep learning | |
CN115620066B (en) | Article detection method and device based on X-ray image and electronic equipment | |
CN111126112B (en) | Candidate region determination method and device | |
CN110020992A (en) | Edge quality Identification Platform | |
CN113033545B (en) | Empty tray identification method and device | |
CN110099237A (en) | Image processing method, electronic device and computer readable storage medium | |
CN107907546A (en) | A kind of defects of vision detection method and system based on intelligent mobile terminal | |
CN117573477A (en) | Abnormal data monitoring method, device, equipment, medium and program product | |
CN104598921A (en) | Video preview selecting method and device | |
CN112037255A (en) | Target tracking method and device | |
CN115527069A (en) | Article identification and article identification system construction method and apparatus | |
CN115050000A (en) | Running scene recognition method and device, computer equipment and storage medium | |
CN109472772A (en) | Image smear detection method, device and equipment | |
CN106535130A (en) | Method, device and equipment for judging indoor and outdoor affiliation of mobile terminal 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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190716 |
|
WD01 | Invention patent application deemed withdrawn after publication |