CN101193248A - Method for sectioning image data according to scenario change - Google Patents
Method for sectioning image data according to scenario change Download PDFInfo
- Publication number
- CN101193248A CN101193248A CNA2006101492649A CN200610149264A CN101193248A CN 101193248 A CN101193248 A CN 101193248A CN A2006101492649 A CNA2006101492649 A CN A2006101492649A CN 200610149264 A CN200610149264 A CN 200610149264A CN 101193248 A CN101193248 A CN 101193248A
- Authority
- CN
- China
- Prior art keywords
- image
- image frame
- image data
- scene
- feature value
- 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 relates to a method for carrying out fragments of image data according to changes of a scene. The method includes the following steps: (a) a first characteristic value of a first image frame in the image data is calculated; (b) a second characteristic value of a second image frame in the image data is calculated; (c) a difference value between the characteristic value of the first image and the characteristic value of the second image is compared to judge the value is more than or equal to a threshold; (d) if the difference value is more than or equal to the threshold, the fragments of the image data are carried out so that the first image frame and the second image frame belong to a first section and a second section respectively.
Description
Technical field
The present invention relates to a kind of method of image data segmentation, particularly relate to a kind of method of carrying out the image data segmentation that changes according to scene.
Background technology
Along with Development of Multimedia Technology, the related application of multi-medium data provides the consumer different user demands, and with digital form storage or play multimedia data, and the user also can be from the medium that store multi-medium data, choose required content according to demand separately, for example use the existing fast function of changeing or reversing to reach the fragment purpose of fast searching multi-medium data.The mode of existing image data segmentation (chapter) is to take every a specific period this image data to be carried out the mechanism of segmentation, for instance, see also Fig. 1, Fig. 1 carries out segmentation for an existing image data every a specific period schematic diagram, by finding out among Fig. 1, if the total length of time of this image data is 20 minutes, can carry out segmentation to this image data every 5 minutes, this image data can be divided into four sections (first section, second section, the 3rd section, the 4th section) thus, and each section time span is 5 minutes.It is not best that yet the shortcoming of this kind image segmented mode is the association between each section, for instance may one section the hitter hits flyball in the baseball game film two sections pictures that connect flyball extremely with wild hand afterwards in a flash by segmentation to different sections, just can cause the user to watch inconvenience with the searching multi-medium data thus, so this mode is the image data segmented mode of impersonalityization.
Summary of the invention
The present invention discloses a kind of method of carrying out the image data segmentation according to the scene variation, it includes the following step: the one first image feature value of (a) calculating one first image frame of an image data; (b) calculate one second image feature value of one second image frame of this image data; (c) whether compare a difference of this first image feature value and this second image feature value more than or equal to a threshold value; And (d) if this difference more than or equal to this threshold value, this image data of segmentation is so that this first image frame and this second image frame belong to one first section and one second section respectively.
The present invention has also disclosed and has a kind ofly changed the method carry out the image data segmentation according to scene, and it includes the following step: (a) judge between one first image frame of an image data and one second image frame whether have scene variation; And (b) if this scene changes to be existed, this image data of segmentation is so that this first image frame and this second image frame belong to one first section and one second section respectively.
Description of drawings
Fig. 1 carries out segmentation for previous image data every a specific period schematic diagram.
Fig. 2 changes the flow chart that carries out the image data segmentation for the present invention according to scene.
Fig. 3 changes the schematic diagram that carries out segmentation with the segmentation condition of time restriction for image data of the present invention according to scene.
The reference numeral explanation
Step 100,102,104,106,108,110,112,114
Embodiment
See also Fig. 2, Fig. 2 changes the flow chart that carries out the image data segmentation for the present invention according to scene, and this method comprises the following step:
Step 100: set an initial time point T0.
Step 102: calculate the image feature value of an image data in one first image frame (imageframe) of very first time point T1.
Step 104: calculate the image feature value of this image data in one second image frame of one second time point T2.
Step 106: whether the difference of image feature value of judging the image feature value of this second image frame and this first image frame more than or equal to a threshold value, if execution in step 108 then; If not, execution in step 110 then.
Step 108: in this image data being carried out segmentation with respect to this very first time point T1 of this first image frame and between with respect to this second time point T2 of this second image frame, meaning is about to this first image frame and divides in different sections with this second image frame, for example makes this first image frame and this second image frame belong to one first section and one second section respectively.
Step 110: whether the difference of judging this second time point T2 and this initial time point T0 more than or equal to a scheduled time, if execution in step 108 then; If not, execution in step 112 then.
Step 112: this image data is not carried out segmentation between this very first time point T1 and this second time point T2, meaning is about to this first image frame and this second image frame is divided in same section.
Step 114: finish.
In this above-mentioned steps is described in detail, main spirit of the present invention is according to the scene variation this image data to be carried out the image data segmentation, and the standard that the scene variation is judged can be with reference to the content of MPEG standard, in present embodiment, be the image feature value of the image frame that utilize to calculate this image data two time points, and whether reach the condition that scene changes between two time points before and after judging by the image feature value of the image frame of two time points before and after relatively in front and back.In addition, the condition of arranging in pairs or groups again and this image data being carried out segmentation every a specific period, and as the foundation that this image data is carried out the image data segmentation.
At first the starting point in this image data can be set at this initial time point T0, just this image data is calculated its image feature value in the image frame of different time points afterwards, wherein the image feature value of each image frame system can be this image frame in the brightness integrated value of each space coordinates point etc.In step 102 and step 104, the two time point T1 in front and back, during T2, calculate respectively this image data in the image feature value of this first image frame of this some T1 and this image data very first time in the image feature value of this second image frame of this second time point T2, whether the difference of image feature value of next judging the image feature value of this second image frame and this first image frame again is more than or equal to this threshold value, wherein this threshold value can be set up on their own by the user, if the difference of the image feature value of the image feature value of this second image frame and this first image frame is during more than or equal to this threshold value, then represent the scene difference of this second image frame and this first image frame excessive, for instance if the image feature value is an image frame when the brightness integrated value of each space coordinates point, then represent the luminance difference of this second image frame and this first image frame excessive, might be to be converted to scene in this outdoor second image frame in this indoor first image frame by scene, or by daytime scene this first image frame be converted to this second image frame of night-time scene etc.This moment just can be in respect to this very first time point T1 of this first image frame and between with respect to this second time point T2 of this second image frame this image data being carried out segmentation, meaning is about to this first image frame segmentation of this very first time point T1 of time point early in last section (this first section), and with this second image frame segmentation of this second time point T2 of later time point in back one section (this second section).
If the difference of image feature value of judging the image feature value of this second image frame and this first image frame is not more than or equal to this threshold value, then represent the scene difference of this second image frame and this first image frame not excessive, can inspect second segmentation condition this moment again, meaning judges that promptly whether the difference of this second time point T2 and this initial time point T0 is more than or equal to this scheduled time, if the difference of this second time point T2 and this initial time point T0 is more than or equal to this scheduled time, then represent the period between this second time point T2 and this initial time point T0 long, also can carry out segmentation to this image data this moment between this very first time point T1 and this second time point T2; If the difference of this second time point T2 and this initial time point T0 is not more than or equal to this scheduled time, then represent the period between this second time point T2 and this initial time point T0 not long, just do not carry out segmentation to this image data this moment, and meaning is about to this first image frame and this second image frame is divided in same section.After between this some T1 and this second time point T2, this image data being carried out segmentation very first time, whether the time span of judging next image data fragment afterwards is more than or equal to this scheduled time, be according to time point and the difference of this second time point T2 and the comparative result of this scheduled time after this second time point T2, no longer describe in detail in this.
The judgement of the segmentation condition that changes by above-mentioned scene, can allow image data be segmented in the scene part that differs greatly, so can allow the user think that the scene consistency of each section is preferable, do not have this scene and do not finish as yet, and the part scene be cut to the discontinuity of next section; The judgement of the segmentation condition of the time restriction of arranging in pairs or groups again can be avoided a section overlong time, and causes the inconvenience of data search.As long as meaning promptly reaches the action that a wherein condition of above-mentioned two image data segmentation conditions (scene variation, time restriction) promptly can be carried out the image data segmentation, and what person is the judgement priority of above-mentioned two image data segmentation conditions do not limit to is after what person of elder generation is, whether the time span that also can judge the image data fragment earlier judges between image frame whether carry out scene change more than or equal to this scheduled time again.
See also Fig. 3, Fig. 3 changes the schematic diagram that carries out segmentation with the segmentation condition of time restriction for the present invention's one image data according to scene, by finding out among Fig. 3, the total length of time of this image data is 20 minutes, promptly carry out the image segmentation and in judging this image data, there is scene to change, and under the condition that each section time span is no more than 5 minutes, if started at the 3rd minute by initial time T0, the 11st minute, and the 18th minute be scene variation place, then this image data can be divided into six sections, first section is T0 to the 3 minutes, second section is the 3rd minute to the 8th minute, and the 3rd section is the 8th minute to the 11st minute, and the 4th section is the 11st minute to the 16th minute, the 5th section is the 16th minute to the 18th minute, and the 6th section is the 18th minute to the 20th minute.Just can take into account the restriction of the scene consistency and the section time span of each section thus.
Compared to previous image data segmentation method, spirit of the present invention is carried out segmentation for the scene according to image frame changes to image data, can consider the scene consistency of each section thus, and then improving Same Scene by the incoherence of segmentation to different sections, it is the image data segmented mode with hommization.
The above only is preferred embodiment of the present invention, and all equalizations of doing according to claim of the present invention change and modify, and all should belong to the covering scope of patent of the present invention.
Claims (11)
1. method that changes to carry out the image data segmentation according to scene, it includes the following step:
(a) one first image feature value of one first image frame of calculating one image data;
(b) calculate one second image feature value of one second image frame of this image data;
(c) whether compare a difference of this first image feature value and this second image feature value more than or equal to a threshold value; And
(d) if this difference more than or equal to this threshold value, this image data of segmentation is so that this first image frame and this second image frame belong to one first section and one second section respectively.
2. the method for claim 1, it also comprises:
(e) if this difference less than this threshold value, is kept this first image frame and this second image frame belongs to same section.
3. the method for claim 1, it also comprises every a specific period this image data is carried out segmentation.
4. the method for claim 1, wherein the image feature value of each image frame is the brightness integrated value of this image frame in each space coordinates point.
5. method that changes to carry out the image data segmentation according to scene, it includes the following step:
(a) judge whether exist a scene to change between one first image frame of an image data and one second image frame; And
(b) if this scene changes to be existed, this image data of segmentation is so that this first image frame and this second image frame belong to one first section and one second section respectively.
6. method as claimed in claim 5, it also comprises:
Do not exist if this scene changes, keep this first image frame and this second image frame belongs to same section.
7. method as claimed in claim 5, wherein step (a) includes:
(a1) calculate one first image feature value of one first image frame of this image data;
(a2) calculate one second image feature value of one second image frame of this image data;
(a3) compare a difference of this first image feature value and this second image feature value not more than or equal to a threshold value; And
(a4) if more than or equal to this threshold value, judging, this difference exist this scene to change between this first image frame and this second image frame.
8. method as claimed in claim 7, it also comprises:
(a5) if this difference, judges that not having this scene between this first image frame and this second image frame changes less than this threshold value.
9. method as claimed in claim 8, it also comprises:
Do not exist if this scene changes, keep this first image frame and this second image frame belongs to same section.
10. method as claimed in claim 5, wherein this first image feature value is the brightness integrated value of this first image frame in each space coordinates point, and this second image feature value is the brightness integrated value of this second image frame in each space coordinates point.
11. method as claimed in claim 5, it also comprises every a specific period this image data is carried out segmentation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2006101492649A CN101193248A (en) | 2006-11-21 | 2006-11-21 | Method for sectioning image data according to scenario change |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2006101492649A CN101193248A (en) | 2006-11-21 | 2006-11-21 | Method for sectioning image data according to scenario change |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101193248A true CN101193248A (en) | 2008-06-04 |
Family
ID=39487958
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2006101492649A Pending CN101193248A (en) | 2006-11-21 | 2006-11-21 | Method for sectioning image data according to scenario change |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101193248A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103596016A (en) * | 2013-11-20 | 2014-02-19 | 韩巍 | Multimedia video data processing method and device |
| CN105631422A (en) * | 2015-12-28 | 2016-06-01 | 北京酷云互动科技有限公司 | Video identification method and video identification system |
| CN106686452A (en) * | 2016-12-29 | 2017-05-17 | 北京奇艺世纪科技有限公司 | Dynamic picture generation method and device |
| CN107005676A (en) * | 2014-12-15 | 2017-08-01 | 索尼公司 | Information processing method, image processor and program |
| CN112383686A (en) * | 2020-11-02 | 2021-02-19 | 浙江大华技术股份有限公司 | Video processing method, video processing device, storage medium and electronic device |
| CN113347471A (en) * | 2021-06-01 | 2021-09-03 | 咪咕文化科技有限公司 | Video playing method, device, equipment and storage medium |
| CN115100739A (en) * | 2022-06-09 | 2022-09-23 | 厦门国际银行股份有限公司 | Man-machine behavior detection method, system, terminal device and storage medium |
-
2006
- 2006-11-21 CN CNA2006101492649A patent/CN101193248A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103596016A (en) * | 2013-11-20 | 2014-02-19 | 韩巍 | Multimedia video data processing method and device |
| CN103596016B (en) * | 2013-11-20 | 2018-04-13 | 韩巍 | A kind of multimedia video data treating method and apparatus |
| CN107005676A (en) * | 2014-12-15 | 2017-08-01 | 索尼公司 | Information processing method, image processor and program |
| CN105631422A (en) * | 2015-12-28 | 2016-06-01 | 北京酷云互动科技有限公司 | Video identification method and video identification system |
| CN106686452A (en) * | 2016-12-29 | 2017-05-17 | 北京奇艺世纪科技有限公司 | Dynamic picture generation method and device |
| CN106686452B (en) * | 2016-12-29 | 2020-03-27 | 北京奇艺世纪科技有限公司 | Method and device for generating dynamic picture |
| CN112383686A (en) * | 2020-11-02 | 2021-02-19 | 浙江大华技术股份有限公司 | Video processing method, video processing device, storage medium and electronic device |
| CN113347471A (en) * | 2021-06-01 | 2021-09-03 | 咪咕文化科技有限公司 | Video playing method, device, equipment and storage medium |
| CN113347471B (en) * | 2021-06-01 | 2023-05-02 | 咪咕文化科技有限公司 | Video playing method, device, equipment and storage medium |
| CN115100739A (en) * | 2022-06-09 | 2022-09-23 | 厦门国际银行股份有限公司 | Man-machine behavior detection method, system, terminal device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101193248A (en) | Method for sectioning image data according to scenario change | |
| CN108028054B (en) | Synchronizing audio and video components of an automatically generated audio/video presentation | |
| CN104809198B (en) | Method, device and the terminal unit of picture cleaning | |
| US8421819B2 (en) | Pillarboxing correction | |
| AU2008264197B2 (en) | Image selection method | |
| CN104317952B (en) | The scan method and device of memory space in mobile terminal | |
| KR101373020B1 (en) | The method and system for generating animated art effects on static images | |
| CN103475789B (en) | A kind of mobile terminal and control method thereof | |
| US8006201B2 (en) | Method and system for generating thumbnails for video files | |
| US20100315522A1 (en) | Image processing apparatus, dynamic picture reproduction apparatus, and processing method and program for the same | |
| CN105898520A (en) | Video frame interception method and device | |
| CN105657272A (en) | Terminal equipment and photographing method thereof | |
| CN106101771A (en) | Method for processing video frequency, device and terminal | |
| CN106982368A (en) | video response speed detection method and system | |
| CN105549847B (en) | A kind of image display method and user terminal at playback of songs interface | |
| CN104079926A (en) | Video performance testing method for remote desktop software | |
| JP2000358215A (en) | Data string temporary erasing method, its retrieval method, its restoring method, its permanent erasing method and recording medium for storing additional information for restoring data string separated by temporary erasure | |
| CN106507200A (en) | Video-frequency playing content insertion method and system | |
| EP3105758A1 (en) | Browsing videos via a segment list | |
| JP2002223412A (en) | Image processor, image processing system, image processing method, storage medium, and program software | |
| CN109800675A (en) | A kind of method and device of the identification image of determining face object | |
| CN106815283A (en) | Data processing method, device and electronic equipment | |
| CN106454509B (en) | A kind of detection method and device of advertising image | |
| US11126856B2 (en) | Contextualized video segment selection for video-filled text | |
| KR100650665B1 (en) | A method for filtering video data |
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 |
Open date: 20080604 |