CN102609548A - Video content retrieval method and system based on moving objects - Google Patents
Video content retrieval method and system based on moving objects Download PDFInfo
- Publication number
- CN102609548A CN102609548A CN2012101159853A CN201210115985A CN102609548A CN 102609548 A CN102609548 A CN 102609548A CN 2012101159853 A CN2012101159853 A CN 2012101159853A CN 201210115985 A CN201210115985 A CN 201210115985A CN 102609548 A CN102609548 A CN 102609548A
- Authority
- CN
- China
- Prior art keywords
- moving target
- video
- retrieved
- image
- module
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a video content retrieval method based on moving objects. The method comprises the following steps of: analyzing and processing initial image sequences of all video files in a video library to obtain the positions of the moving objects, calculating the visual statistical characteristics of the moving objects, and storing the visual statistical characteristics in a video index database; inputting objects to be retrieved by a user, selecting moving objects to be retrieved from the objects, and calculating the visual statistical characteristics of the moving objects to be retrieved; looking up the video index record most similar to the moving objects to be retrieved; and selecting and displaying the corresponding video segment from the video library according to the retrieved video index record. The invention also discloses a video content retrieval system based on the moving objects. According to the invention, by detecting and tracking the moving objects in all video files in the video library,,automatic retrieving can be rapidly carried out on the video contents, thus the high efficiency of video content retrieval is improved, and the adaptability of the video retrieval system is improved.
Description
Technical field
The present invention relates to the video content searching field, especially a kind of video content search method and system based on moving target.
Background technology
Along with the development of economy and infotech, people are also increasingly high to the demand of safety precaution, and video monitoring is as the effective means in safety precaution field, and range of application is more and more wider, and application demand is also constantly improving.
In daily application, produce a large amount of video files every day, how from these video files, seeking the user's interest content becomes a difficult problem.Adopt manual type to carry out the retrieval anomalies time and effort consuming, especially under the situation that number of videos sharply increases, this mode is more and more infeasible.Therefore, adopt the research of the automatic retrieve video content of infotech to become current research focus, because the indexing service of video content is still needed artificial the participation, equally too consuming time.The fundamental way that addresses this problem is to set up video content index automatically according to the characteristics of video content; In many instances; People are interested to be the moving target in the video, if can carry out automatic index to the moving target in the video, just can realize the quick retrieval of video content.At present, also there is not to occur to carry out the moving target in the video technology of automatic index.
Summary of the invention
Primary and foremost purpose of the present invention is to provide a kind of video content search method based on moving target that can detect, accuracy fast with index, retrieval rate and strong robustness to the moving target in the video automatically, and this method comprises the step of following order:
(1) obtains the initial pictures sequence of all video files in the video library; The initial pictures sequence is carried out analyzing and processing; Obtain the position of each moving target; Calculate the vision statistical nature of each moving target, and the position and the vision statistical nature information stores of each moving target arrived in the video index database;
(2) user imports object to be retrieved, and therefrom chooses moving target to be retrieved, and calculates the vision statistical nature of moving target to be retrieved;
(3) in the video index database, search the video index record the most similar with moving target to be retrieved;
(4), from video library, select corresponding video segment, and on user monitoring terminal, show according to the video index record that retrieves.
Another object of the present invention is to provide a kind of video content searching system, comprising based on moving target:
The video index module; From video library, obtain the image sequence of all video files; Utilize background subtraction that image sequence is carried out analyzing and processing, obtain the position and the image of each moving target, according to the vision statistical nature of each moving target of image calculation of moving target; Carry out motion target tracking in interframe, and the tracking results of position, vision statistical nature information and each moving target of each moving target is stored in the video index database;
The searching object load module, the user inputs to the searching object load module with object to be retrieved, and from object to be retrieved, chooses moving target to be retrieved;
The Visual Feature Retrieval Process module is calculated the vision statistical nature of moving target to be retrieved automatically;
The video frequency searching module is searched the video index record the most close with moving target to be retrieved in the video index database;
The result for retrieval output module according to the video index record that retrieval obtains, is selected corresponding video segment in video library, show on monitor terminal as result for retrieval.
Can know by technique scheme, the present invention is based on moving target video content is carried out automatic index, and calculate the vision statistical nature of moving target, the video content retrieval is converted into the coupling of vision statistical nature.Through to each motion target detection, tracking in all video contents in the video library, can realize the index fast and automatically of video content, thereby guarantee the high efficiency of video content retrieval, improved the applicability of video frequency search system.Retrieval rate of the present invention is fast, accuracy and strong robustness.
Description of drawings
Fig. 1 is a workflow diagram of the present invention;
Fig. 2 is the workflow diagram of video index module among the present invention;
Fig. 3 is a high-level schematic functional block diagram of the present invention;
Fig. 4 is video index database table structure figure of the present invention.
Embodiment
A kind of video content search method based on moving target; This method comprises the step of following order: (1) obtains the initial pictures sequence of all video files in the video library; The initial pictures sequence is carried out analyzing and processing; Obtain the position of each moving target, calculate the vision statistical nature of each moving target, and the position and the vision statistical nature information stores of each moving target arrived in the video index database; (2) user imports object to be retrieved, and therefrom chooses moving target to be retrieved, and calculates the vision statistical nature of moving target to be retrieved; (3) in the video index database, search the video index record the most similar with moving target to be retrieved; (4), from video library, select corresponding video segment, and on user monitoring terminal, show according to the video index record that retrieves.As shown in Figure 1.
Described object to be retrieved is image or a video segment to be retrieved.If the user imports piece image, then by user's selected subject area to be retrieved from image; If the user imports one section video, then obtain all moving targets in the video through background subtraction by system, therefrom select moving target to be retrieved by the user again.
Like Fig. 1, shown in 2; The video index module is obtained the image sequence of all video files from video library; Utilize background subtraction that image sequence is carried out analyzing and processing, obtain the position and the image of each moving target, according to the vision statistical nature of each moving target of image calculation of moving target; Carry out motion target tracking in interframe, and the tracking results of position, vision statistical nature information and each moving target of each moving target is stored in the video index database; The user inputs to the searching object load module with object to be retrieved; And from object to be retrieved, choose moving target to be retrieved; The Visual Feature Retrieval Process module is calculated the vision statistical nature of moving target to be retrieved automatically, specifically comprises shape facilities such as circularity, length breadth ratio, color characteristics such as color average, color histogram; Textural characteristics such as entropy and gray scale symbiosis square; Obtain the index information of moving target at last, comprise information such as moving target ID, frame ID, the position in two field picture, vision statistical nature vector, video ID, and be saved in the video index storehouse; The video frequency searching module is searched the video index record the most close with moving target to be retrieved in the video index database; Corresponding video segment selected in the video index record that the result for retrieval output module obtains according to retrieval in video library, show on monitor terminal as result for retrieval.
As shown in Figure 2, adopt the formula of background subtraction detection moving target following:
Wherein, The pixel value that
expression
video image is constantly located at
;
is
pixel value located at
of background image constantly;
is the foreground pixel value, and
is the threshold value constant; Background image
adopts following formula to upgrade:
Wherein
is turnover rate.
The foreground image that background subtraction obtains often exists noise and hole, in order to obtain the accurate target profile, so that follow-up target's feature-extraction need be carried out pre-service to foreground image.At first foreground image is carried out morphological erosion, remove the noise in the foreground image, then foreground image is carried out morphology and expand, eliminate the hole in the foreground image.
As shown in Figure 2; After obtaining the initial position of moving target, along with the variation of moving target position, the barycenter through calculating moving target is at the displacement of next frame; Obtain and the minimum moving target of former frame moving target relative displacement; It is regarded as same moving target, thereby realizes the tracking of moving target, algorithm is following:
If certain moving target in t-1 center-of-mass coordinate constantly is in the image (xt-1, yt-1), then at t constantly, the center-of-mass coordinate of this moving target is:
(x
t,y
t)={(x
t,k,y
t,k)|min(sqrt((x
k,t-x
t-1)
2+(y
k,t-y
t-1)
2),k=1,2,...}
Wherein, (x
K, t, y
K, t) be the t center-of-mass coordinate of k moving target constantly, when the trace information with moving target was saved in the video index database, the same moving target of different frame had identical ID, thereby tracking results is preserved.
As shown in Figure 1, the video frequency searching module is right based on the aspect ratio that statistical learning method carries out moving target, therefrom searches the video index record the most close with moving target to be retrieved, and the aspect ratio of moving target is accomplished adopting supporting vector machine model.
As shown in Figure 3; Native system comprises five modules: the video index module; From video library, obtain the image sequence of all video files; Utilize background subtraction that image sequence is carried out analyzing and processing, obtain the position and the image of each moving target, according to the vision statistical nature of each moving target of image calculation of moving target; Carry out motion target tracking in interframe, and the tracking results of position, vision statistical nature information and each moving target of each moving target is stored in the video index database; The searching object load module, the user inputs to the searching object load module with object to be retrieved, and from object to be retrieved, chooses moving target to be retrieved; The Visual Feature Retrieval Process module is calculated the vision statistical nature of moving target to be retrieved automatically; The video frequency searching module is searched the video index record the most close with moving target to be retrieved in the video index database; The result for retrieval output module according to the video index record that retrieval obtains, is selected corresponding video segment in video library, show on monitor terminal as result for retrieval.
As shown in Figure 3, described video index module comprises: the moving object detection module, obtain the initial pictures sequence of all video files in the video library, and according to the initial pictures sequence, adopt background subtraction to detect the position that obtains each moving target.The motion target tracking module is carried out motion target tracking in interframe, and the tracking results of moving target is kept in the video index storehouse; The video index database, the tracking results of storing position, vision statistical nature information and each moving target of each moving target.
The video statistics characteristic can adopt the combination in any mode of color, texture, shape facility; Enumerated a typical combination of vision statistical nature among Fig. 4; Comprise shape facilities such as circularity, length breadth ratio; Color characteristic such as color moment, color histogram, textural characteristics such as entropy, gray scale symbiosis square, the classic algorithm of the employing Flame Image Process of these vision statistical natures is calculated.With the color histogram is example; It is the color characteristic that a kind of quilt extensively adopts; Be used for describing different color in the shared ratio of entire image; The computation process that is adopted is: in the RGB color space, carry out color quantizing, color space is divided between several little chromatic zoneses, the pixel quantity that drops in each quantized interval through the calculating color obtains color histogram.
As shown in Figure 4, the video index database comprises a moving target concordance list and a video file concordance list.In the moving target concordance list; Comprise moving target ID, video ID, frame ID and the position coordinates of barycenter on two field picture; These information descriptions the frame position of moving target place video; And corresponding with the vision statistical nature information of moving target, same moving target has identical ID at different frame, and this is realized by the motion target tracking module; Color moment, color histogram, circularity, length breadth ratio, Hu invariant moments, entropy, gray scale symbiosis square have been described the vision statistical nature of moving target.In the video file concordance list, the recording of video storehouse comprise video ID, deposit information such as path, video file name, creation-time, video duration, video profile, use when supplying video index and showing result for retrieval.
As shown in Figure 3; Searching object load module and result for retrieval output module adopt visual user interface; Through this interface input retrieval of content; Result for retrieval is shown to the user with the tabular form of video segment, and can carries out ordering, screening and the broadcast of video segment, make things convenient for the user in result for retrieval, further to seek most interested video segment.
In sum, the present invention is based on moving target video content is carried out automatic index, and calculate the vision statistical nature of moving target, the video content retrieval is converted into the coupling of vision statistical nature.Through to each motion target detection, tracking in all video contents in the video library, can realize the index fast and automatically of video content, thereby guarantee the high efficiency of video content retrieval, improved the applicability of video frequency search system.Retrieval rate of the present invention is fast, accuracy and strong robustness.
Claims (9)
1. video content search method based on moving target, this method comprises the step of following order:
(1) obtains the initial pictures sequence of all video files in the video library; The initial pictures sequence is carried out analyzing and processing; Obtain the position of each moving target; Calculate the vision statistical nature of each moving target, and the position and the vision statistical nature information stores of each moving target arrived in the video index database;
(2) user imports object to be retrieved, and therefrom chooses moving target to be retrieved, and calculates the vision statistical nature of moving target to be retrieved;
(3) in the video index database, search the video index record the most similar with moving target to be retrieved;
(4), from video library, select corresponding video segment, and on user monitoring terminal, show according to the video index record that retrieves.
2. the video content search method based on moving target according to claim 1 is characterized in that: described object to be retrieved is image or a video segment to be retrieved.
3. the video content search method based on moving target according to claim 1; It is characterized in that: the video index module is obtained the image sequence of all video files from video library; Utilize background subtraction that image sequence is carried out analyzing and processing; Obtain the position and the image of each moving target; Vision statistical nature according to each moving target of image calculation of moving target carries out motion target tracking in interframe, and the tracking results of position, vision statistical nature information and each moving target of each moving target is stored in the video index database; The user inputs to the searching object load module with object to be retrieved, and from object to be retrieved, chooses moving target to be retrieved, and the Visual Feature Retrieval Process module is calculated the vision statistical nature of moving target to be retrieved automatically; The video frequency searching module is searched the video index record the most close with moving target to be retrieved in the video index database; Corresponding video segment selected in the video index record that the result for retrieval output module obtains according to retrieval in video library, show on monitor terminal as result for retrieval.
4. the video content search method based on moving target according to claim 3 is characterized in that: adopt the formula of background subtraction detection moving target following:
Wherein, The pixel value that
expression
video image is constantly located at
;
is
pixel value located at
of background image constantly;
is the foreground pixel value, and
is the threshold value constant; Background image
adopts following formula to upgrade:
Wherein
is turnover rate.
5. the video content search method based on moving target according to claim 3; It is characterized in that: after obtaining the initial position of moving target, along with the variation of moving target position, the barycenter through calculating moving target is at the displacement of next frame; Obtain and the minimum moving target of former frame moving target relative displacement; It is regarded as same moving target, thereby realizes the tracking of moving target, algorithm is following:
If certain moving target in t-1 center-of-mass coordinate constantly is in the image (xt-1, yt-1), then at t constantly, the center-of-mass coordinate of this moving target is:
(x
t,y
t)={(x
t,k,y
t,k)|min(sqrt((x
k,t-x
t-1)
2+(y
k,t-y
t-1)
2),k=1,2,...}
Wherein, (x
K, t, y
K, t) be the t center-of-mass coordinate of k moving target constantly, when the trace information with moving target was saved in the video index database, the same moving target of different frame had identical ID, thereby tracking results is preserved.
6. the video content search method based on moving target according to claim 3; It is characterized in that: the video frequency searching module is right based on the aspect ratio that statistical learning method carries out moving target; Therefrom search the video index record the most close with moving target to be retrieved, the aspect ratio of moving target is accomplished adopting supporting vector machine model.
7. the video content search method based on moving target according to claim 4; It is characterized in that: the foreground image to obtaining through background subtraction carries out pre-service; At first foreground image is carried out morphological erosion; Remove the noise in the foreground image, then foreground image is carried out morphology and expand, eliminate the hole in the foreground image.
8. video content searching system based on moving target comprises:
The video index module; From video library, obtain the image sequence of all video files; Utilize background subtraction that image sequence is carried out analyzing and processing, obtain the position and the image of each moving target, according to the vision statistical nature of each moving target of image calculation of moving target; Carry out motion target tracking in interframe, and the tracking results of position, vision statistical nature information and each moving target of each moving target is stored in the video index database;
The searching object load module, the user inputs to the searching object load module with object to be retrieved, and from object to be retrieved, chooses moving target to be retrieved;
The Visual Feature Retrieval Process module is calculated the vision statistical nature of moving target to be retrieved automatically;
The video frequency searching module is searched the video index record the most close with moving target to be retrieved in the video index database;
The result for retrieval output module according to the video index record that retrieval obtains, is selected corresponding video segment in video library, show on monitor terminal as result for retrieval.
9. the video content searching system based on moving target according to claim 8 is characterized in that: described video index module comprises:
The moving object detection module is obtained the initial pictures sequence of all video files in the video library, according to the initial pictures sequence, adopts background subtraction to detect the position that obtains each moving target;
The motion target tracking module is carried out motion target tracking in interframe, and the tracking results of moving target is kept in the video index storehouse;
The video index database, the tracking results of storing position, vision statistical nature information and each moving target of each moving target.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012101159853A CN102609548A (en) | 2012-04-19 | 2012-04-19 | Video content retrieval method and system based on moving objects |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012101159853A CN102609548A (en) | 2012-04-19 | 2012-04-19 | Video content retrieval method and system based on moving objects |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102609548A true CN102609548A (en) | 2012-07-25 |
Family
ID=46526920
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012101159853A Pending CN102609548A (en) | 2012-04-19 | 2012-04-19 | Video content retrieval method and system based on moving objects |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102609548A (en) |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102930061A (en) * | 2012-11-28 | 2013-02-13 | 安徽水天信息科技有限公司 | Video abstraction method and system based on moving target detection |
| CN102970578A (en) * | 2012-11-19 | 2013-03-13 | 北京十分科技有限公司 | Multimedia information identifying and training method and device |
| CN103020260A (en) * | 2012-12-24 | 2013-04-03 | 中国科学院半导体研究所 | Video query method |
| CN103067692A (en) * | 2012-12-30 | 2013-04-24 | 同方(深圳)云计算技术股份有限公司 | Post-treatment detection method and system based on net-harddisk video recorder dangerous invasion |
| CN103198110A (en) * | 2013-03-28 | 2013-07-10 | 广州中国科学院软件应用技术研究所 | Method and system for rapid video data characteristic retrieval |
| CN103605786A (en) * | 2013-11-27 | 2014-02-26 | 姚领众 | Massive video retrieving method based on sample video clips |
| CN103873955A (en) * | 2012-12-18 | 2014-06-18 | 联想(北京)有限公司 | Video acquisition method and device |
| CN104572804A (en) * | 2013-10-24 | 2015-04-29 | Tcl集团股份有限公司 | Video object retrieval system and method |
| CN105141915A (en) * | 2015-08-26 | 2015-12-09 | 广东创我科技发展有限公司 | Video searching method and video searching system |
| CN107291910A (en) * | 2017-06-26 | 2017-10-24 | 图麟信息科技（深圳）有限公司 | A kind of video segment structuralized query method, device and electronic equipment |
| CN107358622A (en) * | 2017-06-19 | 2017-11-17 | 三峡大学 | A kind of video information processing method and system based on visualization movement locus |
| CN107480228A (en) * | 2017-08-03 | 2017-12-15 | 深圳市置辰海信科技有限公司 | A kind of monitor video searching method towards hard objectives |
| CN107690058A (en) * | 2016-08-04 | 2018-02-13 | 惠州学院 | A kind of multiple target video frequency following system and method |
| CN108600779A (en) * | 2018-05-18 | 2018-09-28 | 紫光华山信息技术有限公司 | A kind of target object operating method and device based on video content |
| CN109408671A (en) * | 2018-07-23 | 2019-03-01 | 中国联合网络通信集团有限公司 | The searching method and its system of specific objective |
| CN109951688A (en) * | 2018-10-23 | 2019-06-28 | 王龙 | A kind of video monitoring system and method realizing continuity scene relating and excavating |
| CN110083742A (en) * | 2019-04-29 | 2019-08-02 | 腾讯科技（深圳）有限公司 | A kind of video query method and device |
| CN110225282A (en) * | 2019-05-31 | 2019-09-10 | 努比亚技术有限公司 | A kind of video record control method, equipment and computer readable storage medium |
| CN111538858A (en) * | 2020-05-06 | 2020-08-14 | 英华达（上海）科技有限公司 | Method and device for establishing video map, electronic equipment and storage medium |
| CN112019928A (en) * | 2019-05-30 | 2020-12-01 | 杭州海康威视数字技术股份有限公司 | Video playback method and device and electronic equipment |
| WO2020238789A1 (en) * | 2019-05-30 | 2020-12-03 | 杭州海康威视数字技术股份有限公司 | Video replay |
| CN113761239A (en) * | 2021-09-08 | 2021-12-07 | 武汉工程大学 | Index database establishing and retrieving method based on massive geographic position information |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101159859A (en) * | 2007-11-29 | 2008-04-09 | 北京中星微电子有限公司 | Motion detection method, device and an intelligent monitoring system |
| US7720301B2 (en) * | 2004-02-18 | 2010-05-18 | Huaya Microelectronics, Ltd. | Structure characterization of images |
| CN201766663U (en) * | 2010-03-30 | 2011-03-16 | 苏州市职业大学 | Residential property monitoring system |
| CN102117313A (en) * | 2010-12-29 | 2011-07-06 | 天脉聚源(北京)传媒科技有限公司 | Video retrieval method and system |
-
2012
- 2012-04-19 CN CN2012101159853A patent/CN102609548A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7720301B2 (en) * | 2004-02-18 | 2010-05-18 | Huaya Microelectronics, Ltd. | Structure characterization of images |
| CN101159859A (en) * | 2007-11-29 | 2008-04-09 | 北京中星微电子有限公司 | Motion detection method, device and an intelligent monitoring system |
| CN201766663U (en) * | 2010-03-30 | 2011-03-16 | 苏州市职业大学 | Residential property monitoring system |
| CN102117313A (en) * | 2010-12-29 | 2011-07-06 | 天脉聚源(北京)传媒科技有限公司 | Video retrieval method and system |
Non-Patent Citations (1)
| Title |
|---|
| 卢燕斌: "监控视频中运动目标检测和追踪的研究及应用", 《上海交通大学工程硕士专业学位论文》, 15 June 2007 (2007-06-15) * |
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102970578A (en) * | 2012-11-19 | 2013-03-13 | 北京十分科技有限公司 | Multimedia information identifying and training method and device |
| CN102930061A (en) * | 2012-11-28 | 2013-02-13 | 安徽水天信息科技有限公司 | Video abstraction method and system based on moving target detection |
| CN103873955A (en) * | 2012-12-18 | 2014-06-18 | 联想(北京)有限公司 | Video acquisition method and device |
| CN103020260A (en) * | 2012-12-24 | 2013-04-03 | 中国科学院半导体研究所 | Video query method |
| CN103067692A (en) * | 2012-12-30 | 2013-04-24 | 同方(深圳)云计算技术股份有限公司 | Post-treatment detection method and system based on net-harddisk video recorder dangerous invasion |
| CN103067692B (en) * | 2012-12-30 | 2016-01-20 | 同方(深圳)云计算技术股份有限公司 | The post-processing detection method and system of DVR danger invasion Network Based |
| CN103198110A (en) * | 2013-03-28 | 2013-07-10 | 广州中国科学院软件应用技术研究所 | Method and system for rapid video data characteristic retrieval |
| CN104572804A (en) * | 2013-10-24 | 2015-04-29 | Tcl集团股份有限公司 | Video object retrieval system and method |
| CN104572804B (en) * | 2013-10-24 | 2019-04-16 | Tcl集团股份有限公司 | A kind of method and its system of video object retrieval |
| CN103605786A (en) * | 2013-11-27 | 2014-02-26 | 姚领众 | Massive video retrieving method based on sample video clips |
| CN105141915A (en) * | 2015-08-26 | 2015-12-09 | 广东创我科技发展有限公司 | Video searching method and video searching system |
| CN107690058A (en) * | 2016-08-04 | 2018-02-13 | 惠州学院 | A kind of multiple target video frequency following system and method |
| CN107358622A (en) * | 2017-06-19 | 2017-11-17 | 三峡大学 | A kind of video information processing method and system based on visualization movement locus |
| CN107291910A (en) * | 2017-06-26 | 2017-10-24 | 图麟信息科技（深圳）有限公司 | A kind of video segment structuralized query method, device and electronic equipment |
| CN107480228A (en) * | 2017-08-03 | 2017-12-15 | 深圳市置辰海信科技有限公司 | A kind of monitor video searching method towards hard objectives |
| CN108600779A (en) * | 2018-05-18 | 2018-09-28 | 紫光华山信息技术有限公司 | A kind of target object operating method and device based on video content |
| CN108600779B (en) * | 2018-05-18 | 2021-04-06 | 新华三信息技术有限公司 | Target object operation method and device based on video content |
| CN109408671A (en) * | 2018-07-23 | 2019-03-01 | 中国联合网络通信集团有限公司 | The searching method and its system of specific objective |
| CN109951688A (en) * | 2018-10-23 | 2019-06-28 | 王龙 | A kind of video monitoring system and method realizing continuity scene relating and excavating |
| CN109951688B (en) * | 2018-10-23 | 2021-01-29 | 和宇健康科技股份有限公司 | Video monitoring system and method for realizing continuous scene association mining |
| CN110083742A (en) * | 2019-04-29 | 2019-08-02 | 腾讯科技（深圳）有限公司 | A kind of video query method and device |
| CN110083742B (en) * | 2019-04-29 | 2022-12-06 | 腾讯科技（深圳）有限公司 | Video query method and device |
| CN112019928B (en) * | 2019-05-30 | 2022-05-06 | 杭州海康威视数字技术股份有限公司 | Video playback method and device and electronic equipment |
| CN112019928A (en) * | 2019-05-30 | 2020-12-01 | 杭州海康威视数字技术股份有限公司 | Video playback method and device and electronic equipment |
| WO2020238789A1 (en) * | 2019-05-30 | 2020-12-03 | 杭州海康威视数字技术股份有限公司 | Video replay |
| CN110225282A (en) * | 2019-05-31 | 2019-09-10 | 努比亚技术有限公司 | A kind of video record control method, equipment and computer readable storage medium |
| CN110225282B (en) * | 2019-05-31 | 2023-05-30 | 山西仨仁行文化传媒有限公司 | Video recording control method, device and computer readable storage medium |
| CN111538858A (en) * | 2020-05-06 | 2020-08-14 | 英华达（上海）科技有限公司 | Method and device for establishing video map, electronic equipment and storage medium |
| CN111538858B (en) * | 2020-05-06 | 2023-06-23 | 英华达（上海）科技有限公司 | Method, device, electronic equipment and storage medium for establishing video map |
| TWI823018B (en) * | 2020-05-06 | 2023-11-21 | 英華達股份有限公司 | Method of video graph developing |
| CN113761239A (en) * | 2021-09-08 | 2021-12-07 | 武汉工程大学 | Index database establishing and retrieving method based on massive geographic position information |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102609548A (en) | Video content retrieval method and system based on moving objects | |
| KR102560308B1 (en) | System and method for exterior search | |
| CN105141903B (en) | A kind of method for carrying out target retrieval in video based on colouring information | |
| CN110414559B (en) | Construction method of intelligent retail cabinet commodity target detection unified framework and commodity identification method | |
| US9323785B2 (en) | Method and system for mobile visual search using metadata and segmentation | |
| US8385646B2 (en) | Image processing | |
| CN103207898B (en) | A kind of similar face method for quickly retrieving based on local sensitivity Hash | |
| GB2607749A (en) | Fine-grained visual recognition in mobile augmented reality | |
| US9224211B2 (en) | Method and system for motion detection in an image | |
| CN104081435A (en) | Image matching method based on cascading binary encoding | |
| CN106937114A (en) | Method and apparatus for being detected to video scene switching | |
| Priya et al. | Edge strength extraction using orthogonal vectors for shot boundary detection | |
| CN104715251A (en) | Salient object detection method based on histogram linear fitting | |
| CN103428537A (en) | Video processing method and video processing device | |
| Hung et al. | A fast algorithm of temporal median filter for background subtraction. | |
| CN111368867A (en) | Archive classification method and system and computer readable storage medium | |
| CN113051984A (en) | Video copy detection method and apparatus, storage medium, and electronic apparatus | |
| CN113674049A (en) | Commodity shelf position identification method and system based on picture search and storage medium | |
| CN111160099B (en) | Intelligent segmentation method for video image target | |
| Kethsy Prabavathy et al. | Histogram difference with fuzzy rule base modeling for gradual shot boundary detection in video cloud applications | |
| Uke et al. | Motion tracking system in video based on extensive feature set | |
| Sreedevi et al. | Fast image retrieval with feature levels | |
| CN105120335A (en) | A method and apparatus for processing television program pictures | |
| Lin et al. | Analysis of shot boundary based on color and texture features of frame | |
| Srilakshmi et al. | Shot boundary detection using structural similarity index |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120725 |
|
| WD01 | Invention patent application deemed withdrawn after publication |