CN105159923A - Video image based article extraction, query and purchasing method - Google Patents
Video image based article extraction, query and purchasing method Download PDFInfo
- Publication number
- CN105159923A CN105159923A CN201510470933.1A CN201510470933A CN105159923A CN 105159923 A CN105159923 A CN 105159923A CN 201510470933 A CN201510470933 A CN 201510470933A CN 105159923 A CN105159923 A CN 105159923A
- Authority
- CN
- China
- Prior art keywords
- article
- video
- frame
- picture
- library file
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/73—Querying
- G06F16/735—Filtering based on additional data, e.g. user or group profiles
Abstract
The invention provides a video image based article extraction, query and purchasing method. The method comprises the following steps of: decomposing video works; dividing each frame of image; obtaining a library file containing a frame number and a region number; obtaining a URL library file of an article appearing in a video; and establishing a corresponding relation between the library file containing the frame number and the region number and the URL library file of the article appearing in the video. The method has the beneficial effects that with the method, a user watching the video can be helped to perform operations of query, collection, reservation, interaction, purchasing and the like on corresponding articles, such as but not limited to clothes, cosmetics, hair styles, places, characters and the like, appearing in the video through clicking of the user; at the same time, a video producer can be helped to insert various advertisements of articles, scenes and the like without causing audience disgust; and in addition, big data analysis is performed by utilizing the articles, the characters, the scenes and the like clicked or slid by the user, thereby providing feedback data for video producers or commodity providers.
Description
Technical field
The present invention relates to generation information technical field, more specifically to a kind of method extracting article, inquiry article and purchase article based on video pictures, user can directly by clicking the article occurred in video, such as clothes, commodity, scene, the relevant information of personage etc., and come such as to inquire about by clicking corresponding article, collection, predetermined, interactive, the operations such as purchase.
Background technology
Existing technology such as adopt the method for template matches constantly Pan and Zoom template multiple image significant points is aimed at, utilize these frame video image lap information, comprehensively go out width target object image clearly.
The method of template matches: template is exactly a known little image.Template matches is exactly searching target in the large image of a width, have the target that will look for, and this target has identical size, direction and image with template, can find target in the drawings by certain algorithm, determine its coordinate position in this figure known.
As shown in Figure 1, for 8 bit images (its 1 pixel is described by 1 byte), template T (m*n pixel) overlays the upper translation of searched figure S (W × H pixel), and template covers that block region whistle figure Sij of searched figure.I, j are the coordinate of the subgraph upper left corner on searched figure S.Hunting zone is:
1≤i≤W–M
1≤j≤H–N
By comparing the similarity of T and Sij, complete template matches process.
The major defect of template matching method is the Measures compare complexity of information extraction, and accuracy is not high, and hardware spending is comparatively large, cannot realize the object after to identification and inquire about, collection and the function bought.
Summary of the invention
Instant invention overcomes deficiency of the prior art, provide a kind of method extracting article, inquiry article and purchase article based on video pictures, solving prior art cannot carry out information extraction to dynamic menu, or it is low to extract accuracy, the problems such as extraction scheme is complicated, the information of article in the video that simultaneously can utilize extraction, such as clothes, commodity, scene, personages etc., carry out the inquiry of respective articles by the click of user, collection, the operations such as predetermined and purchase.
Object of the present invention is achieved by following technical proposals.
Extract article, inquiry article based on video pictures and buy the method for article, comprising the steps:
Step 1, the decomposition of video work: continuous print video pictures is resolved into independent static frame shape picture, any video is all made up of the image frame of static state, and every frame has independent numbering information, and different frame pictures is numbered;
Step 2, splits every frame picture: carry out coordinate numbering with the pixel of image frame to every frame picture, and then carries out picture segmentation, to obtain picture subregion to the every frame picture finishing coordinate numbering;
Step 3, comprise the library file of frame number and zone number: by having marked every frame picture that coordinate is numbered in step 2, the edge coordinate of picture subregion is marked one by one, edge coordinate in picture subregion whole in same frame picture is gathered, obtains the library file comprising frame number and zone number;
Step 4, the URL library file of video article: the whole URL address corresponding to each article occurred in video is gathered, to obtain the URL address base of same article, and then the URL address base of article all in video is gathered, to form the URL library file of video article;
Step 5, the URL library file comprising frame number and zone number and video article sets up corresponding relation, to reach the object extracted article, inquiry article and buy article.
In described step 2, the shape of picture segmentation adopts rectangle, circle, oval, the mode of polygon or irregular figure.
In described step 5, corresponding relation can be one_to_one corresponding, one-to-many is corresponding, many-one is corresponding or multi-to-multi corresponding in one.
In described step 5, corresponding method can adopt manual method or fuzzy diagnosis and the method manually combined.
Such as be not limited to mobile phone, PAD, computer when user utilizes, when the hardware device such as TV plays video work, click or draw when getting video pictures, user can be obtained and click or draw the position coordinates getting region, and calculate the frame number at it.By the regional location coordinate of gained and frame number, find the corresponding library file comprising frame number and zone number, according to this library file and the corresponding relation of URL library file comprising Item Information, by user provide from click or stroke to get the relevant information of article.
Beneficial effect of the present invention is: utilize the present invention that video-see user can be helped to realize the article occurred in screen, be such as not limited to clothing, cosmetics, hair style, place, personages etc., are inquired about corresponding article by the click of user, collection, predetermined, interactive, the operations such as purchase.Meanwhile, video production side can be helped to insert various article, the advertisements such as scene, and dislike can not be brought to spectators.In addition, utilize the click of user or draw the article got, personage, scene etc. carry out large data analysis, for video production side or goods providers provide feedback data.
Accompanying drawing explanation
Fig. 1 is the schematic diagram of existing template matching method;
Fig. 2 is schematic flow sheet of the present invention.
In figure: 1 is video work, 2 is frame process, and 3 is the segmentation of every frame picture, and 4 is frame page number, and 5 is user operation, and 6 is feedback user operation coordinate, and 7 is information bank file, and 8 is merchandise news and purchase link, and 9 is scene information, and 10 is that personage is interactive.
Embodiment
Below by specific embodiment, technical scheme of the present invention is further described.
As shown in Figure 2, wherein, 1 is video work, and 2 is frame process, 3 is the segmentation of every frame picture, and 4 is frame page number, and 5 is user operation, and 6 is feedback user operation coordinate, 7 is information bank file, and 8 is merchandise news and purchase link, and 9 is scene information, and 10 is that personage is interactive.
Extract article, inquiry article based on video pictures and buy the method for article, comprising the steps:
Step 1, the decomposition of video work: continuous print video pictures is resolved into independent static frame shape picture, any video is all made up of the image frame of static state, and every frame has independent numbering information, and different frame pictures is numbered;
Step 2, splits every frame picture: carry out coordinate numbering with the pixel of image frame to every frame picture, and then carries out picture segmentation, to obtain picture subregion to the every frame picture finishing coordinate numbering;
Step 3, comprise the library file of frame number and zone number: by having marked every frame picture that coordinate is numbered in step 2, the edge coordinate of picture subregion is marked one by one, edge coordinate in picture subregion whole in same frame picture is gathered, obtains the library file comprising frame number and zone number;
Step 4, the URL library file of video article: the whole URL address corresponding to each article occurred in video is gathered, to obtain the URL address base of same article, and then the URL address base of article all in video is gathered, to form the URL library file of video article;
Step 5, the URL library file comprising frame number and zone number and video article sets up corresponding relation, to reach the object extracted article, inquiry article and buy article.
In step 2, the shape of picture segmentation adopts rectangle, circle, oval, the mode of polygon or irregular figure.
In steps of 5, corresponding relation can be one_to_one corresponding, one-to-many is corresponding, many-one is corresponding or multi-to-multi corresponding in one.
In steps of 5, corresponding method can adopt manual method or fuzzy diagnosis and the method manually combined.
First video work is decomposed, continuous print video pictures is resolved into independent static frame shape picture, and different frame pictures is numbered.Secondly be divided into some regions to by the picture of every frame, and this region is numbered, and form the library file comprising frame number and zone number.To the article occurred in video, be such as not limited to clothing, cosmetics, hair style, place, the URL of the relevant informations such as personage sets up unified library file.Frame number and zone number will be comprised and comprise article, being such as not limited to clothing, cosmetics, hair style, place, the URL of the relevant informations such as personage sets up unified library file and sets up corresponding relation, its relation can be one_to_one corresponding, and one-to-many is corresponding, and many-one is corresponding or multi-to-multi is corresponding.Such as be not limited to mobile phone, PAD, computer when user utilizes, when the hardware device such as TV plays video work, click or draw when getting video pictures, user can be obtained and click or draw the position coordinates getting region, and calculate the frame number at it.By the regional location coordinate of gained and frame number, find the corresponding library file comprising frame number and zone number, according to this library file and the corresponding relation of URL library file comprising Item Information, by user provide from click or stroke to get the relevant information of article.
Adopt the software of the method including, without being limited to apple IOS, in the operating systems such as Android operation system, utilize mobile phone, panel computer, notebook, desktop computer, the hardware devices such as televisor are play.
Such as be not limited to mobile phone, PAD, computer when user utilizes, when the hardware device such as TV plays video work, click or draw when getting video pictures, user can be obtained and click or draw the position coordinates getting region, and calculate the frame number at it.By the regional location coordinate of gained and frame number, find the corresponding library file comprising frame number and zone number, according to this library file and the corresponding relation of URL library file comprising Item Information, by user provide from click or stroke to get the relevant information of article.
Above to invention has been exemplary description; should be noted that; when not departing from core of the present invention, any simple distortion, amendment or other those skilled in the art can not spend the equivalent replacement of creative work all to fall into protection scope of the present invention.
Claims (4)
1. extract article, inquiry article based on video pictures and buy the method for article, it is characterized in that: comprise the steps:
Step 1, the decomposition of video work: continuous print video pictures is resolved into independent static frame shape picture, any video is all made up of the image frame of static state, and every frame has independent numbering information, and different frame pictures is numbered;
Step 2, splits every frame picture: carry out coordinate numbering with the pixel of image frame to every frame picture, and then carries out picture segmentation, to obtain picture subregion to the every frame picture finishing coordinate numbering;
Step 3, comprise the library file of frame number and zone number: by having marked every frame picture that coordinate is numbered in step 2, the edge coordinate of picture subregion is marked one by one, edge coordinate in picture subregion whole in same frame picture is gathered, obtains the library file comprising frame number and zone number;
Step 4, the URL library file of video article: the whole URL address corresponding to each article occurred in video is gathered, to obtain the URL address base of same article, and then the URL address base of article all in video is gathered, to form the URL library file of video article;
Step 5, the URL library file comprising frame number and zone number and video article sets up corresponding relation, to reach the object extracted article, inquiry article and buy article.
2. the method extracting article, inquiry article and purchase article based on video pictures according to claim 1, it is characterized in that: in described step 2, the shape of picture segmentation adopts rectangle, circle, oval, the mode of polygon or irregular figure.
3. the method extracting article, inquiry article and purchase article based on video pictures according to claim 1, it is characterized in that: in described step 5, corresponding relation can be one_to_one corresponding, one-to-many is corresponding, many-one is corresponding or multi-to-multi corresponding in one.
4. according to claim 1ly extract article, inquiry article based on video pictures and buy the method for article, it is characterized in that: in described step 5, corresponding method can adopt manual method or fuzzy diagnosis and the method manually combined.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510470933.1A CN105159923A (en) | 2015-08-04 | 2015-08-04 | Video image based article extraction, query and purchasing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510470933.1A CN105159923A (en) | 2015-08-04 | 2015-08-04 | Video image based article extraction, query and purchasing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105159923A true CN105159923A (en) | 2015-12-16 |
Family
ID=54800780
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510470933.1A Pending CN105159923A (en) | 2015-08-04 | 2015-08-04 | Video image based article extraction, query and purchasing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105159923A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107392697A (en) * | 2017-07-03 | 2017-11-24 | 李洪海 | order data processing method, system and server |
| CN108769808A (en) * | 2018-05-24 | 2018-11-06 | 安徽质在智能科技有限公司 | Interactive video playback method and system |
| CN108960005A (en) * | 2017-05-19 | 2018-12-07 | 内蒙古大学 | The foundation and display methods, system of subjects visual label in a kind of intelligent vision Internet of Things |
| CN109902195A (en) * | 2019-01-31 | 2019-06-18 | 深圳市丰巢科技有限公司 | Monitoring image querying method, device, equipment and medium |
| CN110543582A (en) * | 2019-08-05 | 2019-12-06 | 百度在线网络技术（北京）有限公司 | image-based query method and device |
| CN114187556A (en) * | 2021-12-14 | 2022-03-15 | 养哇(南京)科技有限公司 | High-definition video intelligent segmentation method based on picture features |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070261079A1 (en) * | 2000-04-12 | 2007-11-08 | Lg Electronics Inc. | Method and video device for accessing information |
| CN102509235A (en) * | 2011-10-31 | 2012-06-20 | 福州博远无线网络科技有限公司 | Purchase interactive method based on mobile video of intelligent handheld device |
| CN103020172A (en) * | 2012-11-28 | 2013-04-03 | 北京京东世纪贸易有限公司 | Method and device utilizing video information to search articles |
| CN103207862A (en) * | 2012-01-13 | 2013-07-17 | 阿尔卡特朗讯公司 | Method and device for injecting additional information in video |
| CN103443816A (en) * | 2011-03-17 | 2013-12-11 | 电子湾有限公司 | Video processing system for identifying items in video frames |
-
2015
- 2015-08-04 CN CN201510470933.1A patent/CN105159923A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070261079A1 (en) * | 2000-04-12 | 2007-11-08 | Lg Electronics Inc. | Method and video device for accessing information |
| CN103443816A (en) * | 2011-03-17 | 2013-12-11 | 电子湾有限公司 | Video processing system for identifying items in video frames |
| CN102509235A (en) * | 2011-10-31 | 2012-06-20 | 福州博远无线网络科技有限公司 | Purchase interactive method based on mobile video of intelligent handheld device |
| CN103207862A (en) * | 2012-01-13 | 2013-07-17 | 阿尔卡特朗讯公司 | Method and device for injecting additional information in video |
| CN103020172A (en) * | 2012-11-28 | 2013-04-03 | 北京京东世纪贸易有限公司 | Method and device utilizing video information to search articles |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108960005A (en) * | 2017-05-19 | 2018-12-07 | 内蒙古大学 | The foundation and display methods, system of subjects visual label in a kind of intelligent vision Internet of Things |
| CN107392697A (en) * | 2017-07-03 | 2017-11-24 | 李洪海 | order data processing method, system and server |
| CN108769808A (en) * | 2018-05-24 | 2018-11-06 | 安徽质在智能科技有限公司 | Interactive video playback method and system |
| CN108769808B (en) * | 2018-05-24 | 2021-01-29 | 浙报融媒体科技(浙江)有限责任公司 | Interactive video playing method and system |
| CN109902195A (en) * | 2019-01-31 | 2019-06-18 | 深圳市丰巢科技有限公司 | Monitoring image querying method, device, equipment and medium |
| CN110543582A (en) * | 2019-08-05 | 2019-12-06 | 百度在线网络技术（北京）有限公司 | image-based query method and device |
| CN114187556A (en) * | 2021-12-14 | 2022-03-15 | 养哇(南京)科技有限公司 | High-definition video intelligent segmentation method based on picture features |
| CN114187556B (en) * | 2021-12-14 | 2023-12-15 | 华策影视(北京)有限公司 | Intelligent high-definition video segmentation method based on picture characteristics |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105159923A (en) | Video image based article extraction, query and purchasing method | |
| US10134441B2 (en) | Method and system for overlaying image in video stream | |
| US10643264B2 (en) | Method and computer readable medium for presentation of content items synchronized with media display | |
| WO2018095142A1 (en) | Livestream interaction method and apparatus | |
| KR101901454B1 (en) | A real-time shopping method using image recognition in broadcasting and a smart phone equipped with an application for implementing it | |
| TWI559242B (en) | Visual clothing retrieval | |
| US8718369B1 (en) | Techniques for shape-based search of content | |
| CN106792153B (en) | Video identification processing method and device and computer readable storage medium | |
| CN104602128A (en) | Video processing method and device | |
| KR102024818B1 (en) | Method, apparatus and computer program for providing search information from video | |
| US11894021B2 (en) | Data processing method and system, storage medium, and computing device | |
| KR20180024200A (en) | Method, apparatus and computer program for providing search information from video | |
| CN110858134A (en) | Data, display processing method and device, electronic equipment and storage medium | |
| CN102833625A (en) | Device and method for dynamically embedding advertisement into video | |
| CN104268168A (en) | Method and device for pushing information to user | |
| US10091556B1 (en) | Relating items to objects detected in media | |
| CN104519371B (en) | Method for pushing, pusher and server | |
| US20170013309A1 (en) | System and method for product placement | |
| CN114040248A (en) | Video processing method and device and electronic equipment | |
| CN102073668B (en) | Searching and extracting digital images from digital video files | |
| US20150269189A1 (en) | Retrieval apparatus, retrieval method, and computer program product | |
| CN106156732A (en) | Object identifying method and object recognition equipment | |
| CN110765296A (en) | Image searching method, terminal device and storage medium | |
| CN110942056A (en) | Clothing key point positioning method and device, electronic equipment and medium | |
| CA2948298A1 (en) | Marker-based pixel replacement |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20151216 |