CN106127093A - A kind of Quick Response Code identification chip and its implementation - Google Patents
A kind of Quick Response Code identification chip and its implementation Download PDFInfo
- Publication number
- CN106127093A CN106127093A CN201610435710.6A CN201610435710A CN106127093A CN 106127093 A CN106127093 A CN 106127093A CN 201610435710 A CN201610435710 A CN 201610435710A CN 106127093 A CN106127093 A CN 106127093A
- Authority
- CN
- China
- Prior art keywords
- module
- image
- code
- quick response
- response code
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/10544—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation by scanning of the records by radiation in the optical part of the electromagnetic spectrum
- G06K7/10821—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation by scanning of the records by radiation in the optical part of the electromagnetic spectrum further details of bar or optical code scanning devices
- G06K7/10861—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation by scanning of the records by radiation in the optical part of the electromagnetic spectrum further details of bar or optical code scanning devices sensing of data fields affixed to objects or articles, e.g. coded labels
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/146—Methods for optical code recognition the method including quality enhancement steps
- G06K7/1482—Methods for optical code recognition the method including quality enhancement steps using fuzzy logic or natural solvers, such as neural networks, genetic algorithms and simulated annealing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/146—Methods for optical code recognition the method including quality enhancement steps
- G06K7/1491—Methods for optical code recognition the method including quality enhancement steps the method including a reconstruction step, e.g. stitching two pieces of bar code together to derive the full bar code
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Electromagnetism (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Toxicology (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Fuzzy Systems (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Automation & Control Theory (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of Quick Response Code identification chip and its implementation, chip includes that ARM core module, memory module, security module, image processing module, bus interface module and I/O module, ARM core module, memory module, security module, image processing module, bus interface module and I/O module are all integrated on a chips and are all connected with ahb bus or APB bus;Method includes: drive photographic head to obtain image in 2 D code;The image in 2 D code obtained is carried out blind reconstruction processing;Image after blind reconstruction processing is carried out image characteristics extraction;Use the SVM classifier trained to carry out intelligent mode identification according to the characteristics of image extracted, obtain the type belonging to Quick Response Code;Result according to intelligent mode identification carries out code word extraction and decoding, and exports.The advantage that the present invention has low cost, processing speed is fast, volume is little, general, low in energy consumption and recognition efficiency is high, can be widely applied to planar bar code technology field.
Description
Technical field
The present invention relates to planar bar code technology field, especially a kind of Quick Response Code identification chip and its implementation.
Background technology
The bar code that what people were seen be printed in commodity packaging is bar code, and it is widely used in multiple fields, pole
The earth improves the speed of data acquisition and information processing.But it represents data the most in the horizontal, and longitudinal direction does not has information, density
Relatively low, therefore be only capable of marking articles and article can not be described.If it is to be understood that product is for information about, it is necessary to pass through reading bar code
And enter data base.This will ask for help must set up the data base with the code represented by bar code as index field in advance,
Data base and its application of place that can not network is not had to be restricted.It addition, represent Chinese character information the most not by one-dimension code
Possible, this is in the occasion of some application Chinese character, it appears inconvenient, and efficiency is the lowest.In order to make up one-dimension code not
Foot, Quick Response Code arises at the historic moment.
Quick Response Code has that information capacity is big, coding range is wide, fault-tolerant ability is strong, decoding reliability is high, fouling resistance ability is dashed forward
Go out, low cost and other advantages, it has also become one of information automatic identification technology that Present Global is most widely used.Its application drives
Mobile e-business, publication, safety management, transportation, finance, health care, industrial and commercial administration, modern logistics,
The further development of the industry fields such as customs.Due to its applied range, demand is big, is deep into social life each simultaneously
Aspect, therefore its development will drive the development of economy, safety and the scientific and technological level of country.
Chinese market is following huge to the demand of Quick Response Code, most enterprises to Quick Response Code China universal raise one's head and look with
Treat.But the market price of the equipment of Quick Response Code reading at present is between 5000~9000 yuan, considerably beyond the valency of one-dimension code reader
Lattice, so it is typically just used by major company, do not spread to relevant industries.
The design of chip in recent years and manufacturing technology have obtained rapid raising, and China's Internet of Things/Internet enterprises is also made every effort to
Break through in high-end technical fields such as Quick Response Code chip designs, even surmount external planar bar code technology with pursuit, go forward side by side one
Step reduces the threshold of Quick Response Code identification chip designing technique.
The method that traditional Quick Response Code identification equipment have employed software collection image and decoding, its circuit is mostly based on discrete
Components and parts or Programmable Logic Device, although this method is easily achieved, but it still has more defect, such as processing speed
Slowly, volume is big, power consumption is big and recognition efficiency is low etc..
Additionally, the standard of Quick Response Code is various, the Quick Response Code standard of current main flow has QR code (quick response matrix code), Data
Matrix code, PDF417 code etc., these standards are the most incompatible, cause enterprise to need to purchase the reading equipment of multiple standards
And be required to be every kind of standard and all design corresponding recognition methods, add enterprise's totle drilling cost.
Summary of the invention
For solving above-mentioned technical problem, it is an object of the invention to: provide a kind of low cost, processing speed is fast, volume is little,
Low in energy consumption high with recognition efficiency, Quick Response Code identification chip.
Another object of the present invention is to: provide a kind of low cost, general, extensibility good and portability is good, two
The implementation method of dimension code identification chip.
The technical solution used in the present invention is:
A kind of Quick Response Code identification chip, including ARM core module, memory module, security module, image processing module, EBI
Module and I/O module, described ARM core module, memory module, security module, image processing module, bus interface module and I/O
Module is all integrated on a chips, described ARM core module, memory module, security module, image processing module, EBI
Module and I/O module are all connected with ahb bus or APB bus.
Further, described ARM core module includes electrification reset circuit, phaselocked loop and OSC agitator, described electrification reset electricity
Road, phaselocked loop and OSC agitator are all connected with ahb bus or APB bus.
Further, described memory module includes FLASH memory, the first SRAM memory and the second SRAM memory, institute
State FLASH memory, the first SRAM memory and the second SRAM memory to be all connected with ahb bus or APB bus.
Further, described security module includes AES encryption circuit, des encryption circuit, rsa encryption circuit and ECC encryption
Circuit, described AES encryption circuit, des encryption circuit, rsa encryption circuit and ECC encrypted circuit are all total with ahb bus or APB
Line connects.
Further, described bus interface module includes UART interface, SPI interface, ISO7816 interface and I2C interface, described
UART interface, SPI interface, ISO7816 interface and I2C interface are all connected with ahb bus or APB bus.
Further, described I/O module includes universaling I/O port, intervalometer interface and cmos sensor interface, described general purpose I/
O mouth, intervalometer interface and cmos sensor interface are all connected with ahb bus or APB bus.
What the present invention was taked another solution is that
The implementation method of a kind of Quick Response Code identification chip, comprises the following steps:
Photographic head is driven to obtain image in 2 D code;
The image in 2 D code obtained is carried out blind reconstruction processing;
Image after blind reconstruction processing is carried out image characteristics extraction;
Use the SVM classifier trained to carry out intelligent mode identification according to the characteristics of image extracted, obtain belonging to Quick Response Code
Type;
Result according to intelligent mode identification carries out code word extraction and decoding, and information that reduction Quick Response Code is preserved also carries out defeated
Go out.
Further, the step for that the described image in 2 D code to obtaining carrying out blind reconstruction processing, comprising:
If the image in 2 D code obtained noise occurs, obscures, is stained or the situation of miscellaneous speckle, then use blind deconvolution method to obtaining
Image in 2 D code is reconstructed, and described blind deconvolution method includes but not limited to liftering, Wiener filtering, subspace filtering,
A young waiter in a wineshop or an inn takes advantage of filtering, Parameter Estimation Method, iteration blind deconvolution algorithm, NAS-RIF blindly restoring image algorithm, total variation regularization blind multiple
Former algorithm, the blind restoration algorithm of Bayes priori and the blind restoration algorithm of Bussgang;
If the image in 2 D code obtained there occurs the phenomenon of inclination, then first pass through the two dimension that the detection of outline of straight line detection algorithm obtains
The angle of inclination of the outline of straight line of code image, then carries out slant correction according to the angle of inclination of detection, and described outline of straight line is examined
Method of determining and calculating includes but not limited to line Segment Detection based on Freeman chain code and straight line parameter based on Hough transform inspection
Survey method;
If the image in 2 D code obtained there occurs the phenomenon of inclination, then by reflection transformation, the image in 2 D code obtained is carried out
Reconstruct.
Further, the characteristics of image that described basis extracts uses the SVM classifier trained to carry out intelligent mode identification,
The step for of obtaining the type belonging to Quick Response Code, comprising:
Corresponding SVM classifier is designed according to the characteristics of image extracted;
Use training sample that the SVM classifier of design is trained;
Carry out intelligent mode identification according to the SVM classifier the trained characteristics of image to extracting, obtain belonging to Quick Response Code
Type.
Further, the described result according to intelligent mode identification carries out code word extraction and decoding, and reduction Quick Response Code is preserved
Information and the step for export, comprising:
The Quick Response Code code word extraction module of the call by result respective standard according to intelligent mode identification carries out code word extraction, obtains phase
The sequence of modules answered;
The decoding module calling respective standard according to the sequence of modules obtained decodes, the two-dimensional code data after being decompressed
Stream, described decoding process include but not limited to judgements of data query, error correction code word, code word error correcting and detecting, code word data judgement and
Code word decompresses;
Data stream after decompressing is sent to gui interface and shows, and exports.
The chip of the present invention provides the benefit that: include ARM core module, memory module, security module, image procossing mould
Block, bus interface module and I/O module, ARM core module, memory module, security module, image processing module, EBI mould
Block and I/O module are all integrated on a chips, use advanced semiconductor integrated circuit technology to existing Quick Response Code chip
Carry out the architecture innovation of chip-scale, all of image procossing and decoding circuit have all been realized by a chips, takes
For traditional discrete device or Programmable Logic Device, effectively reduce the power consumption of Quick Response Code whole set equipment, volume and become
This, and processing speed and recognition efficiency can be greatly improved.
The beneficial effects of the method for the present invention is: first drives photographic head to obtain image in 2 D code, then depends on the image obtained
Secondary carry out blind reconstruction processing, feature extraction, intelligent mode identification, code word are extracted and decoding, and final reduction identifies Quick Response Code institute
Preserve information and export, be additionally arranged the process of intelligent mode identification, the identification of multiple standards Quick Response Code can be supported, no longer
Enterprise is needed to purchase the reading equipment of multiple standards and be required to be every kind of standard and all design corresponding recognition methods, low cost is general
Property strong and extensibility and portability more preferable.
Accompanying drawing explanation
Fig. 1 is the frame diagram of the present invention a kind of Quick Response Code identification chip;
Fig. 2 is the overall flow figure of the implementation method of the present invention a kind of Quick Response Code identification chip;
Fig. 3 is the functional realiey flow chart of Quick Response Code identification chip in the embodiment of the present invention two;
Fig. 4 is the flow chart of embodiment of the present invention two modes identification process;
Fig. 5 is the flow chart of the embodiment of the present invention two Quick Response Code decoding process.
Detailed description of the invention
With reference to Fig. 1, a kind of Quick Response Code identification chip, including ARM core module, memory module, security module, image procossing mould
Block, bus interface module and I/O module, described ARM core module, memory module, security module, image processing module, bus connect
Mouth die block and I/O module are all integrated on a chips, described ARM core module, memory module, security module, image procossing mould
Block, bus interface module and I/O module are all connected with ahb bus or APB bus.
With reference to Fig. 1, being further used as preferred embodiment, described ARM core module includes electrification reset circuit, phaselocked loop
With OSC agitator, described electrification reset circuit, phaselocked loop and OSC agitator are all connected with ahb bus or APB bus.
With reference to Fig. 1, being further used as preferred embodiment, described memory module includes FLASH memory, a SRAM
Memorizer and the second SRAM memory, described FLASH memory, the first SRAM memory and the second SRAM memory are all and AHB
Bus or APB bus connect.
With reference to Fig. 1, being further used as preferred embodiment, described security module includes AES encryption circuit, des encryption
Circuit, rsa encryption circuit and ECC encrypted circuit, described AES encryption circuit, des encryption circuit, rsa encryption circuit and
ECC encrypted circuit is all connected with ahb bus or APB bus.
With reference to Fig. 1, being further used as preferred embodiment, described bus interface module includes that UART interface, SPI connect
Mouthful, ISO7816 interface and I2C interface, described UART interface, SPI interface, ISO7816 interface and I2C interface are all and ahb bus
Or APB bus connects.
With reference to Fig. 1, being further used as preferred embodiment, described I/O module includes universaling I/O port, intervalometer interface
With cmos sensor interface, described universaling I/O port, intervalometer interface and cmos sensor interface all with ahb bus or APB bus
Connect.
With reference to Fig. 2, the implementation method of a kind of Quick Response Code identification chip, comprise the following steps:
Photographic head is driven to obtain image in 2 D code;
The image in 2 D code obtained is carried out blind reconstruction processing;
Image after blind reconstruction processing is carried out image characteristics extraction;
Use the SVM classifier trained to carry out intelligent mode identification according to the characteristics of image extracted, obtain belonging to Quick Response Code
Type;
Result according to intelligent mode identification carries out code word extraction and decoding, and information that reduction Quick Response Code is preserved also carries out defeated
Go out.
Being further used as preferred embodiment, the described image in 2 D code to obtaining carries out this step of blind reconstruction processing
Suddenly, comprising:
If the image in 2 D code obtained noise occurs, obscures, is stained or the situation of miscellaneous speckle, then use blind deconvolution method to obtaining
Image in 2 D code is reconstructed, and described blind deconvolution method includes but not limited to liftering, Wiener filtering, subspace filtering,
A young waiter in a wineshop or an inn takes advantage of filtering, Parameter Estimation Method, iteration blind deconvolution algorithm, NAS-RIF blindly restoring image algorithm, total variation regularization blind multiple
Former algorithm, the blind restoration algorithm of Bayes priori and the blind restoration algorithm of Bussgang;
If the image in 2 D code obtained there occurs the phenomenon of inclination, then first pass through the two dimension that the detection of outline of straight line detection algorithm obtains
The angle of inclination of the outline of straight line of code image, then carries out slant correction according to the angle of inclination of detection, and described outline of straight line is examined
Method of determining and calculating includes but not limited to line Segment Detection based on Freeman chain code and straight line parameter based on Hough transform inspection
Survey method;
If the image in 2 D code obtained there occurs the phenomenon of inclination, then by reflection transformation, the image in 2 D code obtained is carried out
Reconstruct.
Being further used as preferred embodiment, the characteristics of image that described basis extracts uses the svm classifier trained
Device carries out intelligent mode identification, the step for of obtaining the type belonging to Quick Response Code, comprising:
Corresponding SVM classifier is designed according to the characteristics of image extracted;
Use training sample that the SVM classifier of design is trained;
Carry out intelligent mode identification according to the SVM classifier the trained characteristics of image to extracting, obtain belonging to Quick Response Code
Type.
Being further used as preferred embodiment, the described result according to intelligent mode identification carries out code word extraction and translates
Code, the reduction information that preserved of Quick Response Code the step for export, comprising:
The Quick Response Code code word extraction module of the call by result respective standard according to intelligent mode identification carries out code word extraction, obtains phase
The sequence of modules answered;
The decoding module calling respective standard according to the sequence of modules obtained decodes, the two-dimensional code data after being decompressed
Stream, described decoding process include but not limited to judgements of data query, error correction code word, code word error correcting and detecting, code word data judgement and
Code word decompresses;
Data stream after decompressing is sent to gui interface and shows, and exports.
Below in conjunction with Figure of description and specific embodiment the present invention it is further explained and illustrates.
Embodiment one
Reference Fig. 1, the first embodiment of the present invention:
The present invention uses the semiconductor integrated circuit technology of advanced person, is tied by the system of low power consumption and low cost technology and chip-scale
Structure is innovated, and can all of image procossing and decoding circuit be realized with a chips, instead of traditional discrete device or can
Program logic circuit, effectively reduces the power consumption of Quick Response Code whole set equipment, volume and cost, and can be greatly improved recognition speed,
Recognition efficiency.
The Quick Response Code identification chip of the present embodiment mainly includes ARM core module, memory module, security module, image procossing
Module, bus interface module, I/O module, AHB/APB bus.
Wherein, ARM core module, the control core of whole chips, for carrying out at data according to the image in 2 D code gathered
Reason, it is provided that the control logic of whole Quick Response Code identification process and control signal.ARM core module mainly includes electrification reset circuit,
Phaselocked loop and OSC agitator.Electrification reset circuit, is used for carrying out electrification reset.Phaselocked loop, for the reference according to outside input
Signal (the image in 2 D code signal i.e. gathered) controls frequency and the phase place of internal oscillator signal.OSC agitator, for according to ginseng
Examine signal produce internal oscillator signal and export.
Memory module, for storing the data of Quick Response Code identification process.Memory module include FLASH memory, first
SRAM memory and the second SRAM memory.Wherein, FLASH memory is 32Kx32 FLASH, and the first SRAM memory is
8Kx32 SRAM, the second SRAM memory is 1Kx32 SRAM.
Security module, is used for carrying out safe encryption.Security module includes AES encryption circuit, des encryption circuit, rsa encryption
Circuit and ECC encrypted circuit, support multiple safe encryption standard.Security module can use Ling Ke core peace science and technology (Beijing) limited
The high-end encryption chip of LTK4201N that company produces realizes.
Image processing module, mainly processes the image in 2 D code gathered according to the control signal of ARM core module, with
Auxiliary ARM core module carries out Quick Response Code identification.Image processing module, can use GPU to realize.
Bus interface module and I/O module, for coordinating ahb bus or APB bus to realize the data communication inside and outside chip.
Bus interface module includes UART interface, SPI interface, ISO7816 interface and I2C interface, support SPI, ISO7816, UART and
I2C etc..I/O module includes universaling I/O port, intervalometer interface and cmos sensor interface.
The main working process of the chip of the present invention is: first, and under the control signal effect of ARM core module, image passes
Sensor gathers image in 2 D code to be identified, and the image in 2 D code gathered is passed through I/O module and ahb bus or APB bus
It is sent to image processing module.Then, image processing module is according to image in 2 D code to gathering of the control signal of ARM core module
Carry out processing (including feature extraction and pattern recognition etc.), and the result of process is sent to ARM core module.Followed by, ARM core
Module carries out follow-up code word according to the result of image processing module and extracts and decoding process, the letter that reduction Quick Response Code is preserved
Breath.2 D code information after reduction can be stored by memory module, or shows through display interface, or at encrypted mould
It is sent out by I/O module after block encryption.
The Quick Response Code identification chip of the present invention has a following fundamental characteristics:
A. 1024 ECC AESs are supported;
B. 2304 RSA cryptographic algorithms are supported;
C. AES 128, AES 192, AES 256 AES are supported;
D. des encryption algorithm is supported;
E. EAL5+ highest level secure authentication technology is supported;
F. RGB 565 CMOS interface is supported;
G. multiple coded format is supported: Code39, Code 39 Full ASCII, Trioptic Code 39, Code 32,
Code 11,Code93,Code128,Matrix 2 of 5,Data Matrix (1234567890),Discrete 2 of
5,MSI, Chinese 2 of 5,GS1 DataBar variants,Korean 3 of 5,ISBT Concat
(CODE128);
H. the coded format commonly used is supported: PDF417, QR Code, DM Code;
I. 8Kx32 SRAM is supported;
J. the embedded flash of 32Kx32 is supported;
K. 16 true random numbers are supported;
L. SPI, ISO7816, UART EBI are supported;
M. PLL:10Mhz~120Mhz is supported.
Embodiment two
Reference Fig. 2-5, the first embodiment of the present invention:
For the various and incompatible problem brought of prior art Quick Response Code standard, the present invention proposes a kind of brand-new Quick Response Code
The implementation method of identification chip.As it is shown on figure 3, the method restore based on distortion Image Blind restructing algorithm, support multiple standards
Decoding algorithm and Quick Response Code feature mode Recognition Theory, specifically include procedure below;
(1) the image in 2 D code acquisition module including imageing sensor drives photographic head to obtain image in 2 D code.
(2) acquisition image in 2 D code is called Image Blind reconstructed module and carry out blind reconstruction processing.
Because the complexity of Quick Response Code identification conditions for equipment use, its image collected is likely because light is uneven
Weigh and cause noise jamming phenomenon, or the problems such as deformation that are distorted because camera lens and image are uneven, need to call Image Blind
Reconstructed module carries out deblurring, white balance, denoising, distortion correction etc. to it and processes, and improves the image quality of image, thus improves
Discrimination.
(3) extract the feature of the image in 2 D code after blind reconstruct, utilize the SVM classifier trained to enter according to characteristics of image
Row intelligent mode identification, identifies which kind of standard this Quick Response Code belongs to.
(4) process with decoder according to the Quick Response Code code word extraction module that the call by result identified is correct.
This process decoder of the number of modules sequence obtained through Quick Response Code code word extraction module being made a gift to someone carries out codeword matching, entangles
Mistake, decompression etc. operate, and are finally reduced to the information that bar code preserves.
Code word array after error correction is carried out decoding by coding inverse process and decompresses by decoder, reduction two-dimensional code display information.
Decoder calls Text Mode, byte mode, figure pattern to code word after error correction respectively according to different lockings or transfer mode
Decompress, obtain the data stream decompressed, finally export and show to GUI module.
Mainly the main process involved by the method and correlation theory are described in detail below.
(1) Image Blind reconstruct
Quick Response Code is printed in packaging, thus the image that gets of image capture module it some times happens that: 1) tilt or inverted
Phenomenon;2) cause, because Quick Response Code is uneven with photographic head object lens, the phenomenon deformed upon up and down, even there will be distortion existing
As;3) there is phenomenon of book contamination in Quick Response Code self.Additionally, the impact of light environment also results in the image that image capture module obtains
Produce noise, the phenomenon such as fuzzy.These phenomenons will cause image to cut recombination error, eventually produces the decoding of mistake.And
Blind image restoration technology can only use the artwork degenerated and the partial information of picture system, to estimate from the characteristics of image degenerated
Artwork and then recovery artwork.Therefore, in the research and development of Quick Response Code terminal, the algorithm of research Image Blind reconstruct is to have very much actual meaning
Justice.
Wherein, image generation noise, obscure, be stained, the phenomenon such as miscellaneous speckle time, the method for available blind deconvolution is weighed
Structure.In many cases, the observed image of degenerationg(x,y) available be really imagef(x,y) and a linear shift invariant system
Point spread function h (x,y) two-dimensional convolution plus additional noisen(x,y) represent, i.e.g(x,y)= f(x,y)* h(x,y)+n(x,y), and point spread functionh(x,y) it is the most all explicitly known.Such issues that obtained substantial amounts of research,
Which includes the technology that many is widely known by the people, such as liftering, Wiener filtering, subspace filtering and least squares filtering
Deng.
During image run-off the straight, the angle of inclination of graph line profile can be detected, be then corrected for, be primarily present with
Lower two kinds of methods:
A kind of is line Segment Detection based on Freeman chain code.The image object that it processes is edge image, including basis
Chain code estimates curvature, detects chain code angle point, splits chain code at corner point, and detection of straight lines section chain code also carries out straight line parameter estimation,
Connect criterion according to the straightway specified and carry out straight line connection (i.e. splicing).The straightway that this method detects easily is made an uproar
Sound interference, and be difficult to suitable, the general straightway of definition and connect criterion.
Another kind is straight line parameter detection method based on Hough transform.The essence of Hough transform be achieve a kind of from
Image space is to the mapping relations of parameter space.The basic thought of Hough straight line conversion is the duality of dotted line: on the one hand, figure
The corresponding line intersected in parameter space of point in image space;On the other hand, in parameter space, intersect at the institute of same point
Straight line is had to have the point of conllinear the most corresponding in image space.Therefore Hough transform is straight-line detection in image space
Problem is transformed in parameter space the test problems to point, completes detection by the statistics that simply adds up in parameter space
Task.Owing to Hough transform is to calculate comprehensive characterising parameter, thus for zone boundary by noise jamming according to Local Metric
Or hidden by other targets and the situation that causes border to occur some to be interrupted, it has good fault-tolerance and robustness.But
Directly carry out Hough transform and there is shortcoming computationally intensive, that the accuracys rate such as missing inspection easily occur.It is thus desirable to Hough transform
Improve, optimize, improve its speed of service.
Twisted phenomena belongs to geometric distortion, and arbitrarily geometric distortion can by non-distorted coordinate system (x,y) transform to mistake
True coordinates system (x’,y’) equation define, the general type of equation is:x’=h 1 (x,y),y’=h 2 (x,y)。
Iff(x,y) it is undistorted original image,g(x’,y’) bef(x,y) result after distortion, this Distortion course
It is known that and use functionh 1 (x,y) andh 2 (x,y) define.Then have:
g(x’,y’)=f(x,y)
Above formula explanation should occur in the picture pixel (x,yGray value on), due to distortion, the most but occurs in
(x’,y’) upper.The recovery problem of this distortion can be solved by mapping transformation.
(2) Quick Response Code feature mode identification
Pattern recognition is through computer automatically (or intervening the most on a small quantity) and pattern to be identified is assigned to respectively
The technology of individual pattern apoplexy due to endogenous wind.The basic framework of image steganalysis is as shown in Figure 4.
Mode identification procedure is considered as a mapping process from sample space to classification space.If one is had
N feature is divided into different regions as the n dimensional feature space of parameter, then each region is corresponding with a quasi-mode class.
Wherein, character selection and abstraction is an important step of pattern recognition, if selected feature can reflect more comprehensively
The substitutive characteristics of class, then grader is just easier design;Otherwise, the difficulty of classifier design is increased by.Therefore to difference
The image in 2 D code of standard carries out feature selection and to extract be an important content of pattern recognition.And extract and select feature it
After, the quality of classification and recognition effect depends on the character of designed grader.Therefore, grader is then of the present invention
Key component.
Can build grader with traditional mode identification method, these methods include: statistic decision method, structural models
Recognition methods, Fuzzy Pattern Recognition Method and method based on artificial intelligence.But these traditional methods are at finite sample especially
In the case of being small sample, it is difficult to obtain preferable effect.And support vector machines (Support Vector Machine) leads to
Cross structural risk minimization principle to improve generalization ability, preferably solve small sample, dimension non-linear, high, local minimum
Deng practical problem, neural metwork training unstable result, the problem being easily trapped into local minimum can be avoided again.Therefore the present invention will adopt
Image in 2 D code feature mode recognition classifier is constructed with SVM.
After constructing the SVM classifier of image in 2 D code feature mode identification, the present invention will be good according to training in advance
Standard belonging to SVM classifier Intelligent Recognition Quick Response Code, and then call relevant Quick Response Code decoding module and carry out follow-up decoding.
(3) Quick Response Code decoding
Having now tens kinds of Quick Response Code standards in the world, common just have a PDF417 Quick Response Code, Datamatrix Quick Response Code,
Maxicode Quick Response Code, QR Code, Code 49, Code 16K, Code one etc., the most different.Therefore, identify
The Quick Response Code of the most various criterions, then need the reading equipment of multiple standards.
It practice, the work decoding of the no matter Quick Response Code of which kind of standard, its flow process or the same, as shown in Figure 5.
As shown in Figure 5, the decoding process of the Quick Response Code of any standard all includes that code word error correcting and detecting, code word decompression etc. are several
Individual step.Except that the error correcting and detecting mode of the Quick Response Code of various criterion is different, the method that code word decompresses not to the utmost as.
Therefore, the present invention only need to first carry out the Quick Response Code decoding algorithm of one of them standard according to Fig. 5, then combines Fig. 5 on this basis
Obtain the Quick Response Code decoding algorithm of other standard.
For a common technology difficult problem for existing Quick Response Code chip design, present invention is primarily based on image in 2 D code blind reconstruct reason
Discuss and technology, Quick Response Code feature mode identification technology, Quick Response Code decoding technique, low power consumption and low cost Quick Response Code mark core can be realized
Sheet design theory with can realize technology, design attainable Quick Response Code identification chip, and propose low cost Quick Response Code mark
The new technique of chip design and new method, reduce complexity, power consumption and the cost of the design of Quick Response Code identification chip, improve knowledge
Not rate and reliability, and solve the problem that Quick Response Code identification chip designs and industrialization is brought.
It is above the preferably enforcement of the present invention is illustrated, but the invention is not limited to described enforcement
Example, those of ordinary skill in the art also can make all equivalent variations on the premise of spirit of the present invention or replace
Changing, deformation or the replacement of these equivalents are all contained in the application claim limited range.
Claims (10)
1. a Quick Response Code identification chip, it is characterised in that: include ARM core module, memory module, security module, image procossing
Module, bus interface module and I/O module, described ARM core module, memory module, security module, image processing module, bus
Interface module and I/O module are all integrated on a chips, described ARM core module, memory module, security module, image procossing
Module, bus interface module and I/O module are all connected with ahb bus or APB bus.
A kind of Quick Response Code identification chip the most according to claim 1, it is characterised in that: described ARM core module includes powering on
Reset circuit, phaselocked loop and OSC agitator, described electrification reset circuit, phaselocked loop and OSC agitator all with ahb bus or APB
Bus connects.
A kind of Quick Response Code identification chip the most according to claim 1, it is characterised in that: described memory module includes FLASH
Memorizer, the first SRAM memory and the second SRAM memory, described FLASH memory, the first SRAM memory and second
SRAM memory is all connected with ahb bus or APB bus.
A kind of Quick Response Code identification chip the most according to claim 1, it is characterised in that: described security module includes that AES adds
Cipher telegram road, des encryption circuit, rsa encryption circuit and ECC encrypted circuit, described AES encryption circuit, des encryption circuit, RSA
Encrypted circuit and ECC encrypted circuit are all connected with ahb bus or APB bus.
A kind of Quick Response Code identification chip the most according to claim 1, it is characterised in that: described bus interface module includes
UART interface, SPI interface, ISO7816 interface and I2C interface, described UART interface, SPI interface, ISO7816 interface and I2C connect
Mouth is all connected with ahb bus or APB bus.
6. according to a kind of Quick Response Code identification chip described in any one of claim 1-5, it is characterised in that: described I/O module bag
Including universaling I/O port, intervalometer interface and cmos sensor interface, described universaling I/O port, intervalometer interface and cmos sensor connect
Mouth is all connected with ahb bus or APB bus.
7. the implementation method of a Quick Response Code identification chip, it is characterised in that: comprise the following steps:
Photographic head is driven to obtain image in 2 D code;
The image in 2 D code obtained is carried out blind reconstruction processing;
Image after blind reconstruction processing is carried out image characteristics extraction;
Use the SVM classifier trained to carry out intelligent mode identification according to the characteristics of image extracted, obtain belonging to Quick Response Code
Type;
Result according to intelligent mode identification carries out code word extraction and decoding, and information that reduction Quick Response Code is preserved also carries out defeated
Go out.
The implementation method of a kind of Quick Response Code identification chip the most according to claim 7, it is characterised in that: described to obtain
The step for that image in 2 D code carrying out blind reconstruction processing, comprising:
If the image in 2 D code obtained noise occurs, obscures, is stained or the situation of miscellaneous speckle, then use blind deconvolution method to obtaining
Image in 2 D code is reconstructed, and described blind deconvolution method includes but not limited to liftering, Wiener filtering, subspace filtering,
A young waiter in a wineshop or an inn takes advantage of filtering, Parameter Estimation Method, iteration blind deconvolution algorithm, NAS-RIF blindly restoring image algorithm, total variation regularization blind multiple
Former algorithm, the blind restoration algorithm of Bayes priori and the blind restoration algorithm of Bussgang;
If the image in 2 D code obtained there occurs the phenomenon of inclination, then first pass through the two dimension that the detection of outline of straight line detection algorithm obtains
The angle of inclination of the outline of straight line of code image, then carries out slant correction according to the angle of inclination of detection, and described outline of straight line is examined
Method of determining and calculating includes but not limited to line Segment Detection based on Freeman chain code and straight line parameter based on Hough transform inspection
Survey method;
If the image in 2 D code obtained there occurs the phenomenon of inclination, then by reflection transformation, the image in 2 D code obtained is carried out
Reconstruct.
The implementation method of a kind of Quick Response Code identification chip the most according to claim 7, it is characterised in that: described according to extraction
The characteristics of image that goes out uses the SVM classifier trained to carry out intelligent mode identification, obtain the type belonging to Quick Response Code this
One step, comprising:
Corresponding SVM classifier is designed according to the characteristics of image extracted;
Use training sample that the SVM classifier of design is trained;
Carry out intelligent mode identification according to the SVM classifier the trained characteristics of image to extracting, obtain belonging to Quick Response Code
Type.
10. according to the implementation method of a kind of Quick Response Code identification chip described in claim 7,8 or 9, it is characterised in that: described
Carrying out code word extraction and decoding according to the result of intelligent mode identification, information that reduction Quick Response Code is preserved also carries out exporting this step
Suddenly, comprising:
The Quick Response Code code word extraction module of the call by result respective standard according to intelligent mode identification carries out code word extraction, obtains phase
The sequence of modules answered;
The decoding module calling respective standard according to the sequence of modules obtained decodes, the two-dimensional code data after being decompressed
Stream, described decoding process include but not limited to judgements of data query, error correction code word, code word error correcting and detecting, code word data judgement and
Code word decompresses;
Data stream after decompressing is sent to gui interface and shows, and exports.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610435710.6A CN106127093A (en) | 2016-06-17 | 2016-06-17 | A kind of Quick Response Code identification chip and its implementation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610435710.6A CN106127093A (en) | 2016-06-17 | 2016-06-17 | A kind of Quick Response Code identification chip and its implementation |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106127093A true CN106127093A (en) | 2016-11-16 |
Family
ID=57470748
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610435710.6A Pending CN106127093A (en) | 2016-06-17 | 2016-06-17 | A kind of Quick Response Code identification chip and its implementation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106127093A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106650411A (en) * | 2016-11-24 | 2017-05-10 | 天津津航计算技术研究所 | Verification system for cryptographic algorithms |
CN106845312A (en) * | 2016-12-28 | 2017-06-13 | 广州智慧城市发展研究院 | A kind of pre- determination methods of image in 2 D code quality |
CN107220577A (en) * | 2017-05-12 | 2017-09-29 | 广州智慧城市发展研究院 | A kind of Quick Response Code localization method and system based on machine learning |
CN107944321A (en) * | 2017-11-28 | 2018-04-20 | 努比亚技术有限公司 | A kind of image-recognizing method, terminal and computer-readable recording medium |
CN109615734A (en) * | 2018-11-16 | 2019-04-12 | 连尚(新昌)网络科技有限公司 | It is a kind of for unlocking the method and apparatus of vehicle arrangement |
CN109753839A (en) * | 2019-01-16 | 2019-05-14 | 上海龙旗科技股份有限公司 | Two dimensional code identifies optimization method and equipment |
CN109858305A (en) * | 2019-01-17 | 2019-06-07 | 柳州康云互联科技有限公司 | A kind of two dimensional code positioning identification system and method based on deep learning |
CN113468446A (en) * | 2021-07-14 | 2021-10-01 | 重庆允成互联网科技有限公司 | Method, system and equipment for supporting identification of third-party two-dimensional code data |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7353484B1 (en) * | 2004-02-09 | 2008-04-01 | Altera Corporation | Methods and apparatus for variable latency support |
TW200841686A (en) * | 2007-04-02 | 2008-10-16 | Simpleact Inc | Two dimensional barcode recognition system on mobile phone |
US20100025472A1 (en) * | 2008-07-29 | 2010-02-04 | Xerox Corporation | Colored barcode decoding |
CN102693409A (en) * | 2012-05-18 | 2012-09-26 | 四川大学 | Method for quickly identifying two-dimension code system type in images |
CN103279776A (en) * | 2013-05-06 | 2013-09-04 | 中山大学 | Chip structure applicable to two-dimensional bar code decoding chip |
CN103336939A (en) * | 2013-06-21 | 2013-10-02 | 张家港保税区润桐电子技术研发有限公司 | System and method for identifying two-dimension code by using mobile phone |
CN104616010A (en) * | 2014-12-04 | 2015-05-13 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | Straight line detection method based on Hough transformation and improved least square method |
CN105654004A (en) * | 2015-12-30 | 2016-06-08 | 福建联迪商用设备有限公司 | Intelligent gating barcode decoding method and chip |
CN205899553U (en) * | 2016-06-17 | 2017-01-18 | 广州中大微电子有限公司 | Two -dimensional code recognition chip |
-
2016
- 2016-06-17 CN CN201610435710.6A patent/CN106127093A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7353484B1 (en) * | 2004-02-09 | 2008-04-01 | Altera Corporation | Methods and apparatus for variable latency support |
TW200841686A (en) * | 2007-04-02 | 2008-10-16 | Simpleact Inc | Two dimensional barcode recognition system on mobile phone |
US20100025472A1 (en) * | 2008-07-29 | 2010-02-04 | Xerox Corporation | Colored barcode decoding |
CN102693409A (en) * | 2012-05-18 | 2012-09-26 | 四川大学 | Method for quickly identifying two-dimension code system type in images |
CN103279776A (en) * | 2013-05-06 | 2013-09-04 | 中山大学 | Chip structure applicable to two-dimensional bar code decoding chip |
CN103336939A (en) * | 2013-06-21 | 2013-10-02 | 张家港保税区润桐电子技术研发有限公司 | System and method for identifying two-dimension code by using mobile phone |
CN104616010A (en) * | 2014-12-04 | 2015-05-13 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | Straight line detection method based on Hough transformation and improved least square method |
CN105654004A (en) * | 2015-12-30 | 2016-06-08 | 福建联迪商用设备有限公司 | Intelligent gating barcode decoding method and chip |
CN205899553U (en) * | 2016-06-17 | 2017-01-18 | 广州中大微电子有限公司 | Two -dimensional code recognition chip |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106650411A (en) * | 2016-11-24 | 2017-05-10 | 天津津航计算技术研究所 | Verification system for cryptographic algorithms |
CN106845312A (en) * | 2016-12-28 | 2017-06-13 | 广州智慧城市发展研究院 | A kind of pre- determination methods of image in 2 D code quality |
CN106845312B (en) * | 2016-12-28 | 2020-08-25 | 广州智慧城市发展研究院 | Two-dimensional code image quality pre-judgment method |
CN107220577A (en) * | 2017-05-12 | 2017-09-29 | 广州智慧城市发展研究院 | A kind of Quick Response Code localization method and system based on machine learning |
CN107944321A (en) * | 2017-11-28 | 2018-04-20 | 努比亚技术有限公司 | A kind of image-recognizing method, terminal and computer-readable recording medium |
CN109615734A (en) * | 2018-11-16 | 2019-04-12 | 连尚(新昌)网络科技有限公司 | It is a kind of for unlocking the method and apparatus of vehicle arrangement |
CN109753839A (en) * | 2019-01-16 | 2019-05-14 | 上海龙旗科技股份有限公司 | Two dimensional code identifies optimization method and equipment |
CN109858305A (en) * | 2019-01-17 | 2019-06-07 | 柳州康云互联科技有限公司 | A kind of two dimensional code positioning identification system and method based on deep learning |
CN113468446A (en) * | 2021-07-14 | 2021-10-01 | 重庆允成互联网科技有限公司 | Method, system and equipment for supporting identification of third-party two-dimensional code data |
CN113468446B (en) * | 2021-07-14 | 2023-08-01 | 重庆允丰科技有限公司 | Method, system and equipment for supporting identification of third party two-dimensional code data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106127093A (en) | A kind of Quick Response Code identification chip and its implementation | |
Liu et al. | Recognition of QR Code with mobile phones | |
Liu et al. | A variational pan-sharpening method based on spatial fractional-order geometry and spectral–spatial low-rank priors | |
Zhang et al. | Demeshnet: Blind face inpainting for deep meshface verification | |
US10185857B2 (en) | Devices, systems, and methods for reading barcodes | |
CN110796080B (en) | Multi-pose pedestrian image synthesis algorithm based on generation countermeasure network | |
EP2393037B1 (en) | Data matrix decoding chip and decoding method thereof | |
Zamberletti et al. | Robust angle invariant 1d barcode detection | |
CN110427792B (en) | Reading matching method of shielding two-dimensional code and application thereof | |
Lin et al. | Real-time automatic recognition of omnidirectional multiple barcodes and dsp implementation | |
CN110147864A (en) | The treating method and apparatus of coding pattern, storage medium, electronic device | |
Liu et al. | Automatic recognition algorithm of quick response code based on embedded system | |
Yang et al. | Multi-scale bidirectional fcn for object skeleton extraction | |
Huang et al. | Data matrix code location based on finder pattern detection and bar code border fitting | |
Li et al. | Learning disentangling and fusing networks for face completion under structured occlusions | |
CN113111674A (en) | Aztec code positioning and decoding method, system, equipment and storage medium | |
Wang et al. | Block sparse representation for pattern classification: theory, extensions and applications | |
Zamberletti et al. | Neural 1D barcode detection using the Hough transform | |
Chai et al. | Local salient patterns—A novel local descriptor for face recognition | |
Lopez-Rincon et al. | Binary Large Object‐Based Approach for QR Code Detection in Uncontrolled Environments | |
CN101923632B (en) | Maxi Code bar code decoding chip and decoding method thereof | |
CN104346596A (en) | Identification method and identification device for QR (Quick Response) code | |
Narayan et al. | Barcode recognition from video by combining image processing and xilinx | |
Liu et al. | Bar code recognition in complex scenes by camera phones | |
Liu et al. | Boosting semantic segmentation via feature enhancement |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20190417 Address after: 510800 No. 88 Tiangui Road, Huadu District, Guangzhou City, Guangdong Province Applicant after: Guangzhou Sysur Microelectronics, Inc. Address before: 510800 No. 88 Tiangui Road, Huadu District, Guangzhou City, Guangdong Province Applicant before: Guangzhou Sysur Microelectronics, Inc. Applicant before: GUANGZHOU SMART CITY DEVELOPMENT RESEARCH INSTITUTE |
|
TA01 | Transfer of patent application right |