CN103581507A - Method dynamically adjusting threshold value through mean square error in de-noising algorithm - Google Patents

Method dynamically adjusting threshold value through mean square error in de-noising algorithm Download PDF

Info

Publication number
CN103581507A
CN103581507A CN201310365141.9A CN201310365141A CN103581507A CN 103581507 A CN103581507 A CN 103581507A CN 201310365141 A CN201310365141 A CN 201310365141A CN 103581507 A CN103581507 A CN 103581507A
Authority
CN
China
Prior art keywords
threshold value
frame
value
filtering
video
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
Application number
CN201310365141.9A
Other languages
Chinese (zh)
Inventor
李红波
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Liu Jing
Original Assignee
CHENGDU YUNYING SCIENCE & TECHNOLOGY Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by CHENGDU YUNYING SCIENCE & TECHNOLOGY Co Ltd filed Critical CHENGDU YUNYING SCIENCE & TECHNOLOGY Co Ltd
Priority to CN201310365141.9A priority Critical patent/CN103581507A/en
Publication of CN103581507A publication Critical patent/CN103581507A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Image Analysis (AREA)

Abstract

The invention discloses a method dynamically adjusting a threshold value through a mean square error in the de-noising algorithm. The mean square error of the inter-frame image absolute error mean value (MAD) is calculated, used for representing current motion states of video sequences and calculated to obtain the next frame threshold value, so that when the de-noising processing is carried out on each frame video image, the spatial domain is subdivided into a movable block and a static block, and the movable block and the static block are respectively processed. Meanwhile, according to the time domain, the threshold value is calculated through the mean square error of the video inter-frame MAD, and the motion conditions of the video sequences are dynamically analyzed so that the threshold value can be calculated accurately.

Description

In a kind of Denoising Algorithm, by mean square deviation, dynamically adjust the method for threshold value
Technical field
The present invention relates to technical field of image processing, especially relate to a kind of dynamically adjust threshold value in Denoising Algorithm by mean square deviation method.
Background technology
Image is easily subject to external interference and produces some impulsive noises in formation, transmission, reception and processing procedure, thereby in image, produces the point of black, white, is called salt-pepper noise.The effect that salt-pepper noise can further affect rim detection, image is cut apart applies with the later image such as feature extraction.Therefore, improving signal to noise ratio is one of vital task of image processing.
At present, the method for image denoising is mainly divided into airspace filter, time-domain filtering and space-time in conjunction with filtering three classes.The normal employing of airspace filter is weighted average mode filtering to neighbor pixel, so airspace filter removal noise effects is poor and can sacrifice image high frequency detail section, makes image produce distortion.Time-domain filtering, owing to having considered video image correlation in time, adopts IIR filtering algorithm, carries out inter frame image weighting processing, and image sequence is more similar, and correlation is stronger, and denoising effect is better.But for moving image, moving target can produce the time domain bloomings such as artifact.Space-time adopts stationary part in video image is carried out to time-domain filtering in conjunction with filtering, and motion parts image is carried out to airspace filter, and this has just solved effectively, and airspace filter removal noise is poor can produce the time domain fuzzy problems such as artifact with time-domain filtering moving target.But how to judge image static in video image and motion, this problem has to be solved.
Existing method adopts prediction threshold method to judge conventionally, and when interframe block of pixels difference is greater than this threshold value, this area pixel is moving mass, and this piece is carried out to airspace filter.When being less than this threshold value, judge that this area pixel is static block, carries out time-domain filtering to this piece.But there are the following problems for existing mode: one, in image freeze region, when noise is larger, interframe block of pixels difference is still larger, this piece will be mistaken for to moving mass and carry out airspace filter.Two, in video image motion region, larger if threshold value is selected, this piece is judged as to static block and carries out time-domain filtering, thereby make image time domain fuzzy.
Summary of the invention
For the problems referred to above, the invention discloses a kind of dynamically adjust threshold value in Denoising Algorithm by mean square deviation method, the method can effectively address the above problem.
In order to reach above-mentioned technique effect, the present invention adopts following technical scheme: in a kind of Denoising Algorithm, by mean square deviation, dynamically adjust the method for threshold value, the method comprises the steps:
A. the raw video signal of input is divided into the frame of video of serial, with the Kuai Wei unit of 8x8 size, processes, judge whether this frame of video is denoising start frame, does not if it is process, former state output; If not, enter next step;
B. to current pending, carry out IIR filtering, obtain piece PBx.y (t), IIR Filtering Formula is:
Figure 2013103651419100002DEST_PATH_IMAGE001
.
Wherein, filter factor b1=0.85, a1=0.15, PBx.y (t-1) is filtered of same position in former frame, CB x.y (t) is current pending;
C. piece PBx.y (t) after current pending CBx.y (t) and filtering is carried out to difference, obtain MAD, thereby obtain mean square deviation
Figure 527835DEST_PATH_IMAGE002
, formula is:
MAD?=?
Figure 2013103651419100002DEST_PATH_IMAGE003
?/64
Figure 453066DEST_PATH_IMAGE002
?=
Figure 615057DEST_PATH_IMAGE004
；
D.: by present frame threshold value and
Figure 942133DEST_PATH_IMAGE002
compare and judge, being greater than
Figure 852583DEST_PATH_IMAGE002
current block is moving mass, is less than
Figure 265109DEST_PATH_IMAGE002
current block is static block, and first processed frame is first set to an initial threshold, and follow-up this threshold value dynamically updates, and judges whether present frame finishes, and if not, returns to step b, if so, enters next step;
E. travel through a two field picture, obtain the attribute list Tc of each piece in a two field picture, and then these piece attribute lists Tc is carried out to the filtering of morphology opening operation, remove isolated attribute block, again obtain piece attribute list Tp in present frame,
Figure 2013103651419100002DEST_PATH_IMAGE005
Wherein B represents to carry out the structural element of opening operation, adopts the cross template of 3x3;
F. by the T that tables look-up p, moving mass is carried out to airspace filter, static block is carried out to IIR filtering, airspace filter adopts non-linear filtering, and the value of certain pixel in digital image sequence is replaced with the Mesophyticum of the value of each point in this vertex neighborhood, and medium filtering is defined as follows:
If the neighborhood territory pixel of object pixel is
Figure 214480DEST_PATH_IMAGE006
, the large minispread according to value of n number is as follows:
Figure 2013103651419100002DEST_PATH_IMAGE007
After filtering, the value of object pixel is:
Figure 333745DEST_PATH_IMAGE008
Interframe IIR filtering implementation is as follows:
PB t(i,j)?=?0.8*PB t-1?(i,j)?+?0.2*CB ?t(i,j)；
G. calculate present frame MAD value variance, variance computational methods are as follows:
Figure 2013103651419100002DEST_PATH_IMAGE009
?；
H. predict next frame threshold value, dynamic noise in actual video frame, mostly is white noise, and obeying average is 0, and variance is
Figure 480343DEST_PATH_IMAGE010
gaussian Profile, be located at the appearance of not moving of certain hour scope, the difference between image is mainly caused have by noise:
Figure 2013103651419100002DEST_PATH_IMAGE011
In formula representing the value of scene, because of without motion, is definite value, and n (t) represents noise, is changing value, and obeying average is 0, and variance is
Figure 883959DEST_PATH_IMAGE010
gaussian Profile, inter-frame difference image is defined as:
Figure 2013103651419100002DEST_PATH_IMAGE013
Suppose to exist without motion, calculate difference image and face mutually the probability that 3 pieces are all greater than threshold value TH or are all less than threshold value TH,
Figure 733098DEST_PATH_IMAGE014
Facing mutually the probability that 3 pieces are greater than threshold value is:
Figure 2013103651419100002DEST_PATH_IMAGE015
Suppose
Figure 296934DEST_PATH_IMAGE010
be 5, when TH gets 1 times time, calculating probability is 0.004, when getting TH, is 2 times time, calculating probability is 1.178x
Figure 754570DEST_PATH_IMAGE016
, this is a minimum probability event, in like manner, gets one and is a bit larger tham variance
Figure 551625DEST_PATH_IMAGE010
threshold value, be also a minimum probability event, consider the feature of actual video sequence, this algorithm is got threshold value
Figure 2013103651419100002DEST_PATH_IMAGE017
;
I. return to step b, by the sequence of steps of b, c, d, e, f, g, h, circulate, traversal frame of video, obtains the sequence of frames of video after denoising.
In described steps d, Initial Hurdle is 4.
Beneficial effect of the present invention is: by calculating the mean square deviation of inter frame image absolute error average (MAD)
Figure 98144DEST_PATH_IMAGE010
, with mean square deviation
Figure 458587DEST_PATH_IMAGE010
characterize current video sequence motion state and calculate next frame threshold value.Make in to every frame video image denoising, in spatial domain, segmented moving mass and static block is processed respectively.Meanwhile, time-domain, with the mean square deviation of frame of video interframe MAD
Figure 469268DEST_PATH_IMAGE010
calculate thresholding, dynamically analyzed the motion conditions of video sequence, make the calculating of thresholding more accurate.
 
Accompanying drawing explanation
Fig. 1 is flow chart of the present invention.
 
Embodiment
Below in conjunction with flow chart, the present invention is described in more detail.
A. the raw video signal of input is divided into the frame of video of serial, with the Kuai Wei unit of 8x8 size, processes, judge whether this frame of video is denoising start frame, does not if it is process, former state output; If not, enter next step;
B. to current pending, carry out IIR filtering, obtain piece PBx.y (t), IIR Filtering Formula is:
Figure 437224DEST_PATH_IMAGE001
Wherein, filter factor b1=0.85, a1=0.15, PBx.y (t-1) is filtered of same position in former frame, CB x.y (t) is current pending;
C. piece PBx.y (t) after current pending CBx.y (t) and filtering is carried out to difference, obtain MAD, thereby obtain mean square deviation
Figure 471039DEST_PATH_IMAGE002
, formula is:
MAD?=?
Figure 917064DEST_PATH_IMAGE003
?/64
Figure 47831DEST_PATH_IMAGE002
?= ；
D.: by present frame threshold value and
Figure 458532DEST_PATH_IMAGE002
compare and judge, being greater than
Figure 708248DEST_PATH_IMAGE002
current block is moving mass, is less than
Figure 631204DEST_PATH_IMAGE002
current block is static block, and first processed frame is first set to an initial threshold, and follow-up this threshold value dynamically updates, and judges whether present frame finishes, and if not, returns to step b, if so, enters next step;
E. travel through a two field picture, obtain the attribute list Tc of each piece in a two field picture, and then these piece attribute lists Tc is carried out to the filtering of morphology opening operation, remove isolated attribute block, again obtain piece attribute list Tp in present frame,
Wherein B represents to carry out the structural element of opening operation, adopts the cross template of 3x3;
F. by the T that tables look-up p, moving mass is carried out to airspace filter, static block is carried out to IIR filtering, airspace filter adopts non-linear filtering, and the value of certain pixel in digital image sequence is replaced with the Mesophyticum of the value of each point in this vertex neighborhood, and medium filtering is defined as follows:
If the neighborhood territory pixel of object pixel is
Figure 198638DEST_PATH_IMAGE006
, the large minispread according to value of n number is as follows:
After filtering, the value of object pixel is:
Figure 91824DEST_PATH_IMAGE008
Interframe IIR filtering implementation is as follows:
PB t(i,j)?=?0.8*PB t-1?(i,j)?+?0.2*CB ?t(i,j)；
G. calculate present frame MAD value variance, variance computational methods are as follows:
?；
H. predict next frame threshold value, dynamic noise in actual video frame, mostly is white noise, and obeying average is 0, and variance is
Figure 802609DEST_PATH_IMAGE010
gaussian Profile, be located at certain hour scope, i.e. the adjacent two frames appearance of not moving, the difference between image is mainly caused have by noise:
In formula
Figure 57057DEST_PATH_IMAGE012
representing the value of scene, because of without motion, is definite value, and n (t) represents noise, is changing value, and obeying average is 0, and variance is
Figure 708618DEST_PATH_IMAGE010
gaussian Profile, inter-frame difference image is defined as:
Figure 753934DEST_PATH_IMAGE013
Suppose to exist without motion, calculate difference image and face mutually the probability that 3 pieces are all greater than threshold value TH or are all less than threshold value TH,
Facing mutually the probability that 3 pieces are greater than threshold value is:
Figure 635620DEST_PATH_IMAGE015
Suppose be 5, when TH gets 1 times
Figure 177645DEST_PATH_IMAGE010
time, calculating probability is 0.004, when getting TH, is 2 times
Figure 110966DEST_PATH_IMAGE010
time, calculating probability is 1.178x
Figure 717528DEST_PATH_IMAGE016
, this is a minimum probability event, in like manner, gets one and is a bit larger tham variance
Figure 976471DEST_PATH_IMAGE010
threshold value, be also a minimum probability event, consider the feature of actual video sequence, this algorithm is got threshold value
Figure 465221DEST_PATH_IMAGE017
;
I. return to step b, by the sequence of steps of b, c, d, e, f, g, h, circulate, traversal frame of video, obtains the sequence of frames of video after denoising.

Claims (2)

1. in Denoising Algorithm, by mean square deviation, dynamically adjust a method for threshold value, it is characterized in that, the method comprises the steps:
A. the raw video signal of input is divided into the frame of video of serial, with the Kuai Wei unit of 8x8 size, processes, judge whether this frame of video is denoising start frame, does not if it is process, former state output; If not, enter next step;
B. press from top to bottom, order from left to right scans successively, obtains current pending, and current pending PBx.y (t) carried out to IIR filtering, and IIR Filtering Formula is:
Figure 2013103651419100001DEST_PATH_IMAGE001
Wherein, filter factor b1=0.85, a1=0.15, PBx.y (t-1) is filtered of same position in former frame, CB x.y (t) is current pending;
C. piece PBx.y (t) after current pending CBx.y (t) and filtering is carried out to difference, obtain MAD, thereby obtain mean square deviation , formula is:
MAD?=?
Figure 2013103651419100001DEST_PATH_IMAGE003
?/64
Figure 830355DEST_PATH_IMAGE002
?=
Figure 993352DEST_PATH_IMAGE004
；
D.: by present frame threshold value and
Figure 534055DEST_PATH_IMAGE002
compare and judge, being greater than
Figure 177526DEST_PATH_IMAGE002
current block is moving mass, is less than
Figure 135861DEST_PATH_IMAGE002
current block is static block, and first processed frame is first set to an initial threshold, and follow-up this threshold value dynamically updates; Judge whether present frame finishes, if not, return to step b, process next piece, if so, enter next step;
E. travel through a two field picture, obtain the attribute list Tc of each piece in a two field picture, and then these piece attribute lists Tc is carried out to the filtering of morphology opening operation, remove isolated attribute block, again obtain piece attribute list Tp in present frame,
Figure DEST_PATH_IMAGE005
Wherein B represents to carry out the structural element of opening operation, adopts the cross template of 3x3;
F. by the T that tables look-up p, moving mass is carried out to airspace filter, static block is carried out to IIR filtering, airspace filter adopts non-linear filtering, and the value of certain pixel in digital image sequence is replaced with the Mesophyticum of the value of each point in this vertex neighborhood, and medium filtering is defined as follows:
If the neighborhood territory pixel of object pixel is
Figure 802466DEST_PATH_IMAGE006
, the large minispread according to value of n number is as follows:
Figure DEST_PATH_IMAGE007
After filtering, the value of object pixel is:
Figure 881280DEST_PATH_IMAGE008
Interframe IIR filtering implementation is as follows:
PB t(i,j)?=?0.8*PB t-1?(i,j)?+?0.2*CB ?t(i,j)；
G. calculate present frame MAD value variance, variance computational methods are as follows:
Figure DEST_PATH_IMAGE009
?；
H. predict next frame threshold value, dynamic noise in actual video frame, mostly is white noise, and obeying average is 0, and variance is
Figure 566208DEST_PATH_IMAGE010
gaussian Profile, establish the appearance of not moving of adjacent two frames, the difference between image is mainly caused have by noise:
Figure DEST_PATH_IMAGE011
In formula
Figure 9959DEST_PATH_IMAGE012
representing the value of scene, because of without motion, is definite value, and n (t) represents noise, is changing value, and obeying average is 0, and variance is
Figure 226177DEST_PATH_IMAGE010
gaussian Profile, inter-frame difference image is defined as:
Figure DEST_PATH_IMAGE013
Suppose to exist without motion, calculate difference image and face mutually the probability that 3 pieces are all greater than threshold value TH or are all less than threshold value TH,
Figure 797098DEST_PATH_IMAGE014
Facing mutually the probability that 3 pieces are greater than threshold value is:
Figure DEST_PATH_IMAGE015
Suppose
Figure 87265DEST_PATH_IMAGE010
be 5, when TH gets 1 times time, calculating probability is 0.004, when getting TH, is 2 times
Figure 467748DEST_PATH_IMAGE010
time, calculating probability is 1.178x
Figure 75315DEST_PATH_IMAGE016
, this is a minimum probability event, in like manner, gets one and is a bit larger tham variance
Figure 547885DEST_PATH_IMAGE010
threshold value, be also a minimum probability event, consider the feature of actual video sequence, this algorithm is got threshold value
Figure DEST_PATH_IMAGE017
;
I. return to step b, by the sequence of steps of b, c, d, e, f, g, h, circulate, traversal frame of video, obtains the sequence of frames of video after denoising.
2. by mean square deviation, dynamically adjust according to claim 1 the method for threshold value in a kind of Denoising Algorithm, it is characterized in that, in described steps d, Initial Hurdle is 4.
CN201310365141.9A 2013-08-21 2013-08-21 Method dynamically adjusting threshold value through mean square error in de-noising algorithm Pending CN103581507A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310365141.9A CN103581507A (en) 2013-08-21 2013-08-21 Method dynamically adjusting threshold value through mean square error in de-noising algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310365141.9A CN103581507A (en) 2013-08-21 2013-08-21 Method dynamically adjusting threshold value through mean square error in de-noising algorithm

Publications (1)

Publication Number Publication Date
CN103581507A true CN103581507A (en) 2014-02-12

Family

ID=50052356

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310365141.9A Pending CN103581507A (en) 2013-08-21 2013-08-21 Method dynamically adjusting threshold value through mean square error in de-noising algorithm

Country Status (1)

Country Link
CN (1) CN103581507A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105225244A (en) * 2015-10-22 2016-01-06 天津大学 Based on the noise detection method that minimum local mean square deviation calculates
CN110072034A (en) * 2018-01-23 2019-07-30 瑞昱半导体股份有限公司 Image treatment method and image processor
CN111191484A (en) * 2018-11-14 2020-05-22 普天信息技术有限公司 Method and device for recognizing human speaking in video image
CN113469893A (en) * 2020-05-08 2021-10-01 上海齐感电子信息科技有限公司 Method for estimating noise of image in video and video noise reduction method
CN114070959A (en) * 2021-10-31 2022-02-18 南京理工大学 FPGA-based video denoising hardware implementation method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090092337A1 (en) * 2007-09-07 2009-04-09 Takefumi Nagumo Image processing apparatus, image processing method, and computer program
CN102497497A (en) * 2011-12-05 2012-06-13 四川九洲电器集团有限责任公司 Method for dynamically adjusting threshold in image denoising algorithm
CN103067647A (en) * 2012-12-25 2013-04-24 四川九洲电器集团有限责任公司 Field programmable gata array (FPGA) based video de-noising method
CN103176425A (en) * 2013-03-13 2013-06-26 西北工业大学 Design method of multivariate exponentially weighted moving average controller

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090092337A1 (en) * 2007-09-07 2009-04-09 Takefumi Nagumo Image processing apparatus, image processing method, and computer program
CN102497497A (en) * 2011-12-05 2012-06-13 四川九洲电器集团有限责任公司 Method for dynamically adjusting threshold in image denoising algorithm
CN103067647A (en) * 2012-12-25 2013-04-24 四川九洲电器集团有限责任公司 Field programmable gata array (FPGA) based video de-noising method
CN103176425A (en) * 2013-03-13 2013-06-26 西北工业大学 Design method of multivariate exponentially weighted moving average controller

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105225244A (en) * 2015-10-22 2016-01-06 天津大学 Based on the noise detection method that minimum local mean square deviation calculates
CN110072034A (en) * 2018-01-23 2019-07-30 瑞昱半导体股份有限公司 Image treatment method and image processor
CN111191484A (en) * 2018-11-14 2020-05-22 普天信息技术有限公司 Method and device for recognizing human speaking in video image
CN113469893A (en) * 2020-05-08 2021-10-01 上海齐感电子信息科技有限公司 Method for estimating noise of image in video and video noise reduction method
CN113469893B (en) * 2020-05-08 2024-03-22 上海齐感电子信息科技有限公司 Noise estimation method of image in video and video noise reduction method
CN114070959A (en) * 2021-10-31 2022-02-18 南京理工大学 FPGA-based video denoising hardware implementation method
CN114070959B (en) * 2021-10-31 2024-04-12 南京理工大学 FPGA-based video denoising hardware implementation method

Similar Documents

Publication Publication Date Title
CN102497497B (en) Method for dynamically adjusting threshold in image denoising algorithm
CN107403413B (en) Video multi-frame denoising and enhancing method
CN106251318B (en) Denoising device and method for sequence image
CN101448077B (en) Self-adapting video image 3D denoise method
CN103581507A (en) Method dynamically adjusting threshold value through mean square error in de-noising algorithm
CN105574891B (en) The method and system of moving target in detection image
KR20060076176A (en) A method of temporal noise reduction in video sequence ans system therefore
CN106210448B (en) Video image jitter elimination processing method
CN101123681A (en) A digital image noise reduction method and device
Jin et al. Quaternion-based impulse noise removal from color video sequences
CN111614965B (en) Unmanned aerial vehicle video image stabilization method and system based on image grid optical flow filtering
CN102014281A (en) Methods and systems for motion estimation with nonlinear motion-field smoothing
CN102868879A (en) Method and system for converting video frame rate
CN102968765A (en) Method for correcting infrared focal plane heterogeneity based on sigma filter
Reeja et al. Real time video denoising
CN103996177A (en) Snow noise removing algorithm free of reference detection
JP3314043B2 (en) Motion detection circuit and noise reduction device
JP4161477B2 (en) Motion detection method and motion detection apparatus
WO2016185708A1 (en) Image processing device, image processing method, and storage medium
CN106027943B (en) A kind of video interlace-removing method
WO2016199418A1 (en) Frame rate conversion system
CN101494787B (en) De-block effect method based on block effect detection
CN101778297B (en) Interference elimination method of image sequence
KR20070098540A (en) Recursive super precision method for smoothly changing area
CN110705492A (en) Stage mobile robot obstacle target detection 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
ASS Succession or assignment of patent right

Free format text: FORMER OWNER: CHENGDU YUNYING SCIENCE + TECHNOLOGY CO., LTD.

Effective date: 20150108

C41 Transfer of patent application or patent right or utility model
COR Change of bibliographic data

Free format text: CORRECT: ADDRESS; FROM: 610000 CHENGDU, SICHUAN PROVINCE TO: 610041 CHENGDU, SICHUAN PROVINCE

TA01 Transfer of patent application right

Effective date of registration: 20150108

Address after: Wuhou District city of Sichuan province Chengdu 610041 triclosan Lane 8 No. 2 Building 1 unit 7 Building No. 14

Applicant after: Liu Jing

Address before: 610000 No. 28, Xin Cheng Road, hi tech Zone, Sichuan, Chengdu

Applicant before: Chengdu Yunying Science & Technology Co., Ltd.

RJ01 Rejection of invention patent application after publication

Application publication date: 20140212

RJ01 Rejection of invention patent application after publication