CN107124642A - The detection method and system of captions in continuous moving image - Google Patents
The detection method and system of captions in continuous moving image Download PDFInfo
- Publication number
- CN107124642A CN107124642A CN201710133750.XA CN201710133750A CN107124642A CN 107124642 A CN107124642 A CN 107124642A CN 201710133750 A CN201710133750 A CN 201710133750A CN 107124642 A CN107124642 A CN 107124642A
- Authority
- CN
- China
- Prior art keywords
- logo
- captions
- colourity
- brightness
- region
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 21
- 238000007689 inspection Methods 0.000 claims abstract description 16
- 238000000034 method Methods 0.000 claims abstract description 9
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000003068 static effect Effects 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 3
- 230000004224 protection Effects 0.000 abstract description 4
- 230000000694 effects Effects 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 5
- VJTAZCKMHINUKO-UHFFFAOYSA-M chloro(2-methoxyethyl)mercury Chemical compound [Cl-].COCC[Hg+] VJTAZCKMHINUKO-UHFFFAOYSA-M 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 206010019133 Hangover Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs
- H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/431—Generation of visual interfaces for content selection or interaction; Content or additional data rendering
- H04N21/4312—Generation of visual interfaces for content selection or interaction; Content or additional data rendering involving specific graphical features, e.g. screen layout, special fonts or colors, blinking icons, highlights or animations
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/435—Processing of additional data, e.g. decrypting of additional data, reconstructing software from modules extracted from the transport stream
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
Abstract
The invention provides the detection method and system of captions in a kind of continuous moving image, including detecting step:Detect the logo parts in video pictures;Colourity and brightness statistics step:The brightness for the logo that statistic mixed-state is arrived and colourity;Frequency statistics step:There is the frequency of occurrences of logo in logo region in picture in statistics;Captions discriminating step:If different logo can be repeatedly detected in some region or specified inspection area, and difference logo brightness, colourity are approximate, then it is assumed that the region occurs in that captions logo.Method in the present invention can increase the protection to moving captions, improve interpolation image effect, it is to avoid move the broken of captions, and only checked in the normal region of captions set in advance, rather than full frame inspection, save resource, and detection efficiency is high.
Description
Technical field
The present invention relates to video caption processing technology field, in particular it relates in continuous moving image captions detection side
Method and system.
Background technology
Current LCD TV has replaced traditional CRT TVs to turn into the main flow in market, and the display yet with liquid crystal molecule is special
Property, there is picture conditions of streaking in liquid crystal display viewing TV, if refresh rate is not enough, and some motion pictures, which also can be appreciated that, significantly to be trembled
It is dynamic.One of the common methods for solving shake and hangover are that frame per second conversion is generally used in lifting tab source frame rate, current TV chip
Motion estimation motion compensates (Motion estimation and motion compensation, MEMC) technology.MEMC technologies
Application drastically increase viewing experience, but algorithm has limitation under many scenes, and motion search can not be found just
True motion vector, for example compound movement, repetitive structure, captions logo etc..Wherein, image content is being moved, but Logo
Transfixion, logo motion vector is easy to the vector influence of passive movement background, it is impossible to converge to zero vector.
Current solution is detection of the increase to static logo, uses zero vector in logo place, interpolation when
Time is protected to it.However, captions logo had both possessed common logo inactive, and change in time than common logo
It hurry up, therefore traditional logo protections are difficult to protect captions logo to crush in motion estimation motion compensation to fly out.
The content of the invention
For defect of the prior art, it is an object of the invention to provide a kind of detection side of captions in continuous moving image
Method and system.
The detection method of captions, comprises the following steps in the continuous moving image provided according to the present invention:
Detecting step:Detect the logo parts in video pictures;
Colourity and brightness statistics step:The brightness for the logo that statistic mixed-state is arrived and colourity;
Frequency statistics step:There is the frequency of occurrences of logo in logo region in picture in statistics;
Captions discriminating step:If different logo can be repeatedly detected in some region or specified inspection area,
The brightness to the front and rear logo detected, the similitude of colourity are judged, if similar, then it is assumed that the region occurs in that captions logo.
Preferably, the detecting step includes:Static logo is detected using based on local feature, specifically, bag
Include following steps:
Step A1:Divide the image into as M × N blocks, the block detection logo based on segmentation;
Step A2:The local feature for each block split in calculation procedure A1, the local feature includes:SIFT feature,
SURF features, ORB features, HOG features, or any one local feature algorithm for description block particularity;
Step A3:Compare local feature description's of block at same position in two field pictures, with local feature description's
Distance weighs similitude of the block in two interframe, and the similitude distinguishing rule includes:Absolute difference, similitude is higher, then corresponds to
Block is bigger for logo possibility;
Step A4:If on consecutive hours countershaft, local feature description's son difference of adjacent two frame of corresponding position block is at certain
It is less than given threshold in one setting time section always, then it is assumed that the block detected is logo.
Preferably, the frequency statistics step includes:Check whether certain region can be repeatedly detected logo, and judge inspection
Whether the logo measured is that the content being continually changing (counts the frequency of occurrences of the logo in the region, if occurring repeatedly, corresponded to
High frequency), if, then it is assumed that detect captions.
Preferably, the specified inspection area in the captions discriminating step includes:Below screen or screen top position.
Different logo brightness, colourity are compared in captions discriminating step, comprised the following steps that:
If different logo can be repeatedly detected in some region or specified inspection area, detected to front and rear
Logo brightness, colourity ask poor respectively, and colour difference is designated as:Diff_C, luminance difference is designated as:Diff_Y, if diff_Y<Th1, and
diff_C<th2;The logo detected before and after then thinking brightness is similar with colourity, wherein, th1, th2 is the threshold that user gives
It is worth higher limit, if the front and rear logo detected brightness is similar with colourity, then it is assumed that the region occurs in that captions logo.
The detecting system of captions in the continuous moving image provided according to the present invention, including following module:
Detection module:For detecting the logo parts in video pictures;
Colourity and brightness statistic:The logo arrived for statistic mixed-state brightness and colourity;
Frequency statistics module:Occurs the frequency of occurrences of logo in logo region in picture for counting;
Captions discrimination module:For discriminating whether that in some region or specified inspection area difference can be repeatedly detected
Logo, if the different logo detected brightness, colourity are similar, then it is assumed that the region occurs in that captions logo.
Compared with prior art, the present invention has following beneficial effect:
1st, the detection method of captions can increase the protection to moving captions in the continuous moving image provided in the present invention,
Improve interpolation image effect, it is to avoid move the broken of captions
2nd, the detection method of captions only often goes out in the captions set in advance in the continuous moving image of the offer in the present invention
Existing region is checked, rather than full frame inspection, saves resource, efficiency high.
Brief description of the drawings
By reading the detailed description made with reference to the following drawings to non-limiting example, further feature of the invention,
Objects and advantages will become more apparent upon:
The method flow schematic diagram that Fig. 1 provides for the present invention.
Embodiment
With reference to specific embodiment, the present invention is described in detail.Following examples will be helpful to the technology of this area
Personnel further understand the present invention, but the invention is not limited in any way.It should be pointed out that to the ordinary skill of this area
For personnel, without departing from the inventive concept of the premise, some changes and improvements can also be made.These belong to the present invention
Protection domain.
The detection method of captions, comprises the following steps in the continuous moving image provided according to the present invention:
Step 1:Detect the logo parts in video pictures;
Step 2:The brightness for the logo that statistic mixed-state is arrived and colourity;
Step 3:There is the frequency of occurrences of logo in logo region in picture in statistics;
Step 4:If different logo can be repeatedly detected in some region or specified inspection area, judge to preceding
The logo detected afterwards brightness, the similitude of colourity, if similar, then it is assumed that the region occurs in that captions logo.
The step 1 includes:Static logo is detected using based on local feature, specifically, comprised the following steps:
Step 1.1:Divide the image into as M × N blocks, the block detection logo based on segmentation;
Step 1.2:The local feature for each block split in calculation procedure 1.1, the local feature includes:SIFT is special
Levy, SURF features, ORB features, HOG features, or any one be used for description block particularity local feature algorithm;
Step 1.3:Compare local feature description's of block at same position in two field pictures, with local feature description's
Distance weighs similitude of the block in two interframe, and the similitude distinguishing rule includes:Absolute difference, similitude is higher, then corresponds to
Block is bigger for logo possibility;
Step 1.4:If on consecutive hours countershaft, local feature description's son difference of adjacent two frame of corresponding position block is at certain
It is less than given threshold in one setting time section always, then it is assumed that the block detected is logo.
The step 3 includes:Check that (statistics is made in the region that only captions can be selected often to occur, and for example shields in certain region
Curtain lower section) logo whether can be repeatedly detected, and judge whether the logo detected is the content being continually changing, if so, then
Think to detect captions, count the frequency of occurrences of the logo in the region.
Specified inspection area in the step 4 includes:Below screen or screen top position.
Different logo brightness, colourity are compared in step 4, comprised the following steps that:
If different logo can be repeatedly detected in some region or specified inspection area, detected to front and rear
Logo brightness, colourity ask poor respectively, and colour difference is designated as:Diff_C, luminance difference is designated as:Diff_Y, if diff_Y<Th1, and
diff_C<th2;The logo detected before and after then thinking brightness is similar with colourity, wherein, th1, th2 is the threshold that user gives
It is worth higher limit, if the front and rear logo detected brightness is similar with colourity, then it is assumed that the region occurs in that captions logo.
The detecting system of captions in the continuous moving image provided according to the present invention, including following module:
Detection module:For detecting the logo parts in video pictures;
Colourity and brightness statistic:The logo arrived for statistic mixed-state brightness and colourity;
Frequency statistics module:Occurs the frequency of occurrences of logo in logo region in picture for counting;
Captions discrimination module:For discriminating whether that in some region or specified inspection area difference can be repeatedly detected
Logo, if the different logo detected brightness, colourity are similar, then it is assumed that the region occurs in that captions logo.
More detailed explanation is done to the technical scheme in the present invention with reference to specific embodiment.
Embodiment using method in the present invention comprises the following steps:
Step S1:Logo is detected based on local feature algorithm, image is divided into block, the hog features of calculation block;
Step S2:The otherness of frame same position block hog features before and after calculating:Diff=abs (HOGpre-HOGcur),
diff<Th then Logo_Flag=1;
Step S3:On consecutive hours countershaft, time>In th, Logo_Flag is considered Logo for 1 block always;
Step S4:All pixels to present frame labeled as Logo ask its brightness, colourity average value;
Step S5:Th in step S3 is set to a less value, in time<In P period, P>Th, is referring to
Determine in region, detect logo more than n times, and front and rear logo brightness, colourity difference are smaller, diff_Y<th_Y,diff_C
<Th_C, then think that the region occurs in that motion captions logo;
Step S6:Conclusion in step S5 is dynamically adapted the th in step S3, it is believed that when having motion captions at this,
Th can set sufficiently small, you can detect logo quickly, and captions are protected;During without captions, th sets larger, it is to avoid mistake is protected
Shield.
The specific embodiment of the present invention is described above.It is to be appreciated that the invention is not limited in above-mentioned
Particular implementation, those skilled in the art can make a variety of changes or change within the scope of the claims, this not shadow
Ring the substantive content of the present invention.In the case where not conflicting, feature in embodiments herein and embodiment can any phase
Mutually combination.
Claims (5)
1. the detection method of captions in a kind of continuous moving image, it is characterised in that comprise the following steps:
Detecting step:Detect the logo parts in video pictures;
Colourity and brightness statistics step:The brightness for the logo that statistic mixed-state is arrived and colourity;
Frequency statistics step:There is the frequency of occurrences of logo in logo region in picture in statistics;
Captions discriminating step:If different logo can be repeatedly detected in some region or specified inspection area, judge
The similitude of brightness, colourity to the front and rear logo detected, if similar, then it is assumed that the region occurs in that captions logo.
2. the detection method of captions in continuous moving image according to claim 1, it is characterised in that the detecting step
Including:Static logo is detected using based on local feature, specifically, comprised the following steps:
Step A1:Divide the image into as M × N blocks, the block detection logo based on segmentation;
Step A2:The local feature for each block split in calculation procedure A1, the local feature includes:SIFT feature, SURF
Feature, ORB features, HOG features, or any one local feature algorithm for description block particularity;
Step A3:Compare local feature description's of block at same position in two field pictures, with the distance of local feature description's
To weigh similitude of the block in two interframe, the similitude distinguishing rule includes:Absolute difference, similitude is higher, then corresponding blocks are
Logo possibility is bigger;
Step A4:If on consecutive hours countershaft, local feature description's son difference of adjacent two frame of corresponding position block is set a certain
Fix time and be less than given threshold in section always, then it is assumed that the block detected is logo.
3. the detection method of captions in continuous moving image according to claim 1, it is characterised in that the frequency statistics
Step includes:Check whether certain region can be repeatedly detected logo, and judge whether the logo detected is continually changing
Content, if, then it is assumed that detect captions.
4. the detection method of captions in continuous moving image according to claim 1, it is characterised in that the captions differentiate
Specified inspection area in step includes:Below screen or screen top position.
Different logo brightness, colourity are compared in captions discriminating step, comprised the following steps that:
If different logo can be repeatedly detected in some region or specified inspection area, to the front and rear logo detected
Brightness, colourity ask poor respectively, colour difference is designated as:Diff_C, luminance difference is designated as:Diff_Y, if diff_Y<Th1, and
diff_C<th2;The logo detected before and after then thinking brightness is similar with colourity, wherein, th1, th2 is the threshold that user gives
It is worth higher limit, if the front and rear logo detected brightness is similar with colourity, then it is assumed that the region occurs in that captions logo.
5. the detecting system of captions in a kind of continuous moving image, it is characterised in that including following module:
Detection module:For detecting the logo parts in video pictures;
Colourity and brightness statistic:The logo arrived for statistic mixed-state brightness and colourity;
Frequency statistics module:Occurs the frequency of occurrences of logo in logo region in picture for counting;
Captions discrimination module:It is different for discriminating whether can be repeatedly detected in some region or specified inspection area
Logo, if the different logo detected brightness, colourity are similar, then it is assumed that the region occurs in that captions logo.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710133750.XA CN107124642A (en) | 2017-03-08 | 2017-03-08 | The detection method and system of captions in continuous moving image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710133750.XA CN107124642A (en) | 2017-03-08 | 2017-03-08 | The detection method and system of captions in continuous moving image |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107124642A true CN107124642A (en) | 2017-09-01 |
Family
ID=59717391
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710133750.XA Pending CN107124642A (en) | 2017-03-08 | 2017-03-08 | The detection method and system of captions in continuous moving image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107124642A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103607635A (en) * | 2013-10-08 | 2014-02-26 | 十分(北京)信息科技有限公司 | Method, device and terminal for caption identification |
CN103870795A (en) * | 2012-12-13 | 2014-06-18 | 北京捷成世纪科技股份有限公司 | Automatic detection method and device of video rolling subtitle |
CN104602096A (en) * | 2014-12-26 | 2015-05-06 | 北京奇艺世纪科技有限公司 | Detecting method and device for video subtitle area |
CN104780362A (en) * | 2015-04-24 | 2015-07-15 | 宏祐图像科技(上海)有限公司 | Video static logo detecting method based on local feature description |
CN104980765A (en) * | 2015-06-15 | 2015-10-14 | 北京博威康技术有限公司 | Multi-channel plain text frame monitoring method |
US20160189380A1 (en) * | 2014-12-24 | 2016-06-30 | Samsung Electronics Co., Ltd. | Method and apparatus for processing image |
-
2017
- 2017-03-08 CN CN201710133750.XA patent/CN107124642A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103870795A (en) * | 2012-12-13 | 2014-06-18 | 北京捷成世纪科技股份有限公司 | Automatic detection method and device of video rolling subtitle |
CN103607635A (en) * | 2013-10-08 | 2014-02-26 | 十分(北京)信息科技有限公司 | Method, device and terminal for caption identification |
US20160189380A1 (en) * | 2014-12-24 | 2016-06-30 | Samsung Electronics Co., Ltd. | Method and apparatus for processing image |
CN104602096A (en) * | 2014-12-26 | 2015-05-06 | 北京奇艺世纪科技有限公司 | Detecting method and device for video subtitle area |
CN104780362A (en) * | 2015-04-24 | 2015-07-15 | 宏祐图像科技(上海)有限公司 | Video static logo detecting method based on local feature description |
CN104980765A (en) * | 2015-06-15 | 2015-10-14 | 北京博威康技术有限公司 | Multi-channel plain text frame monitoring method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US8144255B2 (en) | Still subtitle detection apparatus and image processing method therefor | |
CN105282475B (en) | Crawl detection and compensation method and system | |
US10540791B2 (en) | Image processing apparatus, and image processing method for performing scaling processing based on image characteristics | |
US8355079B2 (en) | Temporally consistent caption detection on videos using a 3D spatiotemporal method | |
JP2006072359A (en) | Method of controlling display apparatus | |
CA2920834A1 (en) | Legibility enhancement for a logo, text or other region of interest in video | |
JP2004080252A (en) | Video display unit and its method | |
TWI504248B (en) | Image processing apparatus and image processing method | |
CN102044067A (en) | Method and system for reducing ringing artifacts of image deconvolution | |
CN102724458A (en) | Video picture full-screen display subtitle processing method and video terminal | |
CN104010114B (en) | Video denoising method and device | |
US20120182311A1 (en) | Image displaying apparatus | |
US9516260B2 (en) | Video processing method and apparatus | |
CN102254544A (en) | Method for automatically adjusting video signal proportion and television using same | |
KR20140046370A (en) | Method and apparatus for detecting a television channel change event | |
CN102629969A (en) | Smear eliminating method during shooting of plane objects | |
CN107124642A (en) | The detection method and system of captions in continuous moving image | |
WO2016199418A1 (en) | Frame rate conversion system | |
CN108074248B (en) | OSD automatic detection method and device based on image content | |
US9715736B2 (en) | Method and apparatus to detect artificial edges in images | |
JP5188272B2 (en) | Video processing apparatus and video display apparatus | |
JP2008147826A (en) | Black stripe region detection circuit | |
US8879000B2 (en) | Method and system for detecting analog noise in the presence of mosquito noise | |
US20130201404A1 (en) | Image processing method | |
US8233085B1 (en) | Method and system for interpolating a pixel value of a pixel located at an on-screen display |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20190712 Address after: 266000 Shandong Province, Qingdao city Laoshan District Songling Road No. 399 Applicant after: Qingdao Xinxin Microelectronics Technology Co., Ltd. Address before: Room 316, Building 88 Chenhui Road, China (Shanghai) Free Trade Pilot Area, Pudong New Area, Shanghai, 201203 Applicant before: HONGYOU IMAGE TECHNOLOGY (SHANGHAI) CO., LTD. |
|
TA01 | Transfer of patent application right | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170901 |
|
RJ01 | Rejection of invention patent application after publication |