CN100496113C - Film characteristic-image capturing system and method - Google Patents
Film characteristic-image capturing system and method Download PDFInfo
- Publication number
- CN100496113C CN100496113C CNB2004100015963A CN200410001596A CN100496113C CN 100496113 C CN100496113 C CN 100496113C CN B2004100015963 A CNB2004100015963 A CN B2004100015963A CN 200410001596 A CN200410001596 A CN 200410001596A CN 100496113 C CN100496113 C CN 100496113C
- Authority
- CN
- China
- Prior art keywords
- picture
- image
- index
- pictures
- image data
- 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.)
- Expired - Fee Related
Links
Images
Abstract
The invention provides a characteristic video picking system and method for a film, where the method cuts digital video data using the detection for content change of adjacent video pictures, and cuts them as a very large change of the content of video picture suddenly happens, and outputs the video picture in this place as index picture to express the content of this segment of plot or scene, making the digital video data contained by the film able to be automatically cut according to the content, and simultaneously provides index picture set to let a user be able to rapidly browse digital video data; the system includes: a video decoding module, a preprocessing module, a video cutting module and a database. The system and method can let the user simply and effectively search the needed segments in the video data and obtain the paragraph information, thus more conveniently and effectively using video data.
Description
Technical field
The present invention relates to a kind of image acquisition system and method thereof, particularly a kind of characteristic image acquisition system and method thereof of film.
Background technology
Along with the arriving of digital Age, the digitized forms that information miscellaneous such as sound, image can be more efficient, convenient stores, scatters and circulation, but the problem that also causes a large amount of audio and video information accumulations to spread unchecked.It is more and more higher to the quality requirements of image and audio to add the user, and therefore the storage area demand of single video/audio archive also increases thereupon.Yet in one whole section digital image data of prerecording, perhaps the user is only interesting to a bit of or number piece of digital image data wherein, but existing systems needs whole section content play in chronological order, make the user to spend to put in order the piece of digital image data for a long time and finish watching, just known whether interested content, lacked function for user's fast browsing digital image data.
Therefore, there are some systems that the information segment that digital image data is divided into set time length according to the time is handled or preserved, rather than carry out cutting according to the plot of digital image data.This cutting result is that each segment information might not comprise one section complete plot, and the user has no way of also learning which segment information is its interested content be positioned at.Or for the convenience of the users, video-signal system is made the content annotations and comments for each segment information, with manual type a large amount of substance films is done annotations and comments fully and is dependent on very big manpower.Again because each segment information might not comprise one section complete plot, thus be difficult to fix objective and detailed text description at each paragraph in the film, and cause traditionally in the mode of text query, can't satisfy the demand that digital image data is searched.
Therefore, make system that effective index can be provided, allow the user can easily obtain required information, and the information that provided of the whole piece of digital image data of fast browsing, digital image data is utilized more effectively.
Summary of the invention
For improving the shortcoming of known technology, the present invention proposes a kind of characteristic image acquisition system and method thereof of film, main purpose is that the digital image data that film is comprised can carry out cutting according to content automatically, provide index picture to make the user can carry out the digital image data fast browsing simultaneously, carried out cutting with the solution past based on the time and do not meet user's demand and browse the time-consuming very long shortcoming of video signal.
For achieving the above object, the characteristic image acquisition system and the method thereof of film of the present invention, utilization comes the cutting digital image data to the detection of adjacent image picture content change, when the image image content produces great changes suddenly, just think that variation has taken place for the plot or the scene of digital image data, carries out cutting in this place; The image picture of exporting this place simultaneously as index picture to represent the content of this section plot or scene.According to the result of cutting, digital image data can carry out segment processing and storage by its content, and index picture then can carry out fast browsing for the user.
In order to inquire the particular film paragraph at digital image data, we need earlier digital image data to be deciphered, and take out its all pixel values at each picture that splits again, and further compare.Characteristic image DAQ method provided by the present invention, be used to digital image data and select a plurality of index pictures that comprise correspondence position, and export database to form the index picture collection, include the following step: (a) digital image data is deciphered forming a plurality of image pictures, and obtained the original pixel value of a plurality of image pictures; (b) again the image picture is carried out preliminary treatment and eliminate noise; (c) set a threshold limit value; (d) take out an image picture in regular turn as differentiating picture and an adjacent back image picture thereof as the reference picture; (e) original pixel value that will differentiate picture and reference picture is converted to the GTG pixel value; (f) the accumulated value of pixel difference that confirm to differentiate picture and all correspondence positions of reference picture is greater than threshold limit value, and as this accumulated value execution in step g during greater than this threshold limit value, then skips steps g re-executes steps d when this accumulated value is not more than this threshold limit value; (g) picture and position to index picture collection thereof are differentiated in output; Repeat above-mentioned steps (d) at last to (g), until the identification of finishing N image picture.
In the above-mentioned characteristic image DAQ method, also can comprise the step of an indexing information of corresponding this index picture of generation.
In the above-mentioned characteristic image DAQ method, this correspondence film position can be the cutting time location of this differentiation picture and image picture order one of them.
Characteristic image DAQ provided by the present invention system, be used to digital image data and select the index picture that comprises correspondence position, and exporting database to form the index picture collection, it includes: image decoding module, pretreatment module, image cutting module and database.Wherein, the image decoding module is in order to deciphering for digital image data forming a plurality of image pictures, and obtains the original pixel value of a plurality of image pictures; Pretreatment module then is used for carrying out preliminary treatment with unnecessary influences such as removal noises for a plurality of image pictures; Image cutting module is by the GTG pixel value of more adjacent image picture, detecting the image picture that can be used as index, and exports database to composition index picture collection; Database is in order to storage index picture collection.
In the above-mentioned characteristic image DAQ system, this image cutting module can detect this index picture, simultaneously corresponding generation one indexing information.
Can be divided into a lot of fragments to digital image data according to its content plot by above-mentioned method with complete plot or scene, allow the user can find required fragment image easily, and from each section, select the content that a typical index picture is represented this section plot or scene, be available for users to the main contents of the whole digital image data of fast browsing, this digital image data is done fast and effective ratio is right.
For making purpose of the present invention, structural feature and function thereof there are further understanding, cooperate icon to be described in detail as follows now.
Description of drawings
Fig. 1 is the method flow diagram of the embodiment of the invention; And
Fig. 2 is the characteristic image DAQ system schematic of the embodiment of the invention.
Wherein, description of reference numerals is as follows:
Step 110: digital image data is deciphered forming the 1st to N image picture, and obtained the 1st original pixel value to N image picture
Step 120: the image picture is carried out preliminary treatment eliminate noise and unnecessary influence
Step 130: set a threshold limit value
Step 140: take out M image picture as differentiating picture, M+1 image picture is as the reference picture
Step 150: the original pixel value that will differentiate picture and reference picture is converted to the GTG pixel value
Step 160: the accumulated value of the GTG pixel difference of all correspondence positions of computational discrimination picture and reference picture
Step 170: an accumulated value of the pixel difference of affirmation differentiation picture and all correspondence positions of reference picture is greater than threshold limit value
Step 180: export M image picture and position thereof as index picture, and transfer to an index picture collection
Step 190: the M value is added 1 and confirm M+1≤N
200: digital image data
210: the image decoding module
220: pretreatment module
230: image cutting module
240: database
Embodiment
Please refer to Fig. 1, it is the method flow diagram of the embodiment of the invention.Include the following step: at first, digital image data is deciphered forming the 1st to N image picture, and obtained the 1st original pixel value (step 110) to N image picture; Again the image picture is carried out preliminary treatment and eliminate noise and unnecessary influence (step 120); Set a threshold limit value (step 130); Take out M image picture as differentiating picture, M+1 image picture is as reference picture (step 140); The original pixel value of differentiating picture and reference picture is converted to GTG pixel value (step 150); The accumulated value (step 160) of the GTG pixel difference of all correspondence positions of computational discrimination picture and reference picture; One accumulated value of the pixel difference of affirmation differentiation picture and all correspondence positions of reference picture is greater than threshold limit value (step 170); Export M image picture and position thereof as index picture, and transfer to an index picture collection (step 180); The M value is added 1 and confirm M+1≤N (step 190).
Wherein, an accumulated value of pixel difference of confirming this differentiation picture and these all correspondence positions of reference picture is greater than the step of this threshold limit value (step 170), and when being "No" as the result, then skips steps 180 is to step 190.And the M value is added 1 and confirm the step of M+1≤N (step 190), and when being "Yes", then get back to step 140 as the result, proceed the identification of other image picture, be "No" as the result, then finish.After all image pictures all passed through cutting and finish, the image picture of all outputs was promptly formed the index picture collection, and each image picture is all represented the episode or the scene of image data.Wherein, corresponding film position can be the cutting time location or the image picture order of the index picture selected.The image picture of all outputs is arranged the main contents that just can represent whole image data in chronological order, but the user is as long as concentrate the just quicklook ground understanding image data content of browsing of image picture by index picture.In addition, the present invention also comprises the step of an indexing information of corresponding this index picture of generation, and index data can comprise numbering, start picture position, end picture position, time span, representative picture position and comment Text or the like.
Please refer to Fig. 2, it is the characteristic image DAQ system schematic of the embodiment of the invention, be used to digital image data 200 and select the index picture that comprises correspondence position, and exporting database 240 to form the index picture collection, it includes: image decoding module 210, pretreatment module 220, image cutting module 230 and database 240.
When carrying out method of the present invention, digital image data 200 is input into image decoding module 210, and 210 of image decoding modules are deciphered forming a plurality of image pictures for digital image data 200, and obtain the original pixel value of a plurality of image pictures; A plurality of image pictures carry out preliminary treatment with unnecessary influences such as removal noises by pretreatment module 220 again, and transfer the original pixel value of a plurality of image pictures to the GTG pixel value; Then, utilize the GTG value differences of the more adjacent image picture of image cutting module 250, detecting image picture and the corresponding film position thereof that can be used as index, and export image picture and position thereof to database and form the index picture collection.And image cutting module 230 produces corresponding indexing information when detecting this index picture.Wherein, the source of digital image data can be from Storage Media or Network Transmission, and corresponding film position can be the cutting time location or the image picture order of the index picture of selecting.
Because the present invention lists each cutting paragraph of image data in regular turn with a representative picture, the person of being easy to use carries out fast browsing.The user who particularly desires the construction image database can do processing such as artificial annotations and comments at faster speed to important cutting paragraph.The indexing information of each cutting paragraph in the image data (comprising numbering, start picture position, end picture position, time span, representative picture position and comment Text or the like), also can produce by the form of Data Base Pile, and export the index picture collection to, wherein only there is comment Text to add, other can produce by Automatic Program through manual type.By this index picture collection, the user not only can simply and effectively compare in image data and hunt out required fragment, can obtain many cutting paragraph information simultaneously, and image data can more conveniently be used by more people effectively.
Though preferred embodiment of the present invention discloses as mentioned above; yet be not in order to limit the present invention; any those skilled in the art; without departing from the spirit and scope of the present invention; when can doing a little change and retouching, therefore scope of patent protection of the present invention must be looked defining of the appended claim of this specification and is as the criterion.
Claims (5)
1. a characteristic image DAQ method is selected a plurality of index pictures that comprise correspondence position by the identification of image feature in a digital image data, and wherein this method comprises the following step:
A. this digital image data is deciphered forming a plurality of image pictures, and obtained the respectively original pixel value of this image picture;
B. this image picture is respectively carried out preliminary treatment and eliminate noise;
C. set a threshold limit value, with the restriction of allowing as image feature difference;
D. take out this image picture in regular turn and differentiate picture and adjacent time this image picture thereof as a reference picture as one;
E. the original pixel value that will differentiate picture and this reference picture is converted to the GTG pixel value;
F. an accumulated value of pixel difference of confirming this differentiation picture and these all correspondence positions of reference picture is greater than this threshold limit value, as this accumulated value execution in step g during greater than this threshold limit value, then skips steps g re-executes steps d when this accumulated value is not more than this threshold limit value; And
G. export this a differentiation picture and a corresponding film position as this index picture and be stored to an index picture collection;
Wherein, this method will continue the step of repeating step d to step g, up to finishing the respectively differentiation of this image picture.
2. characteristic image DAQ method as claimed in claim 1, wherein this method also comprises the step of an indexing information of corresponding this index picture of generation.
3. characteristic image DAQ method as claimed in claim 1, wherein should correspondence film position be the cutting time location of this differentiation picture and image picture order one of them.
4. a characteristic image DAQ system selects a plurality of index pictures that comprise correspondence position by the identification of image feature in a digital image data, and wherein this system comprises following modules:
One image decoding module in order to this digital image data being deciphered forming a plurality of image pictures, and obtains the original pixel value of a plurality of image pictures;
One pretreatment module is used for carrying out preliminary treatment with the removal noise for a plurality of image pictures, and transfers the original pixel value of this image picture to the GTG pixel value;
One image cutting module by the GTG value differences of this more adjacent image picture, detecting this index picture, and exports a database to form an index picture collection; And
One database is in order to store this index picture collection.
5. characteristic image DAQ as claimed in claim 4 system, wherein this image cutting module detects this index picture, simultaneously corresponding generation one indexing information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100015963A CN100496113C (en) | 2004-01-14 | 2004-01-14 | Film characteristic-image capturing system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100015963A CN100496113C (en) | 2004-01-14 | 2004-01-14 | Film characteristic-image capturing system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1642256A CN1642256A (en) | 2005-07-20 |
| CN100496113C true CN100496113C (en) | 2009-06-03 |
Family
ID=34867146
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2004100015963A Expired - Fee Related CN100496113C (en) | 2004-01-14 | 2004-01-14 | Film characteristic-image capturing system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100496113C (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101394558B (en) * | 2007-09-20 | 2011-05-11 | 奇扬网科股份有限公司 | Method for detecting and playing imagery zone of computer picture |
| US10218911B2 (en) * | 2017-03-22 | 2019-02-26 | Htc Corporation | Mobile device, operating method of mobile device, and non-transitory computer readable storage medium |
-
2004
- 2004-01-14 CN CNB2004100015963A patent/CN100496113C/en not_active Expired - Fee Related
Non-Patent Citations (2)
| Title |
|---|
| 视频分层组织方案和技术. 章毓晋，陆海斌.中国工程科学，第2卷第3期. 2000 |
| 视频分层组织方案和技术. 章毓晋，陆海斌.中国工程科学，第2卷第3期. 2000 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1642256A (en) | 2005-07-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8503523B2 (en) | Forming a representation of a video item and use thereof | |
| CN111866585B (en) | Video processing method and device | |
| CN101616264B (en) | Method and system for cataloging news video | |
| AU2017204338B2 (en) | Industry first method that uses users search query to show highly relevant poster frames for stock videos thereby resulting in great experience and more sales among users of stock video service | |
| CN108769731B (en) | Method and device for detecting target video clip in video and electronic equipment | |
| US7876381B2 (en) | Telop collecting apparatus and telop collecting method | |
| CN101872346B (en) | Method for generating video navigation system automatically | |
| CN101821734B (en) | Detection and classification of matches between time-based media | |
| CN109684513B (en) | Low-quality video identification method and device | |
| US20040170321A1 (en) | Method and system for segmentation, classification, and summarization of video images | |
| CN103530652A (en) | Face clustering based video categorization method and retrieval method as well as systems thereof | |
| KR101404585B1 (en) | Segment creation device, segment creation method, and computer-readable recording medium having a segment creation program | |
| CN106407358A (en) | Image searching method and device and mobile terminal | |
| CN101499085B (en) | Method and apparatus for fast extracting key frame | |
| KR100644016B1 (en) | Moving picture search system and method thereof | |
| CN100496113C (en) | Film characteristic-image capturing system and method | |
| Satoh | News video analysis based on identical shot detection | |
| CN101770474A (en) | History searching record-based searching method and device | |
| Bartolini et al. | Shiatsu: semantic-based hierarchical automatic tagging of videos by segmentation using cuts | |
| CN100496112C (en) | Multi-gradation film browsing system and method therefor | |
| JP4270118B2 (en) | Semantic label assigning method, apparatus and program for video scene | |
| CN114385859A (en) | Multi-modal retrieval method for video content | |
| TWI261192B (en) | Multi-level film browsing system and method thereof | |
| Nabi et al. | Detection and extraction of videos using decision trees | |
| CN102479194A (en) | WEB information processing method and device |
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 | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090603 Termination date: 20110114 |