CN108257172A - Integrated circuit diagram open circuit critical area extracting method based on Hadoop - Google Patents
Integrated circuit diagram open circuit critical area extracting method based on Hadoop Download PDFInfo
- Publication number
- CN108257172A CN108257172A CN201810034620.5A CN201810034620A CN108257172A CN 108257172 A CN108257172 A CN 108257172A CN 201810034620 A CN201810034620 A CN 201810034620A CN 108257172 A CN108257172 A CN 108257172A
- Authority
- CN
- China
- Prior art keywords
- image
- domain image
- gauze
- value
- edge
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/13—File access structures, e.g. distributed indices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/16—File or folder operations, e.g. details of user interfaces specifically adapted to file systems
- G06F16/164—File meta data generation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/18—File system types
- G06F16/182—Distributed file systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/39—Circuit design at the physical level
- G06F30/392—Floor-planning or layout, e.g. partitioning or placement
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/39—Circuit design at the physical level
- G06F30/398—Design verification or optimisation, e.g. using design rule check [DRC], layout versus schematics [LVS] or finite element methods [FEM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- 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
- G06T2207/30108—Industrial image inspection
- G06T2207/30148—Semiconductor; IC; Wafer
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Databases & Information Systems (AREA)
- Geometry (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Architecture (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The present invention discloses a kind of integrated circuit diagram open circuit critical area extracting method based on Hadoop, realizes that step is：(1) the domain image of integrated circuit is read；(2) the domain image of integrated circuit is uploaded；(3) piecemeal storage integrated circuit diagram image；(4) by the domain image slices of back end；(5) input key-value pair of the domain image into mapping class Map is converted；(6) domain image is pre-processed；(7) extraction gauze edge；(8) gauze edge is expanded；(9) open circuit critical area is calculated；(10) output open circuit critical area；(11) setting abbreviation class Reduce；(12) task is submitted.The present invention carries out integrated circuit diagram parallel open circuit critical area using distributed treatment frame Haoop and extracts, and can complete large scale integrated circuit domain open circuit critical area rapid extraction, can improve the open circuit critical area extraction efficiency of integrated circuit diagram.
Description
Technical field
The invention belongs to field of computer technology, further relate to integrated circuit and computer distribution type data processing skill
A kind of open circuit of large scale integrated circuit domain image gauze based on distributed treatment frame Hadoop in art field is crucial
Area extraction method.Application distribution formula file system HDFS (Hadoop Distributed File System) of the present invention is to defeated
Enter bitmap domain carry out it is distributed storage and application mapping-abbreviation (MapReduce) framework contraposition plate figure in gauze into
Row open circuit critical area extraction operation, available for expeditiously extracting the open circuit critical area of gauze in extensive bitmap domain.
Background technology
The random defect occurred in ic manufacturing process can cause the open fault of circuit, so as to reduce integrated circuit
The yield rate of chip.Open circuit critical area refers in integrated circuit the key of easily generation circuit open fault due to random defect
Region area, if there is random defect in these regions, circuit just necessarily will appear open fault.Critical area of opening a way is pair
One of domain open fault caused by random defect quantization, thus the distribution for critical area of opening a way and size to improve chip into
Product rate has vital effect.Open circuit critical area extracting method based on image processing techniques be be likely to occur with
Machine defect is structural element, calculates the open circuit critical area of integrated circuit diagram image using the method for mathematical morphology, still
Operation time complexity is big, and efficiency is low, it is impossible to be extracted suitable for the open circuit critical area of large scale integrated circuit.
Patented technology " a kind of method of rapid extraction critical area of layout " (number of patent application that Zhejiang University possesses at it
201010108651.4 Authorization Notice No. 101789048B) in disclose it is a kind of extraction critical area of layout method.This method
Realizing step is, 1. extraction layout informations classify to the fundamental figure unit of integrated circuit diagram；2. it is orderly to establish piecemeal
Multiple index table；3. it is layered traversal domain tree using the orderly multiple index table of piecemeal；4. extract all and defect polygon
The figure of coincidence；5. calculate critical area of layout.It is to establish Multilevel Block Ordered indices table to need in place of this method Shortcomings
A large amount of memory headroom and operand are wanted, with the increase of domain scale, this method consumes the resources such as machine memory and CPU very much, needs
The computer for wanting hardware configuration very high could stablize extraction.
(2012, Xi'an was electric for the paper " the open circuit critical area extraction based on image processing techniques " that Wang Le is delivered at it
Sub- University of Science and Technology's Master's thesis) in propose it is a kind of extract domain open circuit critical area method.The basic thought of this method is
The bitmap containing domain gauze information is read in, gauze topological path curve is obtained by refining operation in mathematical morphology；By
The line end of the borderline topological path of hit or miss transform identification line net in mathematical morphology obtains gauze and flows to axis, into
And it extracts gauze and flows to side.Side is flowed to gauze to carry out dilation operation and be overlapped expansion results；Based on mathematical morphology
Set operation extraction superposition after overlapping region, calculate open circuit key area area, that is, critical area of opening a way.This method is deposited
Shortcoming be that the corresponding gauze position in defect center is done as unit of a grid of domain gauze in expansion process
Repetitive operation, with the expansion of designated critical net scale, open circuit critical area extraction working efficiency is not high.
Invention content
It is a kind of based on distributed treatment frame it is an object of the invention in view of the above shortcomings of the prior art, propose
The integrated circuit diagram open circuit critical area extracting method of Hadoop, this method can improve large scale integrated circuit domain center line
The efficiency of the open circuit critical area extraction of net solves the problems, such as that efficiency is low when unit serially extracts open circuit critical area.
Realize that the object of the invention is as follows：
(1) the domain image of integrated circuit is read：
(1a) reads the whole in the integrated circuit diagram of the open circuit critical area to be extracted of normative document image BMP forms
Domain image；
Each width domain image of reading is saved as customized form X_Y_Z.bmp by (1b)；
(2) the domain image of integrated circuit is uploaded to distributed treatment frame Hadoop：
(2a) starts distributed treatment using the startup order start-all.sh of distributed treatment frame Hadoop clusters
Frame Hadoop clusters；
(2b) uploads to the domain image for being fully integrated circuit of reading in distributed file system HDFS；
(3) under distributed treatment frame Hadoop piecemeal storage integrated circuit domain image：
(3a) distributed file system HDFS carries out piecemeal to uploading to the domain image in the system；
(3b) fifty-fifty domain after memory partitioning in each back end of distributed treatment frame Hadoop clusters
Image；
(4) by the domain image slices of distributed treatment frame Hadoop back end：
(4a) randomly selects a back end from distributed treatment frame Hadoop clusters, is inputted using composition file
Domain image in selected back end is divided into fragment size as 64M by form CombineFileInputFormat
Combination fragment CombineFileSplit image data sets；
(4b) is concentrated from combination fragment CombineFileSplit image datas, randomly selects a combination fragment；
(5) input key-value pair key1/value1 of the domain image into mapping class Map is converted：
(5a) obtains selected combination point by the routing information function getPath of distributed treatment frame Hadoop
The routing information of domain image in piece；
(5b) using the image pixel data transfer function cvDecodeImage in the JavaCV of image procossing library, by path
Corresponding domain image is converted to the picture number of image type ImageWritable in distributed treatment frame Hadoop in information
According to by the filename of domain image in routing information, as the key key1 of key-value pair in mapping class Map, by the number of domain image
According to as the corresponding value1 of key key1；
(6) domain image is pre-processed：
The corresponding value1 of key key1 are converted to the version in the JavaCV picture formats IplImage of image procossing library by (6a)
Figure image data；
(6b) calculates the gray scale of each pixel in the domain image of IplImage forms using gray value calculation formula
Value by the gray value of all pixels point after calculating, forms the domain image after gray processing；
(6c) calculates the global threshold of domain image after gray processing using maximum variance between clusters；
(6d) calculates the two-value of domain image slices vegetarian refreshments after each gray processing, by the institute of calculating using two-value calculation formula
There is the two-value of pixel, form the domain image after binaryzation；
(7) domain image line network edge is extracted：
(7a) is carried out edge detection to the domain image after binaryzation, is obtained domain figure using method for detecting image edge
As center line network edge；
(7b) serial number successively since the row coordinate at the gauze edge in domain image, obtains connected region and gauze
The sum at edge；
(8) the horizontal direction edge of domain image gauze and vertical direction edge are extracted：
Define respectively the element value of one 1 × 3 be all 1 matrix and the element value of one 3 × 1 be all 1 matrix, with 1
The matrix that × 3 element value is all 1 is structural element, carries out etching operation to the edge of current number gauze, is currently compiled
The horizontal direction edge of number gauze is all 1 matrix as structural element, to the edge of current number gauze using 3 × 1 element value
Etching operation is carried out, obtains the vertical direction edge of current number gauze；
(9) expansive working is carried out to domain image line network edge：
(9a) defines one according to transformation rule the defects of can result in integrated circuit diagram gauze open circuit on matrix
Defect matrix DD1×D2, wherein, D1Represent the row of defect matrix, D2Represent defect matrix column；
(9b) defines a horizontal direction edge image matrix and vertical direction edge image matrix respectively, and according to gauze
The number order at edge extracts the horizontal direction edge of gauze and vertical direction edge, is stored respectively in horizontal direction side successively
In edge image array and vertical direction edge image matrix, by the head of two horizontal direction edge image matrixes of current number gauze
The abscissa of element subtracts each other, using difference as the minimum range at current number gauze horizontal direction edge；By current number gauze
The abscissa of the header element of two vertical direction edge image matrixes subtracts each other, using difference as current number gauze vertical direction side
The minimum range of edge；
(9c) judges whether the minimum range at current number gauze horizontal direction edge is more than the sum of defect row matrix, if
It is then to perform step (9e), otherwise, performs step (9d)；
(9d) carries out mathematical morphology using defect matrix as structural element, to the horizontal direction edge of current number gauze
Expansive working；
(9e) judges whether the minimum range at the vertical direction edge of current number gauze is more than the sum of the row of defect, if
It is then to perform step (9g), otherwise, performs step (9f)；
(9f) carries out the expansion of mathematical morphology using defect matrix as structural element, to the gauze vertical edge of current number
Operation；
(9g) is square vertically by the image pixel value after current number gauze horizontal direction edge swell and current number gauze
Pixel value after edge swell, which corresponds to, to be added；
(9h) defines the image storage matrix of a processed image matrix form cvMat, and storage current number gauze is horizontal
Image pixel value after the edge swell of direction is corresponding with the pixel value after current number gauze vertical direction edge swell to be added
As a result；
(10) judge whether the number at current gauze edge is equal with the sum at gauze edge, if so, performing step
(10), step (8) is performed after the number at gauze edge otherwise, is added 1；
(11) open circuit critical area is calculated：
The area in region of the pixel value more than 1 in the image storage matrix of storage gauze edge swell result is calculated, as
Open circuit critical area；
(12) judge whether the domain image in present combination fragment all extracts open circuit critical area, if so, holding
Row step (13) otherwise, performs step (5)；
(13) judge whether the combination fragment in the back end currently chosen all extracts open circuit critical area, if
It is then to perform step (14), otherwise, performs step (4)；
(14) pass through distributed treatment frame Hadoop output open circuit critical areas：
Using the filename of domain image in routing information as the key key2 of mapping class Map output key-value pairs, by extraction
Open a way key key2 corresponding value2 of the critical area as text type Text；
(15) the abbreviation class Reduce of setting distributed treatment frame Hadoop：
(15a) summarizes the output key-value pair key2/value2 of the mapping class Map in total data node；
Key key2 is output in the glue file folder for having set outgoing route by (15b)；
The open circuit critical area for calculating gained with the value2 in key-value pair key2/value2, is output to distribution by (15c)
In the open circuit critical area file of formula file system HDFS；
(16) task of extraction open circuit critical area is submitted：
The task Job of back end in (16a) initialization distributed treatment frame Hadoop clusters；
The extracting open circuit critical area of the task is submitted in distributed treatment frame Hadoop clusters by (16b).
The present invention has the following advantages compared with prior art：
1st, due to the domain image of present invention piecemeal storage integrated circuit under distributed treatment frame Hadoop, will divide
Cloth handles the domain image slices of frame Hadoop back end, converts input key-value pair of the domain image into mapping class Map
Key1/value1, by the filename of domain image in routing information, as the key key1 of key-value pair in mapping class Map, by domain
The data of image calculate open circuit critical area, overcome the prior art and consume very much machine memory as the corresponding value1 of key key1
The problem of extracting open circuit critical area could be stablized with resources, the computers for needing hardware configuration very high such as CPU so that the present invention
With the distributed treatment frame distributed treatment frame Hadoop clusters built using multiple common bare machines, with regard to energy
The advantages of completing the open circuit critical area extraction work of large scale integrated circuit gauze.
2nd, due to the present invention extraction horizontal direction edge of domain image gauze and vertical direction edge, judge current compile
Whether the minimum range at number gauze horizontal direction edge is more than the sum of defect row matrix, judges the vertical side of current number gauze
Whether extrorse minimum range is more than the sum of the row of defect, using defect matrix as structural element, to the gauze of current number
Horizontal direction edge carries out the expansive working of mathematical morphology, and Mathematical Morphology is carried out to the gauze vertical direction edge of current number
Expansive working is overcome in the prior art with the expansion of designated critical net scale, open circuit critical area extraction work
The problem of inefficient so that the present invention improves the efficiency of extraction large scale integrated circuit gauze open circuit critical area.
Description of the drawings
Fig. 1 is the flow chart of the present invention.
Fig. 2 is the analogous diagram of the present invention.
Specific embodiment
The present invention will be further described below in conjunction with the accompanying drawings.
With reference to Fig. 1, the specific implementation step of the present invention is further described.
Step 1, the domain image of integrated circuit is read.
Read in whole domains in the integrated circuit diagram of the short-circuit critical area to be extracted of normative document image BMP forms
Image.
Each width domain image of reading is saved as into customized form X_Y_Z.bmp.
The number of plies according to domain chooses each layer of domain image from top to down, in user-defined format X_Y_Z.bmp
X represent, each layer of domain image is divided into the mark that multiple pixel value sizes are 900*900 according to sequence from left to right
Quasi- document image form BMP images represent the normative document picture format BMP images of each width 900*900 in integrated electricity with Y
The value of the initial row of position in the domain image of road represents the normative document picture format BMP of each width 900*900 with Z
The value of image starting row of position in integrated circuit diagram image.
Step 2, the domain image of integrated circuit is uploaded to distributed treatment frame Hadoop.
Using the startup order start-all.sh of distributed treatment frame Hadoop clusters, start distributed treatment frame
Hadoop clusters.
The domain image for being fully integrated circuit of reading is uploaded in distributed file system HDFS.
Step 3, under distributed treatment frame Hadoop piecemeal storage integrated circuit domain image.
Distributed file system HDFS carries out piecemeal to uploading to the domain image in the system.
The fifty-fifty domain image after memory partitioning in each back end of distributed treatment frame Hadoop clusters.
Step 4, fragment is carried out to the domain image of back end each in distributed treatment frame Hadoop.
A back end is randomly selected from distributed treatment frame Hadoop clusters, uses composition file input format
Domain image in selected back end is divided into the group that fragment size is 64M by CombineFileInputFormat
Close fragment CombineFileSplit image data sets.
It is concentrated from combination fragment CombineFileSplit image datas, randomly selects a combination fragment.
Step 5, input key-value pair key1/value1 of the domain image into mapping function Map is converted.
By the routing information function getPath of distributed treatment frame Hadoop, obtain in selected combination fragment
The routing information of domain image.
Using the image pixel data transfer function cvDecodeImage in the JavaCV of image procossing library, by routing information
In corresponding domain image be converted to the image data of image type ImageWritable in distributed treatment frame Hadoop,
It is using the filename of domain image in routing information as key key1, the data of domain image are corresponding as key key1
value1。
Step 6, domain image is pre-processed.
The corresponding value1 of key key1 are converted into the domain figure in the JavaCV picture formats IplImage of image procossing library
As data.
Using gray value calculation formula, the gray value of each pixel in the domain image of IplImage forms is calculated, it will
The gray value of all pixels point after calculating forms the domain image after gray processing.
Gray value calculation formula is as follows：
Li=0.299 × Ri+0.587×Gi+0.114×Bi
Wherein, LiRepresent the gray value of ith pixel point in domain image, RiRepresent the ith pixel point in domain image
Red component, GiRepresent the green component of the ith pixel point in domain image, BiRepresent the ith pixel in domain image
The blue component of point.
Using maximum variance between clusters, the global threshold of domain image after gray processing is calculated.It is as follows：
1st step, after gray processing in domain image, the gray value of an optional pixel having not been used is as reference
Gray value.
Gray value is more than the pixel of domain image after the gray processing of reference gray level value, as foreground pixel by the 2nd step
Point；Gray value is less than to the pixel of domain image after the gray processing of reference gray level value, as background pixel point.
3rd step, by ratio shared in all pixels point of background pixel point domain image after gray processing, as background
Pixel ratio；By ratio shared in all pixels point of foreground pixel point domain image after gray processing, as prospect picture
Vegetarian refreshments ratio.
4th step, by background pixel point gray average, as background average gray；By foreground pixel point gray average, as
Prospect average gray.
5th step, according to the following formula, the inter-class variance value after calculating gray processing between the foreground and background of domain image.
G=ω0×ω1×(μ0-μ1)2
Wherein, g represents the inter-class variance value between the foreground and background of domain image, ω0Represent background pixel point ratio,
ω1Represent foreground pixel point ratio, μ0Represent background average gray, μ1Expression prospect average gray.
6th step, judge in the pixel having not been used in domain image whether also there are gray scale with it is used
The all different pixel of all reference gray level values if so, performing the 1st step of this step, otherwise, performs the 7th step of this step.
7th step, by reference gray level value used in inter-class variance maximum in inter-class variance, as domain figure after gray processing
The global threshold of picture.
Using two-value calculation formula, the two-value of domain image slices vegetarian refreshments after each gray processing is calculated, by all pictures of calculating
The two-value of vegetarian refreshments forms the domain image after binaryzation.
Two-value calculation formula is as follows：
Wherein, BWiRepresent the two-value of ith pixel point in domain image after gray processing, LiDomain image after expression gray processing
The gray value of middle ith pixel point, T represent the global threshold of domain image after gray processing.
Step 7, image line network edge is extracted.
1st step using method for detecting image edge, carries out edge detection to the domain image after binaryzation, obtains domain
The edge of gauze in image.
Using all 1's matrix of 3*3 as structural element, mathematical morphology etching operation is carried out to binaryzation domain image, is obtained
Domain image after corrosion subtracts pixel value at the domain image corresponding pixel points after corroding, two-value with binaryzation domain image
The edge for changing domain image is made of the pixel after subtracting each other.
2nd step, to the serial number successively since the row coordinate at the gauze edge in domain image, obtain connected region and
The sum at gauze edge.Due to the gauze in domain be all it is independent of each other, can be according to connected domain be asked to acquire gauze
Item number is simultaneously marked gauze, and a connected domain region is exactly the region of a closure gauze, therefore is directed to different gauzes,
The pixel value of gauze is consistent with the mark value to gauze, and the number of connected domain is exactly the number of gauze on required image.
Step 8, the horizontal direction edge of domain image gauze and vertical direction edge are extracted, defines the element of one 1 × 3
Value is all 1 matrix and the element value of one 3 × 1 be all 1 matrix, respectively using this matrix as structural element, to current number
The edge of gauze carries out etching operation, respectively obtains the horizontal direction edge of current number gauze and vertical direction edge, and handle
The horizontal edge of current number gauze is stored in in the equirotal matrix r esult1 of master figure image array, is compiled current
The vertical edge of number gauze be stored in in the equirotal matrix r esult2 of master figure image storage matrix.
Step 9, expansive working is carried out to domain image line network edge.
1st step according to transformation rule the defects of can result in integrated circuit diagram gauze open circuit on matrix, defines one
A defect matrix DD1×D2, wherein, D1Represent the row of defect matrix, D2Represent defect matrix column, D (i, j)=0 or 1,0≤i ＜
D1, 0≤j ＜ D2If in defect set covering theory coordinate be i, the point of j, value D (i, j)=1 of the coordinate points, otherwise D (i, j)=
0, real defect is made of the coordinate points of D (i, j)=1, for random defect D (XC,YC), XC,YCIt is the centre of form, XC=Mx/ A, YC
=My/ A, wherein, Mx, MyBe defect to the first moment of X-axis and Y-axis, A is the area of defect.
2nd step defines a horizontal direction edge image matrix and vertical direction edge image matrix, and according to line respectively
The number order of network edge extracts the horizontal direction edge of gauze and vertical direction edge, is stored respectively in horizontal direction successively
In edge image matrix and vertical direction edge image matrix, by two horizontal direction edge image matrixes of current number gauze
The abscissa of header element subtracts each other, using difference as the minimum range at current number gauze horizontal direction edge；By current number line
The abscissa of the header element of two vertical direction edge image matrixes of net subtracts each other, using difference as current number gauze vertical direction
The minimum range at edge.
3rd step, judges whether the minimum range at current number gauze horizontal direction edge is more than the sum of defect row matrix,
If so, performing the 5th step of this step, otherwise, the 4th step of this step is performed.
4th step, using defect as structural element, the expansion that mathematical morphology is carried out to the gauze horizontal edge of current number is grasped
Make.
5th step, judges whether the minimum range at current number gauze horizontal direction edge is more than the sum of defect rectangular array,
If so, performing step 7, otherwise, the 6th step of this step is performed.
6th step, using defect as structural element, the expansion that mathematical morphology is carried out to the gauze vertical edge of current number is grasped
Make.
7th step, the image pixel value after current number gauze horizontal direction edge swell and current number gauze is vertical
Pixel value after the edge swell of direction, which corresponds to, to be added.
8th step defines the image storage matrix of a processed image matrix form cvMat, stores current number gauze water
Image pixel value square after edge swell is corresponding with the pixel value after current number gauze vertical direction edge swell to be added
Result.
Step 10, judge whether the number at current gauze edge is equal with the sum at gauze edge, if so, performing step
11, otherwise, step 7 is performed after the number at gauze edge is added 1.
Step 11, open circuit critical area is calculated, calculates pixel in the image storage matrix of storage gauze edge swell result
The area in region of the value more than 1 is open circuit critical area.
According to mathematical morphology dilation operation and gauze open circuit critical area definition, in certain specific IC technological process,
By missing material defect (Defect) labeled as D (XC,YC), the wherein centre of form of real defect and the origin position of structural element is
XC,YC；Gauze (Net) is labeled as Ni；The gauze that gauze flows to axis both sides flows to side (Net Edge) labeled as NE1,
NE2...NEn, real defect D (XC,YC) dilation operation result DA is carried out to the side that flows to of single conductive gauzei(expansion area exists
Respective pixel value is 1) as follows in domain matrix:
DAi=DILATE (NEi,XC,YC) i=1,2 ... n
The intersection that the gauze of both sides is flowed to after the expansion of side represents as follows：
DA (i, j)=DAi∩DAj i≠j
In domain matrix, gauze N is flowed into side NEiResult DA after expansioni(i=1,2,3 ... n) are overlapped,
Stack result region is denoted as D, then D is represented as follows：
Region of the pixel value more than 2 i.e. open circuit key area in overlap-add region D.Open circuit key area (expands intersection
Region) area be open a way critical area.
Step 12, judge whether the domain image in present combination fragment has all been handled, if so, performing step
13, otherwise, perform step 5.
Step 13, judge whether the combination fragment in the back end currently chosen all has been handled, if so, performing
Step 14, otherwise, step 4 is performed；
Step 14, by distributed treatment frame Hadoop output open circuit critical areas, by domain image in routing information
Filename as the key key2 in Map functions output key-value pair key2/value2, using the short-circuit critical area of extraction as literary
The value value2 of the key key2 of this Text types.
Step 15, the abbreviation class Reduce of setting distributed treatment frame Hadoop.
Summarize the Map functions output key-value pair key2/value2 in total data node.
Key key2 is output in the glue file folder that outgoing route has been set, key key2 refers to the input road of Text types
File name in diameter information.
The short-circuit critical area of gained will be calculated with key-value pair key2/value2 value value2s corresponding with key key2,
It is output in the short-circuit critical area file of distributed file system HDFS.
Step 16, the task of extraction open circuit critical area is submitted.
Initialize the task Job of back end in distributed treatment frame Hadoop clusters.
A configuration object configuration is created, it is defeated using configuration object configuration setting individual nodes
Enter fragment minimum value for 64M, it is 64M to set single rack input fragment minimum value, and setting input fragment maximum value is 64M.With
Object configuration is configured and creates simultaneously initialization task Job, and specified for task Job and perform address and mapping function
The file path of Map and Reduce function Reduce.
Task job is submitted in distributed treatment frame Hadoop clusters.
The effect of the present invention can be verified by following emulation experiments.
1. simulated conditions：
The emulation of the present invention is carried out on the distributed treatment frame Hadoop clusters built.Distributed treatment frame
Hadoop clusters include 3 nodes：1 host node Master and 2 subordinate computer nodes Slaver1, Slaver2, and pass through office
Domain net three nodes of connection.
Software environment is as follows：
Linux environment：CentOS 6.5.
JDK versions：1.7_80.
Hadoop versions：Hadoop 1.2.1.
Eclipse versions：Eclipse Release 4.3.0.
JavaCV versions：JavaCV 1.1
The IP address and hardware configuration of node are as follows：
The IP address and hardware configuration list of 1 Hadoop cluster interior joints of table
| Machine name | IP address | Hardware configuration |
| Master | 192.168.131.3 | Pentium(R)Dual-Core CPU E5800@3.20GHZ |
| Slaver1 | 192.168.131.4 | Pentium(R)Dual-Core CPU E5800@3.20GHZ |
| Slaver2 | 192.168.131.5 | Pentium(R)Dual-Core CPU E5800@3.20GHZ |
Master in table 1 is the host node in distributed treatment frame Hadoop clusters, and Slaver1, Slaver2 are
Slave node in distributed treatment frame Hadoop clusters, IP address represent the network address of computer, and hardware configuration is to calculate
The model of the processor CPU of machine.
2. emulation content：
The emulation experiment of the present invention is the two methods of method and the prior art using the present invention, i.e. mathematical morphology carries
Road critical area method and the open circuit critical area extracting method based on image processing techniques are taken away, respectively to integrated circuit diagram
Image zooming-out open circuit critical area, and compare its extraction time.
3. the simulation experiment result：
The present invention chooses 7 groups of domain images respectively, in 7 domain images the number of gauze be 3696 respectively, 4990,
6216,6577,12816,23505,30824, export the open circuit critical area corresponding to 7 domain images.It adopts
It is the open circuit critical area extracting method based on image processing techniques with the method and the prior art of the present invention, respectively to 7 versions
Figure image zooming-out open circuit critical area, obtains extraction time comparison diagram as shown in Figure 2.
Abscissa in Fig. 2 represents the item number of domain gauze in the domain image of input, and physical unit is item, ordinate table
Show the extraction time of short-circuit critical area, physical unit is second s.The solid line indicated in Fig. 2 with circle is represented with prior art base
In the open circuit critical area extracting method of image procossing, in different number of domain image line extraction open circuit critical area institute off the net
With the curve of time, the solid line indicated with cross represents open circuit critical area extracting method using the present invention, in different numbers
Domain image line it is off the net, extraction open circuit critical area used in the time curve.
The extraction time curve for comparing method of the invention in Fig. 2 and prior art extraction open circuit critical area method can
：(1) as line screen purpose increases, the time needed for extraction open circuit critical area gradually increases.(2) when input domain gauze
Quantity it is identical when, compare the ordinate of the two methods time graph under same defect size, opening based on image procossing
Time used in the critical area extracting method of road is much larger than under Hadoop clusters used in extraction open circuit critical area method the time,
And when extracting open circuit critical area to large scale integrated circuit gauze, as domain line screen purpose increases, curve linear
Growth is slower, illustrates that superiority is more prominent, therefore the open circuit critical area extraction algorithm effect under the Hadoop clusters of the present invention
Rate is high and can be suitably used for large scale integrated circuit domain net.
The simulation result of the present invention shows：The present invention is the integrated circuit short circuit critical area extraction based on Hadoop, is adopted
With distributed processing system(DPS) a large amount of integrated circuit diagram images are carried out with parallel open circuit critical area extraction, improves extraction collection
Into the efficiency of open circuit road critical area.
Claims (9)
1. a kind of integrated circuit diagram open circuit critical area extracting method based on Hadoop, includes the following steps：
(1) the domain image of integrated circuit is read：
(1a) reads whole domains in the integrated circuit diagram of the open circuit critical area to be extracted of normative document image BMP forms
Image；
Each width domain image of reading is saved as customized form X_Y_Z.bmp by (1b)；
(2) the domain image of integrated circuit is uploaded to distributed treatment frame Hadoop：
(2a) starts distributed treatment frame using the startup order start-all.sh of distributed treatment frame Hadoop clusters
Hadoop clusters；
(2b) uploads to the domain image for being fully integrated circuit of reading in distributed file system HDFS；
(3) under distributed treatment frame Hadoop piecemeal storage integrated circuit domain image：
(3a) distributed file system HDFS carries out piecemeal to uploading to the domain image in the system；
(3b) fifty-fifty domain image after memory partitioning in each back end of distributed treatment frame Hadoop clusters；
(4) by the domain image slices of distributed treatment frame Hadoop back end：
(4a) randomly selects a back end from distributed treatment frame Hadoop clusters, uses composition file input format
Domain image in selected back end is divided into the group that fragment size is 64M by CombineFileInputFormat
Close fragment CombineFileSplit image data sets；
(4b) is concentrated from combination fragment CombineFileSplit image datas, randomly selects a combination fragment；
(5) input key-value pair key1/value1 of the domain image into mapping class Map is converted：
(5a) is obtained by the routing information function getPath of distributed treatment frame Hadoop in selected combination fragment
The routing information of domain image；
(5b) using the image pixel data transfer function cvDecodeImage in the JavaCV of image procossing library, by routing information
In corresponding domain image be converted to the image data of image type ImageWritable in distributed treatment frame Hadoop,
By the filename of domain image in routing information, as the key key1 of key-value pair in mapping class Map, the data of domain image are made
For the corresponding value1 of key key1；
(6) domain image is pre-processed：
The corresponding value1 of key key1 are converted to the domain figure in the JavaCV picture formats IplImage of image procossing library by (6a)
As data；
(6b) is calculated the gray value of each pixel in the domain image of IplImage forms, is incited somebody to action using gray value calculation formula
The gray value of all pixels point after calculating forms the domain image after gray processing；
(6c) calculates the global threshold of domain image after gray processing using maximum variance between clusters；
(6d) calculates the two-value of domain image slices vegetarian refreshments after each gray processing, by all pictures of calculating using two-value calculation formula
The two-value of vegetarian refreshments forms the domain image after binaryzation；
(7) domain image line network edge is extracted：
(7a) is carried out edge detection to the domain image after binaryzation, is obtained in domain image using method for detecting image edge
Gauze edge；
(7b) serial number successively since the row coordinate at the gauze edge in domain image, obtains connected region and gauze edge
Sum；
(8) the horizontal direction edge of domain image gauze and vertical direction edge are extracted：
Define respectively the element value of one 1 × 3 be all 1 matrix and the element value of one 3 × 1 be all 1 matrix, with 1 × 3
The matrix that element value is all 1 is structural element, carries out etching operation to the edge of current number gauze, obtains current number gauze
Horizontal direction edge, 1 matrix is all as structural element using 3 × 1 element value, the edge of current number gauze is carried out rotten
Erosion operation, obtains the vertical direction edge of current number gauze；
(9) expansive working is carried out to domain image line network edge：
(9a) defines a defect according to transformation rule the defects of can result in integrated circuit diagram gauze open circuit on matrix
Matrix DD1×D2, wherein, D1Represent the row of defect matrix, D2Represent defect matrix column；
(9b) defines a horizontal direction edge image matrix and vertical direction edge image matrix respectively, and according to gauze edge
Number order, extract horizontal direction edge and the vertical direction edge of gauze successively, be stored respectively in horizontal direction edge graph
In picture matrix and vertical direction edge image matrix, by the header element of two horizontal direction edge image matrixes of current number gauze
Abscissa subtract each other, using difference as the minimum range at current number gauze horizontal direction edge；By current number gauze two
The abscissa of the header element of vertical direction edge image matrix subtracts each other, using difference as current number gauze vertical direction edge
Minimum range；
(9c) judges whether the minimum range at current number gauze horizontal direction edge is more than the sum of defect row matrix, if so,
Step (9e) is then performed, otherwise, performs step (9d)；
(9d) carries out the expansion of mathematical morphology using defect matrix as structural element, to the horizontal direction edge of current number gauze
Operation；
(9e) judges whether the minimum range at the vertical direction edge of current number gauze is more than the sum of the row of defect, if so,
Step (9g) is then performed, otherwise, performs step (9f)；
(9f) using defect matrix as structural element, the expansion that mathematical morphology is carried out to the gauze vertical edge of current number is grasped
Make；
(9g) is by the image pixel value after current number gauze horizontal direction edge swell and current number gauze vertical direction side
Pixel value after edge expansion, which corresponds to, to be added；
(9h) defines the image storage matrix of a processed image matrix form cvMat, stores current number gauze horizontal direction
The result of image pixel value addition corresponding with the pixel value after current number gauze vertical direction edge swell after edge swell；
(10) judge whether the number at current gauze edge is equal with the sum at gauze edge, if so, step (10) is performed, it is no
Then, step (8) is performed after the number at gauze edge being added 1；
(11) open circuit critical area is calculated：
The area in region of the pixel value more than 1 in the image storage matrix of storage gauze edge swell result is calculated, as open circuit
Critical area；
(12) judge whether the domain image in present combination fragment all extracts open circuit critical area, if so, performing step
Suddenly (13) otherwise, perform step (5)；
(13) judge whether the combination fragment in the back end currently chosen all extracts open circuit critical area, if so,
Step (14) is performed, otherwise, performs step (4)；
(14) pass through distributed treatment frame Hadoop output open circuit critical areas：
Using the filename of domain image in routing information as the key key2 of mapping class Map output key-value pairs, by the open circuit of extraction
Key key2 corresponding value2 of the critical area as text type Text；
(15) the abbreviation class Reduce of setting distributed treatment frame Hadoop：
(15a) summarizes the output key-value pair key2/value2 of the mapping class Map in total data node；
Key key2 is output in the glue file folder for having set outgoing route by (15b)；
The open circuit critical area for calculating gained with the value2 in key-value pair key2/value2, is output to distributed text by (15c)
In the open circuit critical area file of part system HDFS；
(16) task of extraction open circuit critical area is submitted：
The task Job of back end in (16a) initialization distributed treatment frame Hadoop clusters；
The extracting open circuit critical area of the task is submitted in distributed treatment frame Hadoop clusters by (16b).
2. the integrated circuit diagram open circuit critical area extracting method according to claim 1 based on Hadoop, feature
It is, each width domain image of reading is saved as into customized form X_Y_Z.bmp refer to described in step (1b)：It presses
Each layer of domain image is chosen from top to down according to the number of plies of domain, is represented with the X in user-defined format X_Y_Z.bmp,
Each layer of domain image is divided into the normative document image that multiple pixel value sizes are 900*900 according to sequence from left to right
Form BMP images represent the normative document picture format BMP images of each width 900*900 in integrated circuit diagram image with Y
The value of the initial row of middle position represents the normative document picture format BMP images of each width 900*900 integrated with Z
The value of the starting row of position in circuit layout image.
3. the integrated circuit diagram open circuit critical area extracting method according to claim 1 based on Hadoop, feature
It is, the gray value calculation formula described in step (6b) is as follows：
Li=0.299 × Ri+0.587×Gi+0.114×Bi
Wherein, LiRepresent the gray value of ith pixel point in domain image, RiRepresent the red of the ith pixel point in domain image
Colouring component, GiRepresent the green component of the ith pixel point in domain image, BiRepresent the ith pixel point in domain image
Blue component.
4. the integrated circuit diagram open circuit critical area extracting method according to claim 1 based on Hadoop, feature
It is, the specific steps using the global threshold of domain image after maximum variance between clusters calculating gray processing described in step (6c)
It is as follows：
1st step, after gray processing in domain image, the gray value of an optional pixel having not been used is used as with reference to gray scale
Value；
Gray value is more than the pixel of domain image after the gray processing of reference gray level value, as foreground pixel point by the 2nd step；It will
Gray value is less than the pixel of domain image after the gray processing of reference gray level value, as background pixel point；
3rd step, by ratio shared in all pixels point of background pixel point domain image after gray processing, as background pixel
Point ratio；By ratio shared in all pixels point of foreground pixel point domain image after gray processing, as foreground pixel point
Ratio；
4th step, by background pixel point gray average, as background average gray；By foreground pixel point gray average, as prospect
Average gray；
5th step, according to the following formula, the inter-class variance value after calculating gray processing between the foreground and background of domain image：
G=ω0×ω1×(μ0-μ1)2
Wherein, g represents the inter-class variance value between the foreground and background of domain image, ω0Represent background pixel point ratio, ω1Table
Show foreground pixel point ratio, μ0Represent background average gray, μ1Expression prospect average gray；
6th step judges whether also own in the pixel having not been used in domain image there are gray scale with used
The all different pixel of reference gray level value if so, performing the 1st step, otherwise, performs the 7th step；
7th step, by reference gray level value used in inter-class variance maximum in inter-class variance, as domain image after gray processing
Global threshold.
5. the integrated circuit diagram open circuit critical area extracting method according to claim 1 based on Hadoop, feature
It is, the two-value calculation formula described in step (6d) is as follows：
Wherein, BWiRepresent the two-value of ith pixel point in domain image after gray processing, LiRepresent after gray processing in domain image the
The gray value of i pixel, T represent the global threshold of domain image after gray processing.
6. the integrated circuit diagram open circuit critical area extracting method according to claim 1 based on Hadoop, feature
It is, the method for detecting image edge described in step (7a) is as follows：
1st step using all 1's matrix of 3*3 as structural element, carries out mathematical morphology etching operation to binaryzation domain image, obtains
Domain image after to corrosion；
2nd step subtracts pixel value at the domain image corresponding pixel points after corroding, binaryzation domain figure with binaryzation domain image
The edge of picture is made of the pixel after subtracting each other.
7. the integrated circuit diagram open circuit critical area extracting method according to claim 1 based on Hadoop, feature
It is, transformation rule the defects of can result in integrated circuit diagram gauze open circuit on matrix described in step (9a) is
Refer to：Defect is by two-dimensional matrixIt represents, wherein wherein, D1Represent the row of defect matrix, D2Represent defect matrix column, square
Any point element value D (i, j)=0 or 1 in battle array, 0≤i ＜ D1, 0≤j ＜ D2If coordinate is i in defect set covering theory, j's
Point, value D (i, j)=1 of the coordinate points, otherwise D (i, j)=0, real defect are made of the coordinate points of D (i, j)=1.
8. the integrated circuit diagram open circuit critical area extracting method according to claim 1 based on Hadoop, feature
It is, the key key2 described in step (15b) refers to the file name in the input routing information of Text types.
9. the integrated circuit diagram open circuit critical area extracting method according to claim 1 based on Hadoop, feature
It is, the specific step of the task Job of back end in distributed treatment frame Hadoop clusters is initialized described in step (16a)
It is rapid as follows：
1st step creates the configuration object of a distributed treatment frame Hadoop cluster；
2nd step creates task Job, and specified for task Job and perform address and mapping class Map and change with configuration object conf
The file path of simple class Reduce.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810034620.5A CN108257172B (en) | 2018-01-15 | 2018-01-15 | Integrated circuit diagram open circuit critical area extracting method based on Hadoop |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810034620.5A CN108257172B (en) | 2018-01-15 | 2018-01-15 | Integrated circuit diagram open circuit critical area extracting method based on Hadoop |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108257172A true CN108257172A (en) | 2018-07-06 |
| CN108257172B CN108257172B (en) | 2019-08-06 |
Family
ID=62726809
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810034620.5A Active CN108257172B (en) | 2018-01-15 | 2018-01-15 | Integrated circuit diagram open circuit critical area extracting method based on Hadoop |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108257172B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109712181A (en) * | 2018-12-26 | 2019-05-03 | 西安电子科技大学 | The extracting method of open circuit critical area on integrated circuit diagram gauze |
| CN117372567A (en) * | 2023-12-06 | 2024-01-09 | 华芯程(杭州)科技有限公司 | Layout generation method, device, equipment and medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101178809A (en) * | 2007-10-12 | 2008-05-14 | 西安电子科技大学 | Estimation method for integrate circuit function yield |
| CN101419643A (en) * | 2008-10-17 | 2009-04-29 | 西安电子科技大学 | Integrated circuit diagram optimizing method based on mathematical morphologic |
| CN101789048A (en) * | 2010-02-08 | 2010-07-28 | 浙江大学 | Method for quickly extracting critical area of layout |
| CN104794308A (en) * | 2015-05-07 | 2015-07-22 | 西安电子科技大学 | Method for converting layout image into CIF file based on image edge detection |
| CN107067434A (en) * | 2017-04-25 | 2017-08-18 | 西安电子科技大学 | The short-circuit critical area extracting method of integrated circuit based on Hadoop |
-
2018
- 2018-01-15 CN CN201810034620.5A patent/CN108257172B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101178809A (en) * | 2007-10-12 | 2008-05-14 | 西安电子科技大学 | Estimation method for integrate circuit function yield |
| CN101419643A (en) * | 2008-10-17 | 2009-04-29 | 西安电子科技大学 | Integrated circuit diagram optimizing method based on mathematical morphologic |
| CN101789048A (en) * | 2010-02-08 | 2010-07-28 | 浙江大学 | Method for quickly extracting critical area of layout |
| CN101789048B (en) * | 2010-02-08 | 2011-10-19 | 浙江大学 | Method for quickly extracting critical area of layout |
| CN104794308A (en) * | 2015-05-07 | 2015-07-22 | 西安电子科技大学 | Method for converting layout image into CIF file based on image edge detection |
| CN107067434A (en) * | 2017-04-25 | 2017-08-18 | 西安电子科技大学 | The short-circuit critical area extracting method of integrated circuit based on Hadoop |
Non-Patent Citations (1)
| Title |
|---|
| 王乐，: ""基于图像处理技术的开路关键面积提取"", 《万方数据知识服务平台》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109712181A (en) * | 2018-12-26 | 2019-05-03 | 西安电子科技大学 | The extracting method of open circuit critical area on integrated circuit diagram gauze |
| CN109712181B (en) * | 2018-12-26 | 2022-12-06 | 西安电子科技大学 | Method for extracting open-circuit key area on integrated circuit layout line network |
| CN117372567A (en) * | 2023-12-06 | 2024-01-09 | 华芯程(杭州)科技有限公司 | Layout generation method, device, equipment and medium |
| CN117372567B (en) * | 2023-12-06 | 2024-02-23 | 华芯程(杭州)科技有限公司 | Layout generation method, device, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108257172B (en) | 2019-08-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Xiang et al. | Learning rgb-d feature embeddings for unseen object instance segmentation | |
| CN107067434B (en) | Integrated circuit short circuit critical area extracting method based on Hadoop | |
| US11055570B2 (en) | Image processing method and apparatus for implementing image recognition, and electronic device | |
| CN104881466B (en) | The processing of data fragmentation and the delet method of garbage files and device | |
| WO2016119117A1 (en) | Localization and mapping method | |
| CN104866855A (en) | Image feature extraction method and apparatus | |
| Iscen et al. | Panorama to panorama matching for location recognition | |
| CN108257172B (en) | Integrated circuit diagram open circuit critical area extracting method based on Hadoop | |
| CN108304859A (en) | Image-recognizing method and cloud system | |
| KR20150089663A (en) | Device for multi-shape primitives fitting of 3D point clouds using graph-based segmentation and method thereof | |
| TW201523300A (en) | Three-dimension image file serching method and three-dimension image file serching system | |
| CN105023291A (en) | Criminal scene reconstructing apparatus and method based on stereoscopic vision | |
| Giang et al. | TopicFM: Robust and interpretable topic-assisted feature matching | |
| WO2016095068A1 (en) | Pedestrian detection apparatus and method | |
| CN109712181B (en) | Method for extracting open-circuit key area on integrated circuit layout line network | |
| Castro-Gama et al. | A graph theoretical sectorization approach for energy reduction in water distribution networks | |
| CN117422851A (en) | Virtual clothes changing method and device and electronic equipment | |
| CN101504723B (en) | Projection space establishing method and apparatus | |
| Sheng et al. | Detection of content-aware image resizing based on Benford’s law | |
| CN106407538B (en) | Bitmap domain based on Hadoop is to CIF domain conversion method | |
| Liu et al. | Spatiotemporal fusion networks for video action recognition | |
| Jang et al. | Equivalent line limit calculation using available thermal transfer capability | |
| Liu et al. | Point2CN: Progressive two-view correspondence learning via information fusion | |
| Zhu et al. | Parallel image texture feature extraction under hadoop cloud platform | |
| Rawat et al. | Steganalysis for clustering modification directions steganography |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |