CN102005053A - Player position automatic detection method - Google Patents
Player position automatic detection method Download PDFInfo
- Publication number
- CN102005053A CN102005053A CN 201010548695 CN201010548695A CN102005053A CN 102005053 A CN102005053 A CN 102005053A CN 201010548695 CN201010548695 CN 201010548695 CN 201010548695 A CN201010548695 A CN 201010548695A CN 102005053 A CN102005053 A CN 102005053A
- Authority
- CN
- China
- Prior art keywords
- court
- sportsman
- color
- zone
- player
- 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.)
- Granted
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a player position automatic detection method, which is used for automatically detecting the position of a player in a long shot in football video. The method comprises the following steps of: firstly, decoding the football video to obtain continuous long shot frames; secondly, extracting a court color part according to main color from the images of the long shot frames; thirdly, detecting a court area in which the player stays; and finally, detecting the player in the area. The detection method in the invention has very high accuracy and a very high application value, and the whole detection process is simple and rational.
Description
Technical field
The present invention relates in football video the automatic detection of position of Player specifically, is related to a kind of automatic testing method to the position of Player.
Background technology
Football has vast participation always and pay close attention to colony, and the huge commercial value that it contains also makes people that football video is being carried out constantly research as one of most popular ball sports items in the whole world.Wherein the automatic analysis to football video is a main aspect, specifically comprises aspects such as retrieval to video, summary, detection, reinforcement, and then carries out the understanding on the semantic hierarchies, such as tactics with go up form analysis or the like.And for the research of the problems referred to above, the major issue that needs solve is exactly the position of sportsman on the court, and the position of Player provides a lot of semantic informations really surely.On the other hand, popular along with portable mobile terminal can watch football match live whenever and wherever possible on portable terminal, and vast football fan is had very strong temptation undoubtedly.But on the small screen, appreciate football match, the long shot in the match particularly, spectators' visual experience and view and admire and experience and to ensure.A method that solves is spectators' area-of-interest in the displaying video picture, and the delimitation of this area-of-interest also is to need adding sportsman's positional information.
Live football video has the characteristics of himself as a kind of concrete video classification.Activity such as sportsman and ball is all carried out on the football field basically, and there is specific color in the court, and football has its specific shape and cardinal principle color.Sportsman's football shirt generally also has salient feature.In addition, the classification of football video camera lens generally has three kinds: long shot, medium shot and close shot camera lens.Wherein, in medium shot that comprises the sportsman and close shot camera lens, mostly as the main body of camera lens content, the position ratio is easier to determine the sportsman.And in long shot, sportsman's size is smaller with respect to whole camera lens, and the background object in the camera lens is more complicated also, has introduced a lot of noises and can cause strong interference to sportsman's detection.
Summary of the invention
The objective of the invention is to, be the live long shot of football video, provide a kind of automatic testing method, promptly a kind of automatic testing method the position of Player to the accurate detection problem of position of Player.
In order to realize above purpose, a kind of automatic testing method of the present invention to the position of Player, this method is used for detecting automatically the long shot sportsman's of football video position, and described method comprises the steps:
1) to the football video decoding, analyzes its lens type, obtain continuous long shot frame;
2) the long shot two field picture is handled, extracted the court part;
3) zone, court at detection sportsman place;
4) on the zone, court, the sportsman is detected.
Wherein, the described situation of step 1) present football video has usually all been passed through mpeg, has H.264 been waited coding, utilizes ffmpeg just can be decoded into continuous frame.As for determining to lens type, as the method for describing in " Keewon Seo; Jaeseung Ko, Ilkoo Ahn and Changick Kim, in IEEE Trans.CSVT; vol.17; no.10, Oct.2007, An Intelligent Display Scheme of Soccer Video on Mobile Devices ", golden section is carried out in zone to present frame, investigates it and belong to the shared ratio of meadow color pixel in each zone
Technique scheme, described step 2) basis to continuous a plurality of frames (for example in, 30 or 20 frames) statistics of the main areas color on the picture obtains main color, because the major part of big multiframe is exactly the court in the football video, just can determine the cardinal principle color in court according to this main color value, and then extract the court part of image.Described in the document in concrete grammar such as the step 1).
Here, for example can adopt the hsv color model,, also be suitable for differentiating different colors because the hsv color model follows people's perception to suit more.Its color component (Hue) can independently show the color under the different brightness situations, so just can get rid of the interference of court in different weather, light situation.
Technique scheme, described step 3) also comprises following substep:
(3-1) the long shot two field picture is divided into the fritter of 16x16,, delimit according to its ratio that belongs to court colored pixels point and to be court color block or non-court color block for each fritter;
(3-2) again each the non-court color block that obtains is further judged, if most right and wrong court region unit around this piece is to be positioned at outside the court just can assert this piece; Otherwise assert that just this piece is the non-court part that is positioned on the zone, court, so far all court color block and the non-court color block that is positioned on the court lump together, and have formed the zone, court, on this zone, court the sportsman are detected.
Technique scheme, described step 4) also comprises following substep:
(4-1) be set at 1 for non-court color pixel point values all on the zone, court, other pixel point value is set at 0, sets up the binary map in this zone, court;
(4-2) detecting value on the binary map is 1 all objects, screens according to its size, gets rid of undersized and excessive object and noise.
Wherein, described step (4-2) is got rid of at oversize or too small object, and concrete steps are as follows:
(a) size with football is set in 80 to 150 pixel scopes, is not sportsman and directly leaving out if the dimension of object on the binary map during less than 150 pixels, is judged, gets rid of too small object;
(b) remaining object on the binary map being done refinement handles, at after the thinning processing have a few, the height that with each point is the center be 25 pixels wide be in the rectangle frame of 12 pixels, can reach more than 75% if the value of pixel is 1 ratio, so just think this central point and on every side within the rectangle frame all values be that 1 point all may belong to sportsman's part, otherwise just judge that this central point does not belong to sportsman's part;
(c) may do last judgement for sportsman's part to detected in (b), remove the excessive or too small zone of area depth-width ratio too small or its boundary rectangle that the excessive or last step of its area obtains, what finally obtain is sportsman's part.
Beneficial effect to the automatic testing method of position of Player in the long shot of football video of the present invention is: comprehensive various information, guarantee that testing result has higher accuracy and whole testing process all is to carry out in real time, and have bigger using value.
Description of drawings
Fig. 1 is the overview flow chart of the automatic testing method of position of Player in the football video long shot of the present invention;
The synoptic diagram of the long shot frame that figure (2-a) is pending;
The synoptic diagram that figure (2-b) detects the court part according to color;
The synoptic diagram of figure (2-c) 16x16 court part piece;
Figure (2-d) comprises sportsman's court area schematic;
The court area schematic that comprises the sportsman that figure (2-e) obtains based on binary map;
Obtain the synoptic diagram of final position of Player after figure (2-f) screening.
Embodiment
Describe in further detail below in conjunction with accompanying drawing and concrete embodiment automatic testing method position of Player in the football video long shot of the present invention.
Fig. 1 is the automatic testing method overview flow chart of position of Player in the football video long shot of the present invention.Specifically comprise the steps:
1) long shot in the football video is partly decoded, obtain continuous long shot frame, figure (2-a) is the pending long shot frame of intercepting;
Usually, present football video has all passed through mpeg, has H.264 waited coding, utilizes ffmpeg just can be decoded into continuous frame.As for to the determining of lens type, golden section is carried out in the zone of present frame, in each zone investigation its belong to the shared ratio of meadow color pixel.
2) the long shot two field picture is handled, is extracted the court part:
Obtain main color according to statistics to the color on continuous a plurality of frames (30 frames or 20 frames) picture, because the major part of big multiframe is exactly the court in the football video, just can determine the cardinal principle color in court according to this main color value, and then extract the court part of image, shown in figure (2-b).
Here, for example can adopt the hsv color model,, also be suitable for differentiating different colors because the hsv color model follows people's perception to suit more.Its color component (Hue) can independently show the color under the different brightness situations, so just can get rid of the interference of court in different weather, light situation.
3) detect the zone, court:
The long shot two field picture is divided into the fritter of 16x16, for each fritter, delimit according to its ratio that belongs to court colored pixels point and to be court color block and non-court color block, shown in figure (2-c).
Again each non-court color block is judged, if most right and wrong court region unit around this piece is to be positioned at outside the court just can assert this piece; Otherwise assert that just this piece is the non-court part (sportsman, judge or other noise) that is positioned on the zone, court.All like this court color block and the non-court color block that is positioned on the court lump together, and have formed the zone, court, shown in figure (2-d).We just detect the sportsman on this zone, court.
4) on the zone, court, the sportsman is detected:
Be set at 1 for the non-court color pixel point value on the zone, court, other all pixel point values are set at 0 on the image, set up binary map, shown in figure (2-e).
Value is 1 all objects on the detection binary map, screens according to its size, gets rid of undersized object and noise.This can not get rid of oversize object simultaneously, overlaps because bigger object may be white line (sideline, forbidden zone line, center line or middle astragal) and sportsman on the court.In this case, must remove the positional information that white line obtains independent sportsman.Concrete disposal route is as follows:
(a) among the present invention according to experimental data, the size of football is set in 80 to 150 pixel scopes.When the dimension of object on the binary map during less than 150 pixels, sportsman's possibility is just very low during this object, can judge not to be the sportsman and directly to leave out.
(b) remaining object on the binary saliency map being done refinement handles, at after the thinning processing have a few, in high 25 wide 12 the rectangle frame that with this point is the center, can reach more than 75% if the value of pixel is 1 ratio, so just think this central point and on every side the point within the rectangle frame all may belong to sportsman's part, otherwise just judge that this central point does not belong to sportsman's part.
(c) detected sportsman in (b) is partly done last judgement, remove the zone that its area is excessive or too small, the depth-width ratio of its boundary rectangle is excessive or too small, what finally obtain is sportsman's part, shown in figure (2-f).
It should be noted last that above embodiment is only unrestricted in order to technical scheme of the present invention to be described.Although the present invention is had been described in detail with reference to embodiment, those of ordinary skill in the art is to be understood that, technical scheme of the present invention is made amendment or is equal to replacement, do not break away from the spirit and scope of technical solution of the present invention, it all should be encompassed in the middle of the claim scope of the present invention.
Claims (6)
1. automatic testing method to the position of Player, this method are used for detecting automatically the long shot sportsman's of football video position, and described method comprises the steps:
1) to the football video decoding, analyzes its lens type, obtain continuous long shot frame;
2) the long shot two field picture is handled, extracted the court part;
3) zone, court at detection sportsman place;
4) on the zone, court, the sportsman is detected.
2. the automatic testing method described in claim 1 to the position of Player, it is characterized in that, described step 2) in, obtain main color according to statistics to the main areas color on continuous 20 frames or the 30 frame pictures, can determine the cardinal principle color in court according to this main color value, and then extract the court part of image.
3. the automatic testing method to the position of Player as claimed in claim 2 is characterized in that, described main color obtains by the hsv color model.
4. the automatic testing method to the position of Player described in claim 1 is characterized in that, described step 3) also comprises following substep:
(3-1) the long shot two field picture is divided into the fritter of 16x16,, delimit according to its ratio that belongs to court colored pixels point and to be court color block or non-court color block for each fritter;
(3-2) again each the non-court color block that obtains is further judged, if most right and wrong court region unit around this piece is to be positioned at outside the court just can assert this piece; Otherwise assert that just this piece is the non-court part that is positioned on the zone, court, so far all court color block and the non-court color block that is positioned on the court lump together, and have formed the zone, court, on this zone, court the sportsman are detected.
5. the automatic testing method to the position of Player described in claim 1 is characterized in that, described step 4) comprises following substep:
(4-1) be set at 1 for non-court color pixel point values all on the zone, court, other pixel point value is set at 0, sets up the binary map in this zone, court;
(4-2) detecting value on the binary map is 1 all objects, screens according to its size, gets rid of undersized and excessive object and noise.
6. the automatic testing method to the position of Player as claimed in claim 5 is characterized in that, it is as follows that described step (4-2) is got rid of concrete steps at oversize or too small object:
(a) size with football is set in 80 to 150 pixel scopes, is not sportsman and directly leaving out if the dimension of object on the binary map during less than 150 pixels, is judged, is used to get rid of undersized object;
(b) remaining object on the binary map being done refinement handles, at after the thinning processing have a few, the height that with each point is the center be 25 pixels wide be in the rectangle frame of 12 pixels, can reach more than 75% if the value of pixel is 1 ratio, so just think this central point and on every side within the rectangle frame all values be that 1 point all may belong to sportsman's part, otherwise just judge that this central point does not belong to sportsman's part;
(c) may do last judgement for sportsman's part to detected in the step (b), remove the excessive or too small zone of its area area excessive or that step (b) obtains depth-width ratio too small or its boundary rectangle, what finally obtain is sportsman's part.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010548695 CN102005053B (en) | 2010-11-17 | 2010-11-17 | Player position automatic detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010548695 CN102005053B (en) | 2010-11-17 | 2010-11-17 | Player position automatic detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102005053A true CN102005053A (en) | 2011-04-06 |
| CN102005053B CN102005053B (en) | 2012-10-17 |
Family
ID=43812392
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201010548695 Expired - Fee Related CN102005053B (en) | 2010-11-17 | 2010-11-17 | Player position automatic detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102005053B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103745483A (en) * | 2013-12-20 | 2014-04-23 | 成都体育学院 | Mobile-target position automatic detection method based on stadium match video images |
| CN106651952A (en) * | 2016-10-27 | 2017-05-10 | 深圳锐取信息技术股份有限公司 | Football detecting and tracking based video processing method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101324957A (en) * | 2008-07-16 | 2008-12-17 | 上海大学 | Intelligent playing method of football video facing to mobile equipment |
| CN101645137A (en) * | 2009-07-17 | 2010-02-10 | 中国科学院声学研究所 | Method for automatically detecting location of a football in long shot of football video |
| CN101681433A (en) * | 2007-03-26 | 2010-03-24 | 汤姆森特许公司 | Method and apparatus for detecting objects of interest in soccer video by color segmentation and shape anaylsis |
-
2010
- 2010-11-17 CN CN 201010548695 patent/CN102005053B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101681433A (en) * | 2007-03-26 | 2010-03-24 | 汤姆森特许公司 | Method and apparatus for detecting objects of interest in soccer video by color segmentation and shape anaylsis |
| CN101324957A (en) * | 2008-07-16 | 2008-12-17 | 上海大学 | Intelligent playing method of football video facing to mobile equipment |
| CN101645137A (en) * | 2009-07-17 | 2010-02-10 | 中国科学院声学研究所 | Method for automatically detecting location of a football in long shot of football video |
Non-Patent Citations (1)
| Title |
|---|
| 《IEEE Transactions On Circuits and Systems For Video Technology》 20071031 Keewon Seo,Jaeseung Ko An Intelligent Display Scheme of Soccer Video on Mobile Devices 第17卷, 第10期 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103745483A (en) * | 2013-12-20 | 2014-04-23 | 成都体育学院 | Mobile-target position automatic detection method based on stadium match video images |
| CN103745483B (en) * | 2013-12-20 | 2017-02-15 | 成都体育学院 | Mobile-target position automatic detection method based on stadium match video images |
| CN106651952A (en) * | 2016-10-27 | 2017-05-10 | 深圳锐取信息技术股份有限公司 | Football detecting and tracking based video processing method and device |
| CN106651952B (en) * | 2016-10-27 | 2020-10-20 | 深圳锐取信息技术股份有限公司 | Video processing method and device based on football detection and tracking |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102005053B (en) | 2012-10-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101645137B (en) | Method for automatically detecting location of a football in long shot of football video | |
| CN105654471A (en) | Augmented reality AR system applied to internet video live broadcast and method thereof | |
| CN107103606A (en) | A kind of image-recognizing method and device | |
| CN103310422B (en) | Obtain the method and device of image | |
| CN109214999A (en) | A kind of removing method and device of video caption | |
| US10181192B1 (en) | Background modelling of sport videos | |
| CN101324957B (en) | Intelligent playing method of football video facing to mobile equipment | |
| CN107392879B (en) | A kind of low-light (level) monitoring image Enhancement Method based on reference frame | |
| CN105989583B (en) | A kind of image defogging method | |
| CN108961227A (en) | A kind of image quality evaluating method based on airspace and transform domain multiple features fusion | |
| CN100442307C (en) | Goal checking and football video highlight event checking method based on the goal checking | |
| CN102005053B (en) | Player position automatic detection method | |
| CN109583414B (en) | Indoor road occupation detection method, device, medium and processor based on video detection | |
| CN102306153A (en) | Method for detecting goal events based on normalized semantic weighting and regular football video | |
| CN104299234B (en) | The method and system that rain field removes in video data | |
| CN114187664A (en) | Rope skipping counting system based on artificial intelligence | |
| CN113052923B (en) | Tone mapping method, tone mapping apparatus, electronic device, and storage medium | |
| CN102024264B (en) | Dimensional-histogram-statistic-based touch pen color recognition method | |
| CN103729624B (en) | A kind of light measuring method and photometric system based on skin color model | |
| CN107067004A (en) | The adjusting method and device of a kind of electronic equipment and its screen intensity | |
| CN103366368A (en) | Double-truncated-cone-cylinder codebook foreground detection method capable of eliminating shadow and highlight noise | |
| CN102625028A (en) | Method and apparatus for detecting static logo existing in video | |
| CN104469239A (en) | Immersive video presenting method of smart mobile terminal | |
| CN106651952B (en) | Video processing method and device based on football detection and tracking | |
| Shen et al. | A method of billiard objects detection based on Snooker game video |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20121017 Termination date: 20151117 |
|
| EXPY | Termination of patent right or utility model |