CN101950366A - Method for detecting and identifying station logo - Google Patents
Method for detecting and identifying station logo Download PDFInfo
- Publication number
- CN101950366A CN101950366A CN 201010279070 CN201010279070A CN101950366A CN 101950366 A CN101950366 A CN 101950366A CN 201010279070 CN201010279070 CN 201010279070 CN 201010279070 A CN201010279070 A CN 201010279070A CN 101950366 A CN101950366 A CN 101950366A
- Authority
- CN
- China
- Prior art keywords
- station symbol
- identification
- frame
- detection
- 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
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a method for detecting and identifying a station logo. The method comprises the following steps of: 1) setting a station logo example video frame, performing edge detection and filtering on the station logo example video frame, further positioning and segmenting a station logo area, and partitioning the station logo area to obtain a station logo template; 2) setting a video frame to be detected and identified, performing edge detection on the video frame to obtain an edge image of the video frame to be detected and identified; and 3) acquiring a station logo detection and identification result by a sliding window searching and block matching method according to the station logo template and the edge image of the video frame to be detected and identified. Compared with the conventional global matching-based method, the method of the invention can achieve higher station logo detection and identification accuracy rate and time efficiency.
Description
Technical field
The present invention relates to the area of pattern recognition of image, particularly a kind of method of station symbol detection and Identification.
Background technology
TV station's station symbol is the sign of a TV station, has comprised important semantic informations such as this TV station's platform name, program orientation, is one of important meaning of one's words source of realizing video analysis, understanding and retrieval.Along with the develop rapidly of TV tech, the program of TV stations at different levels has reached covers up to a hundred.By manually TV signal being monitored in real time or being selected specific television program to include etc., working strength is big, inefficiency, and mistakes are hard to avoid.Station symbol detection and Identification technology is the key that overcomes above problem, has crucial research and using value.
Template matches is the algorithm of station symbol detection and Identification the most intuitively.2007, the template matching method based on station symbol zone interior pixel color value introduced in the article that Huang surmounts, Lin Jinyao delivers on " Modern Television Technology " " principle of station symbol detection system and application ".This method can reach matching effect preferably for opaque station symbol.But, owing to translucent station symbol zone interior pixel color along with background color changes, this method is subjected to the influence of change of background easily for translucent station symbol detection and Identification.2008, the national inventing patent " a kind of TV station logo training method and recognition methods " of people such as Wang Wenying application (application number: introduced the method for carrying out the station symbol detection and Identification 200810226266.2) by Edge Distance transformation matrix coupling, can adapt to the detection and Identification of translucent station symbol better, less because the edge of station symbol is subjected to the influence of change of background.But, have the mutually approximate situation of station symbol of same TV station different channel in the reality in a large number, as CCTV1 and CCTV10.The method of overall situation edge coupling is difficult to effectively distinguish approximate station symbol, because identical station symbol similarity under noise effect changes in a scope, and approximate station symbol often has higher overall similarity, obscures with identical station symbol easily.In addition, the method for overall edge coupling is proving under the local unmatched situation that still need to finish the coupling of all marginal points in the moving window, time efficiency is lower.
Summary of the invention
The technical problem to be solved in the present invention provides a kind of method of TV station logo detection and Identification, overcomes the deficiency based on station symbol overall situation content match, improves the accuracy rate and the time efficiency of station symbol detection and Identification.
In order to solve the problems of the technologies described above, the invention provides a kind of method of station symbol detection and Identification, may further comprise the steps:
1) given station symbol example video frame carries out rim detection and filtration to it, and further the location is partitioned into the station symbol zone, to the station symbol area dividing, obtains the station symbol template;
2) given to be detected and identification frame of video, it is carried out rim detection, obtain to be detected and identification the frame of video outline map;
3) according to station symbol template and frame of video outline map to be detected and identification, adopt the method for moving window search and piecemeal coupling, obtain the result of station symbol detection and Identification.
Described step 1) realizes the station symbol edge filter by the non-station symbol marginal point among the manual method removal station symbol example video frame border figure.
The station symbol zone of locating in the described step 1) is for comprising the minimum rectangle frame zone of station symbol marginal point.
Described step 1) becomes a plurality of subregions with the station symbol Region Segmentation adaptively, and the sum of station symbol marginal point approximates a given numerical value in each subregion, and the union of all subregions can cover all station symbol marginal points.
The station symbol template comprises described in the described step 1): position in frame of video of station symbol outline map, station symbol zone, station symbol area dividing be three elements as a result.
Described station symbol outline map is the station symbol edge in the station symbol zone.
The scope of moving window search is by the position of station symbol zone in frame of video of determining in the described step 1) in the described step 3), and the maximum fluctuation distance of given station symbol zone on level, vertical both direction is definite, and the hunting zone does not exceed video frame boundary.
In the described step 3), for arbitrary moving window, the method that adopts moving window search and piecemeal coupling is carried out piecemeal to the frame of video edge of wherein to be detected and identification and station symbol edge and is mated, if the minimum value of all segmented areas local similar degree is greater than given threshold value in this moving window, comprise given station symbol in the frame of video of thinking to be detected so and discerning, withdraw from coupling; If matching similarity is less than given threshold value in arbitrary segmented areas, skip the coupling of residue segmented areas in this moving window so, begin the piecemeal coupling in next moving window; If all moving windows traversal finishes, and do not have in the moving window local similar degree minimum value, think so not comprise given station symbol in the frame of video of to be detected and identification greater than given threshold value.
Effect of the present invention is: compare with traditional method based on global registration, the inventive method can obtain the accuracy rate and the time efficiency of higher station symbol detection and Identification.Its reason is: the piecemeal coupling has more effectively been distinguished the difference of similar station symbol regional area, can improve the accuracy rate of station symbol detection and Identification; Piecemeal coupling has made full use of that the result of matching area filters out a large amount of unnecessary zones coupling, can improve the time efficiency of station symbol detection and Identification.
Description of drawings
Fig. 1 is the process flow diagram of the inventive method.
Fig. 2 is the outline map of 6 kinds of TV station logos of employing in the experiment.
Embodiment
In order to make purpose of the present invention, technical scheme and advantage clearer, below in conjunction with accompanying drawing, the present invention is further detailed explanation according to embodiment.
In the present embodiment, at first obtain the station symbol template, extract frame of video edge to be detected and identification then, obtain the result of station symbol detection and Identification at last by moving window piecemeal coupling according to given station symbol example video frame.May further comprise the steps:
Step 1, given station symbol example video frame carries out rim detection and filtration to it, and further the location is partitioned into the station symbol zone, to the station symbol area dividing, obtains the station symbol template.
In the present embodiment, rim detection adopts the Canny operator to realize.Edge filter is realized by non-station symbol marginal point in the outline map of manual method removal station symbol example video frame.Because in the station symbol detection and Identification, the modeling of each station symbol only need be carried out once, manual method is filtered non-station symbol marginal point can't cause too much labor workload.Accurate station symbol outline map plays an important role to the effect of station symbol detection and Identification, and artificial method obtains more accurate station symbol outline map easily, improves the effect of station symbol detection and Identification.
The station symbol zone location obtains according to the station symbol edge after filtering.The station symbol zone of location is for comprising the minimum rectangle frame zone of station symbol marginal point.The left and right border of rectangular area is corresponding station symbol marginal point coordinate minimum value, maximal value in the horizontal direction respectively, and the upper and lower border of rectangular area is minimum value, the maximal value of corresponding station symbol marginal point coordinate in the vertical direction respectively.
The station symbol area dividing is realized by adaptively the station symbol Region Segmentation being become a plurality of subregions.The sum of station symbol marginal point approximates a given numerical value in each subregion, and the union of all subregions can cover all station symbol marginal points.The approximate value N of station symbol marginal point sum from left to right becomes the station symbol Region Segmentation a plurality of rectangular sub-regions territory in given each zone.The left margin in first rectangular sub-regions territory is the left margin in station symbol zone, and right margin moves right gradually, up to wherein station symbol marginal point sum just greater than N; The left margin in second rectangular sub-regions territory is the right margin in first rectangular sub-regions territory, and right margin moves right gradually, up to wherein station symbol marginal point sum just greater than N; Cut apart residue station symbol zone by that analogy, the station symbol number of edge points is less than N in residue station symbol zone; If the station symbol number of edge points is less than N in the residue station symbol zone, and is not equal to 0, increase a rectangular sub-regions territory so, its right margin is the right margin in station symbol zone, left margin is moved to the left gradually, up to wherein station symbol marginal point sum just greater than N.In the present embodiment, N is set at 100.
Step 2, given to be detected and identification frame of video, it is carried out rim detection, obtain to be detected and identification the frame of video outline map.
In the present embodiment, the edge detection method that is adopted in the step 2 is identical with the edge detection method described in the step 1.
Step 3 according to station symbol template and frame of video outline map to be detected and identification, adopts the method for moving window search and piecemeal coupling, obtains the result of station symbol detection and Identification.
The scope of in moving window, carrying out piecemeal coupling by the station symbol zone in frame of video the position and given station symbol regional location in level, vertically the maximum fluctuation distance on the both direction is determined, and matching range does not exceed video frame boundary.In the present embodiment, the position of station symbol zone in frame of video obtained by station symbol zone location in the described step 1, and the maximum fluctuation distance of station symbol regional location on level, vertical both direction all is set at 10.
For arbitrary moving window, to the frame of video edge and the station symbol edge of wherein to be detected and identification carrying out the piecemeal coupling.If in this moving window the minimum value of all segmented areas local similar degree greater than given threshold value (described threshold value automatically according to etc. wrong rate selected), think so to comprise given station symbol in the frame of video of to be detected and identification, withdraw from coupling; If matching similarity is less than given threshold value in arbitrary segmented areas, skip the coupling of residue segmented areas in this moving window so, begin the piecemeal coupling in next moving window; If all moving windows traversal finishes, and do not have in the moving window local similar degree minimum value, think so not comprise given station symbol in the frame of video of to be detected and identification greater than given threshold value.In the present embodiment, segmented areas calculation of similarity degree method is shown in formula one, two.
Formula one:
Formula two: nUni=|T ∩ W|+ α * nNeighbor
Wherein, S is the segmented areas similarity; | T| is a station symbol number of edge points in the segmented areas; | W| is the frame of video number of edge points of to be detected and identification in the segmented areas; | T ∩ W| is the number of edge points that the frame of video edge of the interior station symbol edge of segmented areas and to be detected and identification occurs simultaneously; NNeighbor is the frame of video number of edge points of to be detected and identification in the neighbours territory, station symbol edge in the segmented areas.
Following experimental result shows that compare with traditional method based on global registration, the present invention can obtain the accuracy rate and the time efficiency of higher station symbol detection and Identification.
The database of setting up in the present embodiment comprises 6 kinds of different station symbols, every kind of 100 frame of video to be detected and discern that the TV station logo correspondence is sampled from the associated video intermediate reach, totally 600.These 6 kinds of station symbols have embodied the approximate mutually situation of station symbol of same TV station different channel from the different channel of 3 TV stations.The background that comprise this station symbol more single frame of video of the example video frame of every kind of TV station logo for selecting in advance.Fig. 2 has showed the outline map of 6 kinds of TV station logos.
As a sub regions, other steps are constant with the station symbol regional integration, so, just can obtain the method for corresponding station symbol detection and Identification based on global registration.
In the experiment, wrong rates such as employing are estimated the accuracy rate of station symbol detection and Identification; Adopt detection and Identification one frame to treat to estimate the averaging time of frame of video the time efficiency of station symbol detection and Identification; Etc. wrong rate and averaging time all is low more good more.Etc. wrong rate is mistake (acceptance/refusal) rate of false acceptance rate when equaling false rejection rate.In the present embodiment, false acceptance rate is defined as the frame of video that will not comprise given station symbol and is judged as the probability that contains given station symbol; False rejection rate is defined as the frame of video that will comprise given station symbol and is judged as the probability that does not contain given station symbol.Increase matching threshold, false acceptance rate reduces, but false rejection rate increases; Reduce matching threshold, false rejection rate reduces, but false acceptance rate increases.Etc. wrong rate is the equilibrium point of false acceptance rate and false rejection rate, effectively the accuracy rate of reactive system.
Table 1: experimental result contrast
| Etc. wrong rate | Averaging time (/ frame) | |
| Piecemeal coupling (embodiment of the invention) | 4.69％ | 18.23ms |
| Global registration (classic method) | 26.72％ | 52.46ms |
As can be seen from Table 1, compare with traditional method based on global registration, the present invention can obtain the accuracy rate and the time efficiency of higher station symbol detection and Identification.Its reason is: identical station symbol similarity under noise effect changes in a scope, and approximate station symbol often has higher overall similarity, obscures with identical station symbol easily; The piecemeal coupling has more effectively been distinguished the difference of similar station symbol regional area, can improve the accuracy rate of station symbol detection and Identification; During the piecemeal coupling, matching similarity is less than given threshold value in arbitrary segmented areas, just skip the coupling of residue segmented areas in this moving window, begin the piecemeal coupling in next moving window, made full use of that the result of matching area filters out a large amount of unnecessary zones coupling, can improve the time efficiency of station symbol detection and Identification.
Should illustrate at last: above embodiment is the unrestricted technical scheme of the present invention in order to explanation only.Those of ordinary skill in the art can carry out various changes and modification to the present invention and not break away from the spirit and scope of the present invention.Like this, if of the present invention these are revised and modification belongs within the scope of claim of the present invention and equivalent technologies thereof, then the present invention also is intended to comprise these changes and modification interior.
Claims (8)
1. the method for station symbol detection and Identification may further comprise the steps:
1) given station symbol example video frame carries out rim detection and filtration to it, and further the location is partitioned into the station symbol zone, to the station symbol area dividing, obtains the station symbol template;
2) given to be detected and identification frame of video, it is carried out rim detection, obtain to be detected and identification the frame of video outline map;
3) according to station symbol template and frame of video outline map to be detected and identification, adopt the method for moving window search and piecemeal coupling, obtain the result of station symbol detection and Identification.
2. the method for a kind of station symbol detection and Identification as claimed in claim 1 is characterized in that, described step 1) realizes the station symbol edge filter by the non-station symbol marginal point among the manual method removal station symbol example video frame border figure.
3. the method for a kind of station symbol detection and Identification as claimed in claim 1 is characterized in that, the station symbol zone that the location is partitioned in the described step 1) is for comprising the minimum rectangle frame zone of station symbol marginal point.
4. the method for a kind of station symbol detection and Identification as claimed in claim 1, it is characterized in that, described step 1) becomes a plurality of subregions with the station symbol Region Segmentation adaptively, the sum of station symbol marginal point approximates a given numerical value in each subregion, and the union of all subregions can cover all station symbol marginal points.
5. the method for a kind of station symbol detection and Identification as claimed in claim 1 is characterized in that, the station symbol template comprises described in the described step 1): position in frame of video of station symbol outline map, station symbol zone, station symbol area dividing be three elements as a result.
6. the method for a kind of station symbol detection and Identification as claimed in claim 5 is characterized in that, described station symbol outline map is the station symbol edge in the station symbol zone.
7. the method for a kind of station symbol detection and Identification as claimed in claim 1, it is characterized in that, the scope of moving window search is by the position of station symbol zone in frame of video in the described step 3), and the maximum fluctuation distance of given station symbol zone on level, vertical both direction is definite, and the hunting zone does not exceed video frame boundary.
8. the method for a kind of station symbol detection and Identification as claimed in claim 1 is characterized in that, in the described step 3), adopts the method for moving window search and piecemeal coupling may further comprise the steps:
For arbitrary moving window, if the minimum value of the local matching similarities of all segmented areas is greater than given threshold value in this moving window, judge so in the frame of video of to be detected and identification to comprise given station symbol, withdraw from coupling;
If matching similarity is less than given threshold value in arbitrary segmented areas, skip the coupling of residue segmented areas in this moving window so, begin the piecemeal coupling in next moving window;
If all moving windows traversal finishes, and do not have the local matching similarity minimum value of segmented areas in the moving window, judge so in the frame of video of to be detected and identification not comprise given station symbol greater than given threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010279070 CN101950366A (en) | 2010-09-10 | 2010-09-10 | Method for detecting and identifying station logo |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010279070 CN101950366A (en) | 2010-09-10 | 2010-09-10 | Method for detecting and identifying station logo |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101950366A true CN101950366A (en) | 2011-01-19 |
Family
ID=43453861
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201010279070 Pending CN101950366A (en) | 2010-09-10 | 2010-09-10 | Method for detecting and identifying station logo |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101950366A (en) |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102281384A (en) * | 2011-07-11 | 2011-12-14 | 成都索贝数码科技股份有限公司 | Method and device for obscuring abandoned image regions, and method and system for processing video |
| CN102306279A (en) * | 2011-07-12 | 2012-01-04 | 央视国际网络有限公司 | Method for identifying video scores and device |
| CN102306280A (en) * | 2011-07-12 | 2012-01-04 | 央视国际网络有限公司 | Method and device for detecting video scores |
| CN102426647A (en) * | 2011-10-28 | 2012-04-25 | Tcl集团股份有限公司 | Station identification method and device |
| CN102867170A (en) * | 2012-08-15 | 2013-01-09 | 上海国茂数字技术有限公司 | Method and device for automatically detecting label inserted into video |
| CN103002328A (en) * | 2012-11-30 | 2013-03-27 | 北京十分科技有限公司 | Method and device for identifying channels |
| CN103258187A (en) * | 2013-04-16 | 2013-08-21 | 华中科技大学 | Television station caption identification method based on HOG characteristics |
| CN103353939A (en) * | 2013-06-19 | 2013-10-16 | 电子科技大学 | Method for extracting and identifying image characteristic object template under dynamic background |
| CN103413149A (en) * | 2013-09-06 | 2013-11-27 | 上海美琦浦悦通讯科技有限公司 | Method for detecting and identifying static target in complicated background |
| CN103530655A (en) * | 2013-05-24 | 2014-01-22 | Tcl集团股份有限公司 | Station logo identification method and system |
| CN103544467A (en) * | 2013-04-23 | 2014-01-29 | Tcl集团股份有限公司 | Method and device for detecting and recognizing station captions |
| CN103618905A (en) * | 2013-12-09 | 2014-03-05 | 清华大学深圳研究生院 | Content drawing method for station caption area in video |
| CN104639952A (en) * | 2015-01-23 | 2015-05-20 | 小米科技有限责任公司 | Method and device for identifying station logo |
| CN104918107A (en) * | 2015-05-29 | 2015-09-16 | 小米科技有限责任公司 | Video file identification processing method and device |
| CN104954854A (en) * | 2014-03-27 | 2015-09-30 | 联想(北京)有限公司 | Station logo identification method and electronic equipment |
| CN105760842A (en) * | 2016-02-26 | 2016-07-13 | 北京大学 | Station caption identification method based on combination of edge and texture features |
| CN105869123A (en) * | 2015-11-24 | 2016-08-17 | 乐视致新电子科技（天津）有限公司 | Image processing method and apparatus |
| CN105872591A (en) * | 2016-04-22 | 2016-08-17 | 北京小米移动软件有限公司 | Icon covering method and device |
| CN106203237A (en) * | 2015-05-04 | 2016-12-07 | 杭州海康威视数字技术股份有限公司 | The recognition methods of container-trailer numbering and device |
| CN106454411A (en) * | 2016-10-27 | 2017-02-22 | 北京小米移动软件有限公司 | Station caption processing method and device |
| CN103714314B (en) * | 2013-12-06 | 2017-04-19 | 安徽大学 | Television video station caption identification method combining edge and color information |
| WO2017088463A1 (en) * | 2015-11-24 | 2017-06-01 | 乐视控股（北京）有限公司 | Image processing method and device |
| WO2017088478A1 (en) * | 2015-11-24 | 2017-06-01 | 乐视控股（北京）有限公司 | Number separating method and device |
| CN107943849A (en) * | 2017-11-03 | 2018-04-20 | 小草数语（北京）科技有限公司 | The search method and device of video file |
| CN108280834A (en) * | 2018-01-22 | 2018-07-13 | 西安万像电子科技有限公司 | Video area determines method and device |
| CN109783658A (en) * | 2019-02-19 | 2019-05-21 | 苏州科达科技股份有限公司 | Image processing method, device and storage medium |
| CN111368703A (en) * | 2020-02-29 | 2020-07-03 | 上海电力大学 | Platform logo detection and identification method based on FPN |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005057322A2 (en) * | 2003-11-19 | 2005-06-23 | Nielsen Media Research, Inc. | Methods and apparatus for detecting a television channel change event |
| CN101739561A (en) * | 2008-11-11 | 2010-06-16 | 中国科学院计算技术研究所 | TV station logo training method and identification method |
-
2010
- 2010-09-10 CN CN 201010279070 patent/CN101950366A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005057322A2 (en) * | 2003-11-19 | 2005-06-23 | Nielsen Media Research, Inc. | Methods and apparatus for detecting a television channel change event |
| CN101739561A (en) * | 2008-11-11 | 2010-06-16 | 中国科学院计算技术研究所 | TV station logo training method and identification method |
Non-Patent Citations (2)
| Title |
|---|
| 《中国优秀硕士学位论文全文数据库（中国学术期刊(光盘版)电子期刊）》 20090131 李雪龙 数字视频中的语义提取分析与研究 , * |
| 《计算机科学》 20061231 俞星玉,汪林林 基于台标轮廓检测的视频监控系统的研究与实现 第33卷, 第9期 * |
Cited By (39)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102281384A (en) * | 2011-07-11 | 2011-12-14 | 成都索贝数码科技股份有限公司 | Method and device for obscuring abandoned image regions, and method and system for processing video |
| CN102306280B (en) * | 2011-07-12 | 2014-04-02 | 央视国际网络有限公司 | Method and device for detecting video scores |
| CN102306279A (en) * | 2011-07-12 | 2012-01-04 | 央视国际网络有限公司 | Method for identifying video scores and device |
| CN102306280A (en) * | 2011-07-12 | 2012-01-04 | 央视国际网络有限公司 | Method and device for detecting video scores |
| CN102426647A (en) * | 2011-10-28 | 2012-04-25 | Tcl集团股份有限公司 | Station identification method and device |
| CN102867170A (en) * | 2012-08-15 | 2013-01-09 | 上海国茂数字技术有限公司 | Method and device for automatically detecting label inserted into video |
| CN102867170B (en) * | 2012-08-15 | 2015-02-11 | 上海国茂数字技术有限公司 | Method and device for automatically detecting label inserted into video |
| CN103002328A (en) * | 2012-11-30 | 2013-03-27 | 北京十分科技有限公司 | Method and device for identifying channels |
| CN103258187A (en) * | 2013-04-16 | 2013-08-21 | 华中科技大学 | Television station caption identification method based on HOG characteristics |
| CN103544467B (en) * | 2013-04-23 | 2017-09-05 | Tcl集团股份有限公司 | Method for distinguishing and its device are known in a kind of station symbol detection |
| CN103544467A (en) * | 2013-04-23 | 2014-01-29 | Tcl集团股份有限公司 | Method and device for detecting and recognizing station captions |
| CN103530655A (en) * | 2013-05-24 | 2014-01-22 | Tcl集团股份有限公司 | Station logo identification method and system |
| CN103530655B (en) * | 2013-05-24 | 2017-09-15 | Tcl集团股份有限公司 | A kind of TV station symbol recognition method and system |
| CN103353939B (en) * | 2013-06-19 | 2016-05-25 | 电子科技大学 | Dynamic background hypograph feature object masterplate extracts and recognition methods |
| CN103353939A (en) * | 2013-06-19 | 2013-10-16 | 电子科技大学 | Method for extracting and identifying image characteristic object template under dynamic background |
| CN103413149A (en) * | 2013-09-06 | 2013-11-27 | 上海美琦浦悦通讯科技有限公司 | Method for detecting and identifying static target in complicated background |
| CN103714314B (en) * | 2013-12-06 | 2017-04-19 | 安徽大学 | Television video station caption identification method combining edge and color information |
| WO2015085637A1 (en) * | 2013-12-09 | 2015-06-18 | 清华大学深圳研究生院 | Method for supplementarily drawing content at station logo region in video |
| CN103618905A (en) * | 2013-12-09 | 2014-03-05 | 清华大学深圳研究生院 | Content drawing method for station caption area in video |
| CN104954854A (en) * | 2014-03-27 | 2015-09-30 | 联想(北京)有限公司 | Station logo identification method and electronic equipment |
| CN104954854B (en) * | 2014-03-27 | 2019-01-15 | 联想(北京)有限公司 | A kind of TV station symbol recognition method and electronic equipment |
| CN104639952A (en) * | 2015-01-23 | 2015-05-20 | 小米科技有限责任公司 | Method and device for identifying station logo |
| CN106203237A (en) * | 2015-05-04 | 2016-12-07 | 杭州海康威视数字技术股份有限公司 | The recognition methods of container-trailer numbering and device |
| US9924226B2 (en) | 2015-05-29 | 2018-03-20 | Xiaomi Inc. | Method and device for processing identification of video file |
| CN104918107B (en) * | 2015-05-29 | 2018-11-02 | 小米科技有限责任公司 | The identification processing method and device of video file |
| CN104918107A (en) * | 2015-05-29 | 2015-09-16 | 小米科技有限责任公司 | Video file identification processing method and device |
| WO2017088463A1 (en) * | 2015-11-24 | 2017-06-01 | 乐视控股（北京）有限公司 | Image processing method and device |
| WO2017088478A1 (en) * | 2015-11-24 | 2017-06-01 | 乐视控股（北京）有限公司 | Number separating method and device |
| CN105869123A (en) * | 2015-11-24 | 2016-08-17 | 乐视致新电子科技（天津）有限公司 | Image processing method and apparatus |
| CN105760842A (en) * | 2016-02-26 | 2016-07-13 | 北京大学 | Station caption identification method based on combination of edge and texture features |
| CN105872591A (en) * | 2016-04-22 | 2016-08-17 | 北京小米移动软件有限公司 | Icon covering method and device |
| CN106454411A (en) * | 2016-10-27 | 2017-02-22 | 北京小米移动软件有限公司 | Station caption processing method and device |
| CN106454411B (en) * | 2016-10-27 | 2020-06-02 | 北京小米移动软件有限公司 | Station caption processing method and device |
| CN107943849B (en) * | 2017-11-03 | 2020-05-08 | 绿湾网络科技有限公司 | Video file retrieval method and device |
| CN107943849A (en) * | 2017-11-03 | 2018-04-20 | 小草数语（北京）科技有限公司 | The search method and device of video file |
| CN108280834A (en) * | 2018-01-22 | 2018-07-13 | 西安万像电子科技有限公司 | Video area determines method and device |
| CN108280834B (en) * | 2018-01-22 | 2019-11-29 | 西安万像电子科技有限公司 | Video area determines method and device |
| CN109783658A (en) * | 2019-02-19 | 2019-05-21 | 苏州科达科技股份有限公司 | Image processing method, device and storage medium |
| CN111368703A (en) * | 2020-02-29 | 2020-07-03 | 上海电力大学 | Platform logo detection and identification method based on FPN |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101950366A (en) | Method for detecting and identifying station logo | |
| CN101739561B (en) | TV station logo training method and identification method | |
| CN105760842A (en) | Station caption identification method based on combination of edge and texture features | |
| CN102426647B (en) | Station identification method and device | |
| CN103258201A (en) | Form line extraction method integrating global information and local information | |
| CN101826209B (en) | Canny model-based method for segmenting three-dimensional medical image | |
| CN103258198A (en) | Extraction method for characters in form document image | |
| CN107506760A (en) | Traffic signals detection method and system based on GPS location and visual pattern processing | |
| CN102799859A (en) | Method for identifying traffic sign | |
| CN105528609A (en) | Vehicle license plate location method based on character position | |
| CN103324935A (en) | Method for vehicle positioning and region segmenting in image | |
| CN102842039B (en) | Road image detection method based on Sobel operator | |
| CN102842037A (en) | Method for removing vehicle shadow based on multi-feature fusion | |
| CN105117727A (en) | Method for fast positioning license plate | |
| CN107330027A (en) | A kind of Weakly supervised depth station caption detection method | |
| CN103106409A (en) | Composite character extraction method aiming at head shoulder detection | |
| CN104992175B (en) | It is a kind of that adhesion steel billet character segmentation method is crossed based on tree-shaped relation | |
| CN103679146A (en) | Safety belt intelligent detection method based on high-pass filter and Hough conversion | |
| CN104063691B (en) | Lane line quick determination method based on improved Hough transform | |
| CN104182728A (en) | Vehicle logo automatic location and recognition method based on pattern recognition | |
| CN104036516A (en) | Camera calibration checkerboard image corner detection method based on symmetry analysis | |
| Tarabek | Fast license plate detection based on edge density and integral edge image | |
| CN102509109A (en) | Method for distinguishing Thangka image from non-Thangka image | |
| CN111881867A (en) | Video analysis method and device and electronic equipment | |
| CN103488966A (en) | Intelligent mobile phone capable of identifying real-name ticket information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20110119 |