CN102810206A - Real-time loitering detection method based on dynamic programming - Google Patents
Real-time loitering detection method based on dynamic programming Download PDFInfo
- Publication number
- CN102810206A CN102810206A CN2011101485218A CN201110148521A CN102810206A CN 102810206 A CN102810206 A CN 102810206A CN 2011101485218 A CN2011101485218 A CN 2011101485218A CN 201110148521 A CN201110148521 A CN 201110148521A CN 102810206 A CN102810206 A CN 102810206A
- Authority
- CN
- China
- Prior art keywords
- pacing
- candidate
- object pixel
- track
- pixel
- 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
Images
Abstract
The invention discloses a real-time loitering detection method based on dynamic programming, which belongs to the field of real-time video detection. The method comprises: firstly, extracting the feature points of an image of the current frame; secondly, carrying out SSDA (sequential similarity detection algorithm) matching between the extracted feature points and an image of the previous frame to detect the corresponding matched feature points; then, carrying out feature point trajectory tracking of dynamic programming; evaluating each trajectory by an evaluation function; detecting the evaluation function value of each trajectory; determining that a loitering target occurs in the monitored area if the evaluation function value exceeds a set loitering threshold; and displaying the loitering trajectory and sending an alarm instruction. The method provided by the invention can effectively improve the loitering target recognition capability without needing a large amount of prior information, is easy to realize, is less influenced by the background of the monitored area, and has the advantages of stable and reliable performance and less calculation amount.
Description
Technical field
The present invention relates to the real-time video detection method, particularly a kind of detection method of pacing up and down in real time based on dynamic programming.
Background technology
Pacing up and down during video monitoring is paced up and down and detected is meant in video monitoring system have one or more targets in certain zone, to move back and forth.Pace up and down detect can assessment area in the threat level of target, this all plays an important role in many occasions.For example, in bank in front of the door, just do not allow the target of pacing up and down and the detection of at this moment just need pacing up and down occurs; Around dangerous facilities such as hi-line, also not allowing target paces up and down.
Chinese patent CN101770648A has proposed " a kind of pace up and down detection system and method based on video monitoring ".At first detect moving target in the regions in this patent, then target and the target that had occurred mated, confirm target be of short duration static after the target of motion, or the zoarium that bumps of several objects, or fresh target; According to different situations target is carried out different processing then, detect its movement locus, when course length surpasses setting threshold, just send alarm to the terminal.The moving object detection algorithm that the method relates to is the background subtraction point-score, needs storage lot of background view data.Background modeling exists not enough in the background subtraction point-score at aspects such as context update, background disturbance and shade inhibition.Handling the method that of short duration static problem with collision has adopted template matches, also need store a large amount of target datas.The algorithm structure that this patent proposed is complicated, needs a large amount of prior imformations, and real-time is relatively poor, is unfavorable for actual application.
Chinese patent CN101577006A has proposed " pace up and down detection system and method in the video monitoring ".The scheme that proposes in this patent is to carry out the secondary coupling according to target location and histogram, and whether time that occurs according to target or the number of times that appears judge the appearance of the target of pacing up and down greater than setting threshold.The method receives the background disturbing influence bigger, and the accuracy rate of object matching is not high, causes easily to mismatch.
Dynamic programming problems is actually a multistage decision optimization problem.To study a question during processing and be divided into several subproblems that connect each other.Dynamic programming is incorporated in the detection of pacing up and down, can simplifies problem effectively, reduce operand, improve detecting reliability.
Summary of the invention
The object of the present invention is to provide a kind of detection method of pacing up and down in real time based on dynamic programming; Thereby solve the detection algorithm of pacing up and down causes aspect such as TL track loss to exist in the of short duration static or collision of moving object detection, background interference, target deficiency, improve detection accuracy rate, reliability, real-time and the practicality of algorithm.
The technical solution that realizes the object of the invention is: a kind of detection method of pacing up and down in real time based on dynamic programming, and step is following:
(1) two field picture is carried out feature point extraction;
(2) write down the pace up and down positional information of object pixel of each candidate,, promptly might become the pixel of the target of pacing up and down as candidate's the object pixel of pacing up and down;
(3) in current frame image, with all candidates SSDA coupling that object pixel and previous frame image carry out the unique point guiding of pacing up and down;
(4) some or a plurality of candidates in object pixel and previous frame image object pixel of pacing up and down matees if the candidate paces up and down; And matched candidate object pixel and the candidate of the present frame object pixel of pacing up and down of pacing up and down does not overlap; Then it is included into the matched candidate track under the object pixel of pacing up and down, and upgrades the evaluation function of this track; The pace up and down object pixel coupling and overlapping of some candidates in object pixel and the previous frame image if the candidate paces up and down is not then upgraded the pace up and down evaluation function of the affiliated track of pixel of matched candidate; Any candidate object pixel of pacing up and down does not all match if the candidate paces up and down in object pixel and the previous frame image, then is directed against this candidate object pixel of pacing up and down and sets up track;
(5) set the track disappearance threshold, calculate the renewal speed of each bar track evaluation function, if renewal speed be 0 state continuance frame number greater than setting track disappearance threshold threshold value, then delete this track;
(6) set the threshold value of pacing up and down, judge that for each bar track its track evaluation function value is whether greater than the setting threshold value of pacing up and down.If, confirm that then this candidate object pixel of pacing up and down be the target of pacing up and down, show this candidate affiliated track of object pixel of pacing up and down; If not, then repeating step (1) to (6).
The present invention has following advantage than prior art: (1) adopts dynamic programming to follow the tracks of the back detection method earlier, detects the movement locus of all potential target points, has improved the detection probability to the target of pacing up and down effectively; (2) adopt method that the dynamic programming recurrence follows the tracks of and based on the SSDA coupling of unique point, accurate tracking target track, if the collision of a plurality of targets perhaps the of short duration static target signature locus of points that also can not cause of target lose; (2) need not to consider moving object detection, receive the influence of background disturbance less, algorithm performance is reliable and stable; (3) candidate's object pixel of pacing up and down in the main detected image is ignored other pixels, can significantly reduce algorithm operation quantity, improves the algorithm real-time; (4) this algorithm structure is simple, only needs present frame and former frame view data, can carry out the Feature Points Matching and the detection of pacing up and down, and need not a large amount of prior imformations, is easy to realize.
Below in conjunction with accompanying drawing the present invention is described in further detail.
Description of drawings
Fig. 1 is the detection method process flow diagram of pacing up and down in real time that the present invention is based on dynamic programming.
Fig. 2 is the SSDA matching process process flow diagram of pacing up and down in real time and detecting that the present invention is based on dynamic programming.
Embodiment
The present invention takes image characteristic point is carried out the thinking of dynamic programming track following; The guarded region image is not carried out target detection; Directly the object pixel of pacing up and down of candidate in the image is carried out track following; And every track estimated, then be considered to track greater than the pace up and down track of threshold value of setting, and then confirm the target of pacing up and down for the target of pacing up and down.
In conjunction with Fig. 1, the detection method of pacing up and down in real time that the present invention is based on dynamic programming may further comprise the steps:
Step 1: extract the unique point of current K two field picture, as candidate's object pixel of pacing up and down.Described unique point is strong angle point.Strong angle point is defined as X-Y scheme brightness and changes the point that has curvature maximum value on violent point or the pattern edge curve; Set feature point extraction threshold value
, establish Matrix C and be:
(1)
Wherein, f is the gray-scale value of input picture, f
xWith f
yBe respectively single order direction differential along x, y direction,
For standard deviation does
Gauss's smothing filtering function.
and
is the singular value of Matrix C, and
is prior preset threshold.Can find out the pace up and down coordinate (X of object pixel of candidate according to following extraction operator
T, Y
T):
All candidates that extract candidate that object pixel forms K two field picture object pixel collection
of pacing up and down of pacing up and down;
Step 2: pacing up and down for the candidate, each character pixel carries out the SSDA coupling in the object pixel collection
.In conjunction with Fig. 2, step is following:
(a) to image f
KIn each candidate object pixel of pacing up and down, selecting radius is the neighborhood window of r, forms (2r+1) * (2r+1) template image;
(b) definition absolute value error:
Wherein: S is a benchmark image; T is a template image;
is the subimage in the benchmark image under the template image covering;
is the pixel coordinate of current calculating; (i; J) be
top left corner pixel coordinate on benchmark image; M * N is the template image size;
covers the pixel average of benchmark image down for template image, and
is the pixel average of template image:
(c) set detection threshold S
h;
(d) utilize candidate's object pixel collection of pacing up and down
In each candidate template image that object pixel forms of pacing up and down, successively at image f
K-1Last traversal.The absolute value error that the calculation template corresponding point are right
, with somewhat right error accumulation, when the deviation accumulation sum that adds up after R time greater than the detection threshold S that sets
h, then stop to add up, and record accumulative frequency R; SSDA is at f in definition
K-1Last detection curved surface is:
(6)
(E)? Take the point of maximum R-value target pixel set as a candidate hovering
candidate matches the corresponding pixel Pixel
;
(f) candidate matches is put the character pixel that candidate in pixel and the K-1 two field picture paces up and down in the object pixel collection
and carry out the secondary coupling; Calculated candidate pace up and down each candidate in the object pixel collection
pace up and down object pixel and candidate matches point pixel
apart from d, selected distance less than radius of neighbourhood r and for the candidate of the minor increment object pixel of pacing up and down be the matched candidate object pixel of pacing up and down.Can in the K-1 two field picture, confirm like this with the K two field picture in candidate's candidate that object pixel is complementary object pixel of pacing up and down of pacing up and down;
Step 3:, carry out dynamic programming track accumulative total to candidate's each candidate object pixel of pacing up and down of pacing up and down in the object pixel collection
:
If at f
K-1In have 0 matching candidate object pixel of pacing up and down, think that then this unique point is the new motion feature point that gets into guarded region, creates new track sequence:
The evaluation function of track is:
If f
K-1In have one or more and the matching candidate that it does not the overlap object pixel of pacing up and down; Think that then this candidate object pixel of pacing up and down is a pace up and down object pixel and be in motion state of already present candidate in the guarded region, it be included into matching candidate pace up and down in the track sequence under the object pixel:
Wherein,
expression track sequence is created and which frame; And renewal track evaluation function:
If f
K-1In have the matching candidate overlap with it object pixel of pacing up and down, think that then this candidate object pixel of pacing up and down is a pace up and down object pixel and be in stationary state of already present candidate in the guarded region, do not upgrade the evaluation function of the affiliated track of this matching characteristic pixel;
Step 4: set track disappearance threshold V
hCalculate the renewal speed of the evaluation function of every group of track sequence:
If being 0 lasting frame number, renewal speed surpasses prior preset threshold V
h, think that then this track sequence disappears, and deletes this track sequence;
Step 5: set the threshold value P that paces up and down
hDetect the evaluation function of each track sequence, if evaluation function is less than the threshold value P that paces up and down that sets in advance
h, then repeating step 1 to 5; If evaluation function surpasses the threshold value P that paces up and down that sets
h, then thinking has target to pace up and down at guarded region, shows the pace up and down target and the target trajectory of pacing up and down, and instruction gives the alarm.
Claims (6)
1. detection method of pacing up and down in real time based on dynamic programming is characterized in that step is following:
To carrying out feature point extraction when last two field picture;
Write down the pace up and down positional information of object pixel of each candidate; As candidate's object pixel of pacing up and down, promptly might become the pixel of the target of pacing up and down;
In current frame image, with all candidates SSDA coupling that object pixel and previous frame image carry out the unique point guiding of pacing up and down;
Judge: the object pixel of pacing up and down of the some or a plurality of candidates if the candidate paces up and down in object pixel and the previous frame image matees; And matched candidate object pixel and the candidate of the present frame object pixel of pacing up and down of pacing up and down does not overlap; Then it is included into the matched candidate track under the object pixel of pacing up and down, and upgrades the evaluation function of this track; The pace up and down object pixel coupling and overlapping of some candidates in object pixel and the previous frame image if the candidate paces up and down is not then upgraded the pace up and down evaluation function of the affiliated track of object pixel of matched candidate; Any candidate object pixel of pacing up and down does not all match if the candidate paces up and down in object pixel and the previous frame image, then is directed against this candidate object pixel of pacing up and down and sets up track;
Set the track disappearance threshold, calculate the renewal speed of each track evaluation function, if renewal speed be 0 state continuance frame number greater than setting the track disappearance threshold, then delete this track;
The setting threshold value of pacing up and down judges that for each bar track its track evaluation function value is whether greater than the setting threshold value of pacing up and down; If, confirm that then this candidate object pixel of pacing up and down be the target of pacing up and down, show this candidate affiliated track of object pixel of pacing up and down, send alarm command; If not, then repeating step (1) to (6).
2. the detection method of pacing up and down in real time based on dynamic programming according to claim 1 is characterized in that: realize through following method carrying out feature point extraction when last two field picture in the said step (1):
Extract the unique point of current K two field picture, described unique point is strong angle point, and this strong angle point changes the point that has curvature maximum value on violent point or the pattern edge curve for X-Y scheme brightness; Set feature point extraction threshold value
, establish Matrix C and be:
Wherein, f is the gray-scale value of input picture, f
xWith f
yBe respectively single order direction differential along x, y direction,
For standard deviation does
Gauss's smothing filtering function,
With
Be the singular value of Matrix C,
Be prior preset threshold, can find out the pace up and down coordinate (X of object pixel of candidate according to following extraction operator
T, Y
T):
3. the detection method of pacing up and down in real time based on dynamic programming according to claim 1 is characterized in that: in the said step (3) with all candidates step that sequence similarity that object pixel and previous frame image levy a guiding detects the SSDA coupling of pacing up and down be:
(a) to K two field picture f
KIn each candidate object pixel of pacing up and down, selecting radius is the neighborhood window of r, forms (2r+1) * (2r+1) template image;
(b) definition absolute value error:
Wherein: S is a benchmark image; T is a template image;
is the subimage in the benchmark image under the template image covering;
is the pixel coordinate of current calculating; (i; J) be
top left corner pixel coordinate on benchmark image; M * N is the template image size;
covers the pixel average of benchmark image down for template image, and
is the pixel average of template image:
(c) set detection threshold S
h;
(d) utilize candidate's object pixel collection of pacing up and down
In the template image that forms of each character pixel, successively at image f
K-1Last traversal; The absolute value error that the calculation template corresponding point are right
, with somewhat right error accumulation, when the deviation accumulation sum that adds up after R time greater than the detection threshold S that sets
h, then stop to add up, and record accumulative frequency R; SSDA is at f in definition
K-1Last detection curved surface is:
(E)? Take maximum points R-value is a candidate set of the target pixel hovering
candidate matches the corresponding pixel Pixel
;
(f) candidate matches is put pixel and K-1 two field picture f
K-1In candidate's object pixel collection of pacing up and down
In character pixel carry out secondary coupling, the calculated candidate object pixel collection of pacing up and down
In each character pixel and candidate matches point pixel
Apart from d, selected distance less than radius of neighbourhood r and for the candidate of the minor increment object pixel of pacing up and down be the matching candidate object pixel of pacing up and down; Can in the K-1 two field picture, confirm with the K two field picture in the candidate pace up and down that object pixel is complementary candidate's object pixel of pacing up and down;
4. the detection method of pacing up and down in real time based on dynamic programming according to claim 1 is characterized in that: judge in the said step (4) that the matched feature points step is following:
To candidate's each candidate object pixel of pacing up and down of pacing up and down in the object pixel collection
, carry out dynamic programming track accumulative total:
If at image f
K-1In have 0 matching candidate object pixel of pacing up and down, think that then this unique point is the new motion feature point that gets into guarded region, creates new track sequence:
The evaluation function of track is:
If image f
K-1In have one or more and the matching candidate that it does not the overlap object pixel of pacing up and down; Think that then this candidate object pixel of pacing up and down is a pace up and down object pixel and be in motion state of already present candidate in the guarded region, it be included into matching candidate pace up and down in the track sequence under the object pixel:
Wherein,
expression track sequence is created and which frame; And renewal track evaluation function:
If f
K-1In have the matching candidate overlap with it object pixel of pacing up and down, think that then this candidate object pixel of pacing up and down is a pace up and down object pixel and be in stationary state of already present candidate in the guarded region, do not upgrade the evaluation function of the affiliated track of this matching characteristic pixel;
5. the detection method of pacing up and down in real time based on dynamic programming according to claim 1 is characterized in that: judge in the said step (5) that track disappears through following method realization:
Set track disappearance threshold V
hCalculate the renewal speed of the evaluation function of every group of track sequence:
If being 0 lasting frame number, renewal speed surpasses prior preset threshold V
h, think that then this track sequence disappears, and deletes this track sequence.
6. the detection method of pacing up and down in real time based on dynamic programming according to claim 1 is characterized in that: step (6) judgement has or not the target of pacing up and down to realize through following method:
The setting threshold value P that paces up and down
hDetect the evaluation function of each track sequence, if evaluation function is less than the threshold value P that paces up and down that sets in advance
h, then repeating step 1 to 5; If evaluation function surpasses the threshold value P that paces up and down that sets
h, then thinking has target to pace up and down at guarded region, shows the target trajectory of pacing up and down, and instruction gives the alarm.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2011101485218A CN102810206A (en) | 2011-06-03 | 2011-06-03 | Real-time loitering detection method based on dynamic programming |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2011101485218A CN102810206A (en) | 2011-06-03 | 2011-06-03 | Real-time loitering detection method based on dynamic programming |
Publications (1)
Publication Number | Publication Date |
---|---|
CN102810206A true CN102810206A (en) | 2012-12-05 |
Family
ID=47233908
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2011101485218A Pending CN102810206A (en) | 2011-06-03 | 2011-06-03 | Real-time loitering detection method based on dynamic programming |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102810206A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105374031A (en) * | 2015-10-14 | 2016-03-02 | 江苏美的清洁电器股份有限公司 | Household security protection data processing method and system based on robot |
WO2016141641A1 (en) * | 2015-03-11 | 2016-09-15 | 中兴通讯股份有限公司 | Method and device for recognizing suspicious person |
CN106971142A (en) * | 2017-02-07 | 2017-07-21 | 深圳云天励飞技术有限公司 | A kind of image processing method and device |
CN110062210A (en) * | 2019-04-28 | 2019-07-26 | 广东安居宝数码科技股份有限公司 | Monitoring method, device, equipment and the storage medium of object |
CN112633150A (en) * | 2020-12-22 | 2021-04-09 | 中国华戎科技集团有限公司 | Target trajectory analysis-based retention loitering behavior identification method and system |
CN113050643A (en) * | 2021-03-19 | 2021-06-29 | 京东鲲鹏(江苏)科技有限公司 | Unmanned vehicle path planning method and device, electronic equipment and computer readable medium |
CN113255534A (en) * | 2021-05-28 | 2021-08-13 | 河北幸福消费金融股份有限公司 | Early warning method, system, device and storage medium based on video image analysis |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101577006A (en) * | 2009-06-15 | 2009-11-11 | 北京中星微电子有限公司 | Loitering detecting method and loitering detecting system in video monitoring |
CN101770648A (en) * | 2009-01-06 | 2010-07-07 | 北京中星微电子有限公司 | Video monitoring based loitering system and method thereof |
CN101777183A (en) * | 2009-01-13 | 2010-07-14 | 北京中星微电子有限公司 | Method and device for detecting stationary object and method and device for detecting abandoned object |
CN101783020A (en) * | 2010-03-04 | 2010-07-21 | 湖南大学 | Video multi-target fast tracking method based on joint probability data association |
-
2011
- 2011-06-03 CN CN2011101485218A patent/CN102810206A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101770648A (en) * | 2009-01-06 | 2010-07-07 | 北京中星微电子有限公司 | Video monitoring based loitering system and method thereof |
CN101777183A (en) * | 2009-01-13 | 2010-07-14 | 北京中星微电子有限公司 | Method and device for detecting stationary object and method and device for detecting abandoned object |
CN101577006A (en) * | 2009-06-15 | 2009-11-11 | 北京中星微电子有限公司 | Loitering detecting method and loitering detecting system in video monitoring |
CN101783020A (en) * | 2010-03-04 | 2010-07-21 | 湖南大学 | Video multi-target fast tracking method based on joint probability data association |
Non-Patent Citations (1)
Title |
---|
陈浩: "图像制导系统图像匹配算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016141641A1 (en) * | 2015-03-11 | 2016-09-15 | 中兴通讯股份有限公司 | Method and device for recognizing suspicious person |
CN105374031A (en) * | 2015-10-14 | 2016-03-02 | 江苏美的清洁电器股份有限公司 | Household security protection data processing method and system based on robot |
CN106971142A (en) * | 2017-02-07 | 2017-07-21 | 深圳云天励飞技术有限公司 | A kind of image processing method and device |
CN110062210A (en) * | 2019-04-28 | 2019-07-26 | 广东安居宝数码科技股份有限公司 | Monitoring method, device, equipment and the storage medium of object |
CN112633150A (en) * | 2020-12-22 | 2021-04-09 | 中国华戎科技集团有限公司 | Target trajectory analysis-based retention loitering behavior identification method and system |
CN113050643A (en) * | 2021-03-19 | 2021-06-29 | 京东鲲鹏(江苏)科技有限公司 | Unmanned vehicle path planning method and device, electronic equipment and computer readable medium |
CN113255534A (en) * | 2021-05-28 | 2021-08-13 | 河北幸福消费金融股份有限公司 | Early warning method, system, device and storage medium based on video image analysis |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102810206A (en) | Real-time loitering detection method based on dynamic programming | |
US10970859B2 (en) | Monitoring method and device for mobile target, monitoring system and mobile robot | |
Yang et al. | Real-time multiple objects tracking with occlusion handling in dynamic scenes | |
CN101794385B (en) | Multi-angle multi-target fast human face tracking method used in video sequence | |
CN109657575B (en) | Intelligent video tracking algorithm for outdoor constructors | |
Jafari et al. | Real-time RGB-D based people detection and tracking for mobile robots and head-worn cameras | |
CN105023278A (en) | Movable target tracking method and system based on optical flow approach | |
US9582711B2 (en) | Robot cleaner, apparatus and method for recognizing gesture | |
CN101799968B (en) | Detection method and device for oil well intrusion based on video image intelligent analysis | |
Bondi et al. | Real-time people counting from depth imagery of crowded environments | |
CN104008371A (en) | Regional suspicious target tracking and recognizing method based on multiple cameras | |
CN114119676B (en) | Target detection tracking identification method and system based on multi-feature information fusion | |
CN104091348A (en) | Multi-target tracking method integrating obvious characteristics and block division templates | |
CN105405154A (en) | Target object tracking method based on color-structure characteristics | |
CN103927764B (en) | A kind of wireless vehicle tracking of combining target information and estimation | |
CN104424638A (en) | Target tracking method based on shielding situation | |
CN105654508A (en) | Monitoring video moving target tracking method based on self-adaptive background segmentation and system thereof | |
CN111209781B (en) | Method and device for counting indoor people | |
US20200250803A1 (en) | Method for detecting and tracking target object, target object tracking apparatus, and computer-program product | |
CN104063885A (en) | Improved movement target detecting and tracking method | |
CN103729861A (en) | Multiple object tracking method | |
CN104281837A (en) | Pedestrian tracking method combining Kalman filtering with ROI expansion between adjacent frames | |
Pore et al. | Bidirectional people counting system in video surveillance | |
Chen et al. | Research on the detection and tracking algorithm of moving object in image based on computer vision technology | |
Ryoo et al. | Observe-and-explain: A new approach for multiple hypotheses tracking of humans and objects |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20121205 |