CN106529523B - Aluminum-plastic blister image restoration and sequence number recognition methods on a kind of High Speed Transfer band - Google Patents
Aluminum-plastic blister image restoration and sequence number recognition methods on a kind of High Speed Transfer band Download PDFInfo
- Publication number
- CN106529523B CN106529523B CN201611009704.0A CN201611009704A CN106529523B CN 106529523 B CN106529523 B CN 106529523B CN 201611009704 A CN201611009704 A CN 201611009704A CN 106529523 B CN106529523 B CN 106529523B
- Authority
- CN
- China
- Prior art keywords
- image
- camera
- drug
- sequence number
- blister package
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/242—Aligning, centring, orientation detection or correction of the image by image rotation, e.g. by 90 degrees
Abstract
The image restoration of aluminium-plastic blister package drugs and sequence number recognition methods, this method are related to image restoration and field of image recognition on a kind of High Speed Transfer band, solve aluminium-plastic blister package drugs problem of image blurring on High Speed Transfer band.It include an industrial digital camera, light source, photoelectric proximity switch and photoelectric angular encoder in this method system.The specific steps of this method are as follows: camera acquires continuous two target clear images with maximum frame per second when conveyer belt low-speed motion;It is calculated between two images using joint transform related algorithm as shifting amount and many experiments averaged;Conveyer belt high-speed motion increases camera exposure time acquisition drug blurred picture;Using being obtained before as shifting amount average value derives the point spread function of fuzzy drug image;Clear drug image is obtained using Lucy-Richardson algorithm restored image;Finally utilize support vector machine method identification sequence number.This method is simple and image restoration effect is good, and recognition accuracy is high.
Description
Technical field
The present invention relates to motion blur image restorations and field of image recognition, and in particular to plastic-aluminum on a kind of High Speed Transfer band
The image restoration and sequence number recognition methods of blister package drug.
Background technique
Plate containing medicines generally use aluminium-plastic bubble plate packing, and this packaged form advantage is more, wherein being some blister package aluminium
Foil surface is printed on explanatory note and sequence number, improves the safety of dispensing.
In process of production, drug sequence number mainly plays mark action.It, can according to product batch number and the corresponding record of production
To trace the history in this batch of products material source, drug forming process;It, can be with according to sales figure after drug forms finished product
Trace the market whereabouts of drug;Drug enters the quality condition behind market;It can control or recycle when needed this batch of medicine
Product.So will be recorded to drug sequence number during pharmaceutical production, packaging and dispensing etc., manual record mode efficiency
Low, large labor intensity, is just replaced by machine vision technique step by step at present, replaces manual identified with the mode of image recognition.
Pharmaceutical production, packaging and dispensing etc. are usually to carry out on a moving belt, this is just that machine vision method brings difficulty
Degree, because image captured by camera will will appear motion blur and show when speed of moving body is too fast relative to camera frame per second
As.Existing method generally uses high frame per second camera to acquire the method for image with maximum frame per second to avoid problem of image blurring.But
It is that extraneous additional light source is needed to increase brightness of image when camera acquires image with maximum frame per second, otherwise image is partially dark or even complete
It is black, however strong illumination cannot be used on certain special leechdom production lines, and extraneous polishing be easy to cause plastic-aluminum surface
Mirror-reflection is to influence image acquisition quality.
Summary of the invention
The present invention is to solve the problems, such as that image is fuzzy in drug Image Acquisition on high-speed conveyor and drug obtains hardly possible in real time,
The image restoration and sequence number recognition methods of aluminium-plastic blister package drugs on a kind of High Speed Transfer band are provided.The present invention passes through calculating
Point spread function restores fuzzy drug image, solves the problems, such as to beat strong light when shooting using the method for increasing the time for exposure,
And the sequence number on drug image is identified by support vector machine method.Specific step is as follows:
(1) industrial digital camera and light source etc. are built on drug conveyer belt, complete image capturing system, camera and transmission
Band Relative vertical, installation photoelectric proximity switch realizes that camera is triggered when drug appears under camera lens obtains image, and pacifies
Dress photoelectric angular encoder is used to measure the movement velocity of conveyer belt.
(2) conveyer belt low speed V1Movement, camera are adjusted to maximum frame per second i.e. time for exposure P1, camera exposure time and conveyer belt
Movement velocity meets following formula:
Wherein K is camera pixel size, and β is image enlargement ratio, in this case camera shooting clear image, is opened
Light intensity light source, photoelectric proximity switch are closed, camera outer signal generator, and signal generator generates the trigger signal that the period is T
It triggers camera and obtains two continuous target clear image I1, I2, adjacent two targets are calculated using joint transform related algorithm
Body image slices shifting amount Δ x, Δ y,
(Δ x, Δ y)=JTC (I1, I2)
Wherein JTC is the relevant computational algorithm of joint transform, and joint transform correlation has good noise robustness, for
The image of low signal-to-noise ratio can also obtain the image motion measurement of sub-pixel precision, solve upon exposure between adjust to it is very low when obtain it is low
The problem of signal-to-noise ratio image.Repeat the average value of step (2) n computations picture shifting amountWithFormula is as follows:
(3) speed V when conveyer belt is to work normally2Movement weakens light source, increases the camera exposure time to P2, camera exposure
Time should adjust the half that gray value of image captured by camera is maximum gradation value, open photoelectric proximity switch.At this time because
It is improved for conveyer belt movement velocity and the increase of camera exposure time, the aluminium-plastic blister package drugs image that camera takes is mould
Paste image Ib。
(4) it is extrapolated according to obtained in step (2) as shifting amount average value and obscures drug image I in step (3)bPoint expand
Function h (x, y) is dissipated, calculation formula is as follows:
Wherein L is blurred length and θ is fuzzy angle, if blurred length is greater than 10 pixels, often beyond recovery
The limit can suitably reduce time for exposure P at this time2.Lucy-Richardson algorithm restored image is finally utilized, clear aluminium is obtained
Mould blister package drug image.
(5) on the clear aluminium-plastic blister package drugs image for utilizing the method identification step (4) of support vector machines to obtain
Sequence number first has to carry out character normalization, the feature for sequence number character recognition has periphery to preferably extract feature
Profile and thick meshed feature later disturb training sample by the realization of Bagging (Bootstrap Aggregating) algorithm
It is dynamic, to generate the base classifier with otherness, then carry out the integrated study of base classifier, finally with it is trained support to
Amount machine classifier identifies the sequence number on clear aluminium-plastic blister package drugs image.
Beneficial effects of the present invention: on a kind of High Speed Transfer band of the invention the image restoration of aluminium-plastic blister package drugs and
Sequence number recognition methods is only utilized an industrial camera and just completes image restoration, and is not required in drug shooting process
Intense light source is wanted, solves the problems, such as that certain special drugs cannot be easy to cause mirror-reflection by strong illumination and polishing.The present invention
Image collecting device it is simple, image restoration effect is good, drug comment on aluminium-plastic blister package drugs etc. clearly present,
And the information of drug is obtained by support vector machine method identification sequence number content.
Detailed description of the invention
Fig. 1 is the image restoration of aluminium-plastic blister package drugs and sequence number identification side on a kind of High Speed Transfer band of the invention
The flow chart of method;
Fig. 2 is to establish joint image schematic diagram in joint transform related algorithm;
Fig. 3 is total system assembling schematic diagram of the invention;
Fig. 4 is support vector machines identification sequence number flow chart.
Specific implementation step
With reference to the accompanying drawings and in conjunction with the embodiments to aluminium-plastic blister package drugs on a kind of High Speed Transfer band of the invention
Image restoration and sequence number recognition methods are described further.
Step 1: referring to Fig. 3, industrial digital camera (1) and light source (2) etc. are built on drug conveyer belt (5), completes figure
As acquisition system, camera (1) and conveyer belt (5) Relative vertical, installation photoelectric proximity switch (3) are realized when drug appears in camera
(1) under camera lens image is obtained when, and photoelectric angular encoder (4) are installed for measuring the movement velocity of conveyer belt (5).
Step 2: the time for exposure of reduction industrial camera (1) to limit time P as far as possible1, i.e. camera (1) reaches largest frames
Rate, increasing light source (2) brightness makes target (6) blur-free imaging on conveyer belt (3), opens conveyer belt (5) telecontrol equipment, passes at this time
Send band (5) with low speed V1Movement, camera (1) time for exposure and conveyer belt (5) movement velocity meet following formula:
Wherein K is camera pixel size, and β is image enlargement ratio, and camera can be with shooting clear target in this case
Object image, photoelectric proximity switch (3) are closed, camera (1) outer signal generator, and signal generator generates the triggering that the period is T
Signal triggers camera (1) and obtains two continuous target (6) clear image I1, I2, calculated using joint transform related algorithm adjacent
Two target object image slices shifting amount Δ x, Δ y,
(Δ x, Δ y)=JTC (I1, I2)
Wherein JTC is joint transform related algorithm, the method is as follows:
It is assumed that target image I2(x, y) is relative to reference picture I1The displacement of (x, y) on the direction x, y is Δ x, Δ respectively
The center of the two is placed in (- a, 0) and (a, 0), obtains joint input image j (x, y) by y as shown in Fig. 2 are as follows:
J (x, y)=I1(x+a, y)+I2(x-a, y)=I1(x+a, y)+I1(x-a- Δ x, y- Δ y)
A Fourier transformation is carried out to it and takes its power spectrum, and calculation is derived by following formula:
| J (u, v) |2=2 | I '1(u, v) |2+2|I′1(u, v) |2cos[2π(Δx+2a)u+2πΔyv]
I ' in the formula1(u, v) is I1The Fourier transformation of (x, y), | * | it is modulus operation.Fourier is done again to power spectrum
Transformation, obtains joint correlation output:
C (x, y)=2Crr(x, y)+Crr(x+2a+ Δ x, y+ Δ y)+Crr(x-2a- Δ x, y- Δ y)
C in the formularr(x, y) is I1The auto-correlation function of (x, y), CrrThe peak value of (x, y) is located at (0,0), and Crr(x+2a
+ Δ x, y+ Δ y) and Crr(peak value of x-2a- Δ x, y- Δ y) is located at (- 2a- Δ x ,-Δ y) and (2a+ Δ x, Δ y), root
According to Crr(x+2a+ Δ x, y+ Δ y) and Crr(two peak position relationships of x-2a- Δ x, y- Δ y) can obtain target image
Relative to the displacement of reference picture, (Δ x, Δ y), i.e., picture between the two are moved, and the position coordinates of two correlation peaks pass through
Centroid algorithm is calculated, and joint transform correlation has good noise robustness, can also obtain for the image of low signal-to-noise ratio
The image motion measurement of sub-pixel precision, very good solution upon exposure between adjust to it is very low when acquisition low signal-to-noise ratio (SNR) images ask
Topic.Repeat the average value of step (2) n computations picture shifting amountWithFormula is as follows:
Step 3: speed V when conveyer belt (5) is to work normally2Movement weakens light source (2), when increasing camera (1) exposure
Between arrive P2, camera (1) time for exposure should adjust gray value of image captured by camera (1) close to the half of maximum gradation value,
It opens photoelectric proximity switch (3).At this time because conveyer belt (5) movement velocity improves and the increase of camera (1) time for exposure, phase
Aluminium-plastic blister package drugs (6) image that machine (1) takes is blurred picture Ib。
Step 4: it is extrapolated according to obtained in step 2 as shifting amount average value and obscures drug image I in step 3bPoint
Spread function h (x, y), calculation formula are as follows:
Wherein L is blurred length and θ is fuzzy angle, utilizes Lucy-Richardson algorithm restored image later,
Lucy- Richardson algorithm assumes that image obeys Poission distribution, is estimated using maximum likelihood, is a kind of base
In the iterative algorithm of Bayesian analysis, estimate that iterative equation is as follows:
Wherein IbFormer blurred picture is represented, g represents restored image, and h represents point spread function, and k represents the number of iterations.It utilizes
MATLAB software rejuvenation image, then h=fspecial (' motion ', L, θ), g=deconvlucy (Ib, h, 5).This is just completed
Recovery operation to aluminum-plastic blister drug image on conveyer belt.
Step 5: on the clear aluminium-plastic blister package drugs image obtained using the method identification step four of support vector machines
Sequence number first have to carry out character normalization, the feature for sequence number character recognition has outer to preferably extract feature
Profile and thick meshed feature are enclosed, is realized later by Bagging (Bootstrap Aggregating) algorithm to training sample
Disturbance to generate the base classifier with otherness, then carries out the integrated study of base classifier, finally with trained support
Vector machine classifier identifies the sequence number on clear aluminium-plastic blister package drugs image.
Claims (4)
1. the image restoration and sequence number recognition methods of aluminium-plastic blister package drugs on a kind of High Speed Transfer band, it is characterised in that packet
Include following steps:
Step 1 builds image collecting device on drug conveyer belt (5), and image collecting device includes an industrial digital camera
(1) it realizes with light source (2), camera (1) and conveyer belt (5) Relative vertical, installation photoelectric proximity switch (3) when drug (6) occur
Camera (1) synchronous photo taking is triggered when under camera (1) camera lens, and photoelectric angular encoder (4) are installed and are used to measure conveyer belt (5) fortune
Dynamic speed;
Step 2, conveyer belt (5) low speed V1Movement, camera (1) are adjusted to maximum frame per second i.e. time for exposure P1, it opens light intensity light source (2),
Photoelectric proximity switch (3) is closed, camera (1) outer signal generator, and signal generator generates the trigger signal that the period is T and triggers
Camera (1) obtains two continuous target (6) clear images, calculates between image as shifting amount, step 2 is repeated several times and acquires image
The average value of the direction x and y picture shifting amount is respectivelyWith
Speed V when step 3, conveyer belt (5) are to work normally2Movement weakens light source (2), increases camera (1) time for exposure to P2,
Drug gray value of image captured by camera (1) is the half of maximum gradation value, is opened photoelectric proximity switch (3), and plastic-aluminum is obtained
The blurred picture of blister package drug;
The average value of step 4, the direction the image x and y multiple groups picture shifting amount according to step 2WithExtrapolate mould in step 3
The point spread function of poultice product image, restored image obtain clear aluminium-plastic blister package drugs image;
Sequence in step 5, the clear aluminium-plastic blister package drugs image obtained using the method identification step 4 of support vector machines
Number.
2. the image restoration of aluminium-plastic blister package drugs and sequence number are known on a kind of High Speed Transfer band according to claim 1
Other method, it is characterised in that: target is five great circle targets in the step 2, is calculated between adjacent image as shifting amount method
For joint transform related algorithm.
3. the image restoration of aluminium-plastic blister package drugs and sequence number are known on a kind of High Speed Transfer band according to claim 1
Other method, it is characterised in that: drug image point spread function calculation formula is obscured in the step 4 are as follows:
Wherein L is blurred length, and θ is fuzzy angle, and h (x, y) is point spread function, and the method for restored image uses Lucy-
Richardson algorithm.
4. the image restoration of aluminium-plastic blister package drugs and sequence number are known on a kind of High Speed Transfer band according to claim 1
Other method, it is characterised in that: in the step 5 support vector machine method include: character normalization, feature extraction, support to
The training of amount machine classifier, the identification of support vector machine classifier.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611009704.0A CN106529523B (en) | 2016-11-14 | 2016-11-14 | Aluminum-plastic blister image restoration and sequence number recognition methods on a kind of High Speed Transfer band |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611009704.0A CN106529523B (en) | 2016-11-14 | 2016-11-14 | Aluminum-plastic blister image restoration and sequence number recognition methods on a kind of High Speed Transfer band |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106529523A CN106529523A (en) | 2017-03-22 |
CN106529523B true CN106529523B (en) | 2019-04-02 |
Family
ID=58353373
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611009704.0A Active CN106529523B (en) | 2016-11-14 | 2016-11-14 | Aluminum-plastic blister image restoration and sequence number recognition methods on a kind of High Speed Transfer band |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106529523B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107203770B (en) * | 2017-05-27 | 2020-07-31 | 上海航天控制技术研究所 | Optical strapdown seeker image tracking method |
CN111626946B (en) * | 2020-04-23 | 2023-06-06 | 武汉理工大学 | Motion blur kernel measuring method for high-speed material transmission visual detection system |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7848587B2 (en) * | 2006-06-30 | 2010-12-07 | Eastman Kodak Company | Image-processing system and image-processing program |
CN103310427A (en) * | 2013-06-24 | 2013-09-18 | 中国科学院长春光学精密机械与物理研究所 | Image super-resolution and image quality enhancement method |
CN104299202A (en) * | 2014-10-25 | 2015-01-21 | 中国科学院光电技术研究所 | Out-of-focus blurred image blind restoration method based on medium frequency |
-
2016
- 2016-11-14 CN CN201611009704.0A patent/CN106529523B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7848587B2 (en) * | 2006-06-30 | 2010-12-07 | Eastman Kodak Company | Image-processing system and image-processing program |
CN103310427A (en) * | 2013-06-24 | 2013-09-18 | 中国科学院长春光学精密机械与物理研究所 | Image super-resolution and image quality enhancement method |
CN104299202A (en) * | 2014-10-25 | 2015-01-21 | 中国科学院光电技术研究所 | Out-of-focus blurred image blind restoration method based on medium frequency |
Non-Patent Citations (3)
Title |
---|
Blind recovery of truncated blurred image using adaptive masking method;Nobuhito Ishihara etc.;《Proceedings of the winter international synposium on Information and communication technologies》;20041231;第1-6页 |
CCD光学成像系统的点扩散函数及其在亚像素边缘定位中的应用;安岗;《中国优秀硕士学位论文全文数据库 信息科技辑》;20081015(第10期);正文第1-79页 |
运动模糊图像点扩散函数的频谱估计法;贤光等;《液晶与显示》;20141031;第29卷(第5期);第751-755页 |
Also Published As
Publication number | Publication date |
---|---|
CN106529523A (en) | 2017-03-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP3311364B1 (en) | A method and system for tracking an object | |
CN105608671B (en) | A kind of image split-joint method based on SURF algorithm | |
US10255682B2 (en) | Image detection system using differences in illumination conditions | |
CN104089697B (en) | Real-time online visual vibration measurement method based on thread pool concurrent technology | |
US20170124693A1 (en) | Pose Estimation using Sensors | |
CN109583285A (en) | Object identifying method | |
CN106529523B (en) | Aluminum-plastic blister image restoration and sequence number recognition methods on a kind of High Speed Transfer band | |
DE60227111D1 (en) | PICTURE PROCESSING METHOD AND SYSTEM FOR EXPLORING STRUCTURES IN HARMFUL PICTURES | |
JP6261266B2 (en) | Moving body detection device | |
GB0305304D0 (en) | Scanning apparatus and method | |
US9363427B2 (en) | Device and method for calibrating a temporal contrast sensor with a frame-based camera sensor | |
EP3005231B1 (en) | Method and device for counting objects in image data in frames, computer program product | |
US8693784B2 (en) | Eyelid detection device and program | |
CN104054110B (en) | Collision time according to image sensing | |
CN105637852B (en) | A kind of image processing method, device and electronic equipment | |
CN106885622A (en) | A kind of big visual field multiple spot three-dimensional vibrating measuring method | |
FR2998363B1 (en) | METHOD FOR DETERMINING, IN A 3D FIXED REFERENTIAL, THE LOCATION OF A MOVING GEAR, ASSOCIATED DEVICE AND COMPUTER PROGRAM | |
CN112686928A (en) | Moving target visual tracking method based on multi-source information fusion | |
US20140375821A1 (en) | Detection system | |
Seets et al. | Motion adaptive deblurring with single-photon cameras | |
WO2012002731A3 (en) | Sensing device and method for moving object and virtual golf simulation device using the same | |
CN104331907B (en) | A kind of method based on ORB feature detections measurement bearer rate | |
CN109514922A (en) | A kind of vision detection system of the easy-torn line of paper extraction packaging | |
US11410459B2 (en) | Face detection and recognition method using light field camera system | |
JP2004109009A (en) | Object angular speed measuring device and object angular speed measuring method |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |