CN103699679A - Method and equipment for retrieving information of target objects - Google Patents
Method and equipment for retrieving information of target objectsDownload PDFInfo
- Publication number
- CN103699679A CN103699679ACN201310753628.4ACN201310753628ACN103699679ACN 103699679 ACN103699679 ACN 103699679ACN 201310753628 ACN201310753628 ACN 201310753628ACN 103699679 ACN103699679 ACN 103699679A
- Authority
- CN
- China
- Prior art keywords
- picture
- information
- photo
- concerned
- pictures
- 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
Classifications
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
Technical field
The present invention relates to a kind of information retrieval method and information searching device of destination object.
Background technology
At present, information retrieval is generally undertaken by network search engines, such as search engines such as utilizing Google, Baidu is realized the retrieval of information.The information of people in order to obtain, conventional search method first selects several keywords, then inputs keyword and searches in the search box of search engine, to obtain retrieving information.
The above-mentioned search that utilizes network to carry out, has expanded to many portable equipments now, such as smart mobile phone etc.Like this, people can carry out information retrieval very easily anywhere.
Yet the shortcoming of this information retrieval method is that people must be for feature or some understanding of information of searched targets.When not understanding the information of target, will be difficult to adopt this information retrieval method to obtain and want the relevant information obtaining.For instance, when visitor arrives a place (such as monument or tower etc.) with milestone in travelling, in the situation that not seeing any Word message, visitor will understand this target, is also difficult to carry out network retrieval by keyword.That is to say, in prior art, lack a kind of method that can effectively retrieve the information of unknown object.
Summary of the invention
The technical problem to be solved in the present invention is need user to provide search key and be difficult to be applied to the uncomprehending destination object of user in order to overcome the information retrieval method of prior art, cannot carry out the effectively defect of retrieval to the information of unknown object, propose a kind of information retrieval method and information searching device of destination object.
The present invention solves above-mentioned technical matters by following technical proposals:
The invention provides a kind of information retrieval method of destination object, its feature is, comprises the following steps:
S1, photographic subjects object photo, and location obtains current location;
S2, from picture shared data bank, choose all or part of picture that the distance between geographic position and current location is less than the first default distance threshold as co-located picture, the picture wherein having in this picture shared data bank has label and geographic position;
S3, from this photo, extract some characteristics of image;
S4, from co-located picture, filter out there are these some characteristics of image picture as picture concerned;
S5, obtain the relatively many some labels of multiplicity in the label of all picture concerned;
S6, these some labels are carried out to network retrieval as keyword, and the information that obtains of output retrieval.
It will be appreciated by those skilled in the art that S1the current location that middle location obtains is photographer's position, and because the distance of taking pictures is conventionally shorter, thereby current location should be similar to destination object present position in most cases.In fact destination object in the present invention can be any object, and what be especially applicable to is the static constant object in position, as some sight spots, television tower, monument etc.S3in image characteristics extraction can adopt any one image characteristics extraction algorithm of the prior art to realize, the characteristics of image of extraction can be the feature of profile, color etc.S2in larger at picture shared data bank, meet distance between geographic position and current location and be less than the picture number of first default this condition of distance threshold when more, can choose meet this condition part picture as co-located picture, otherwise can choose meet this condition whole pictures as co-located picture.
This picture shared data bank can be that photo is shared website as the picture database of Flickr and Picasa, what in the picture database of the shared website of these photos, store is substantially all the photo that user uploads, and in upload pictures, can add geographic position and some labels.Label is herein appreciated that the keyword arranging for ease of classification.For instance, can be " historical building ", " skyscraper " etc.After the image that has utilized positional information and photo screens, can think to screen in the picture obtaining to have the photo that major part is destination object equally.Thereby S5in these some labels of obtaining have very large probability to be also applicable to this destination object.Thereby these some labels are retrieved as keyword, can obtain the information of this destination object.And, S5in multiplicity mean relatively more, the multiplicity of these some labels is with respect to the multiplicity of other labels in the label of all picture concerned and Yan Gengduo.
At this destination object, for user, be unknown, user cannot initiatively provide in the situation of search key, above-mentioned search method can utilize network picture resource automatically effectively to retrieve according to positional information and photo, thereby obtains the information of this unknown destination object.
Preferably, S2and S3between further comprising the steps of:
S21, from co-located picture, choose picture that the distance between geographic position and current location is less than default second distance threshold value as picture concerned, and remove picture concerned from co-located picture, wherein second distance threshold value is less than the first distance threshold.
Second distance threshold value should be significantly less than the first distance threshold.Like this, as step S21in when the determination result is NO, can think that the position of picture concerned and this destination object is almost identical.Now, have greater probability picture concerned for object and this photo in the destination object taken be identical.
Hold intelligibly step S21after carrying out with before carrying out, compare, in co-located picture, lacked S21in picture concerned.S21in picture concerned and S4in picture concerned itself not identical, but after filtering out from co-located picture, using this two parts picture as a class picture, picture concerned herein should be understood to a class picture.
Preferably, S2by S2areplace S2afor: adopt k-means clustering algorithm by the whole pictures that have in this picture shared data bank according to geographic Location Classification, and using geographic position and the immediate class picture of current location as co-located picture.
K-mean algorithm is a kind of hard clustering algorithm, belongs to the objective function clustering method of local prototype.It has, and time complexity is low, the fireballing advantage of cluster.The processing procedure of k-means algorithm is as follows: first, choose at random k object as the barycenter of individual bunch of initial k; Then, all the other objects are assigned to nearest bunch according to the distance of itself and each bunch barycenter; Finally, even if the barycenter of each bunch again.Said process constantly repeats, until the minimization of object function.The steps flow chart of k-means clustering algorithm is substantially as follows:
Step 1, select k object as the barycenter of initial bunch at random;
The distance of the barycenter ofstep 2, calculating object and each bunch, is divided into object apart from its nearest bunch;
If the barycenter of step 4 bunch no longer changes, return to division result, otherwise return tostep 2.
Preferably, S3for: adopt gabor small wave converting method to extract global feature and the local feature of this photo;
S4for: adopt gabor small wave converting method to extract global feature and the local feature of co-located picture, and filter out co-located picture that global feature is identical with this photo with local feature as picture concerned.
Gabor small wave converting method can provide good set direction and scale selection characteristic, and insensitive for illumination variation, has the adaptability to illumination variation.And in the present invention, need to carry out image characteristics extraction to as if photo and picture, and this part picture is also likely the photo that user uploads more often than not.The characteristics of image of photo is subject to illumination effect, utilizes gabor small wave converting method to extract whole and part characteristic information to the image in the present invention, can more effectively avoid the impact of illumination.It should be noted in the discussion above that above-mentioned global feature and local feature are all for single picture, any in co-located picture processes through gabor small wave converting method global feature and the local feature that the characteristics of image extracting has included this pictures.
The present invention also provides a kind of information searching device, and its feature is, comprising:
One camera module, for the photo of photographic subjects object;
One locating module, obtains current location for location;
One primary importance screening module, for choosing all or part of picture that the distance between geographic position and current location is less than the first default distance threshold from picture shared data bank as co-located picture, the picture wherein having in this picture shared data bank has label and geographic position;
One characteristics of image screening module, for extracting some characteristics of image from this photo, then from co-located picture, filter out there are these some characteristics of image picture as picture concerned;
One keyword retrieval module, for obtaining the relatively many some labels of label multiplicity of all picture concerned, and carries out network retrieval using these some labels as keyword, the information that output retrieval obtains.
Preferably, this primary importance screening module is enabled a second place screening module after choosing co-located picture, this second place screening module for the picture choosing the distance between geographic position and current location from co-located picture and be less than default second distance threshold value as picture concerned, and remove picture concerned from co-located picture, then enable this characteristics of image screening module, wherein second distance threshold value is less than the first distance threshold.
Preferably, this primary importance screening module is replaced by a cluster screening module, this cluster screening module for whole pictures of adopting k-means clustering algorithm this picture shared data bank being had according to geographic Location Classification, and using geographic position and the immediate class picture of current location as co-located picture.
Preferably, this characteristics of image screening module is for adopting gabor small wave converting method to extract global feature and the local feature of this photo and co-located picture, then filters out picture that global feature is identical with this photo with local feature as picture concerned.
Preferably, this information searching device is mobile phone.
Meeting on the basis of this area general knowledge, above-mentioned each optimum condition, can combination in any, obtains the preferred embodiments of the invention.
Positive progressive effect of the present invention is:
The information retrieval method of destination object of the present invention and information searching device, by photograph image and the positional information of destination object, thereby suitable searched key word is searched in the picture shared data bank from network, to screen suitable picture Automatic-searching, thereby provide search key without user, just can effectively retrieve the information of unknown object, and user's use is very convenient.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of information retrieval method of the destination object of the embodiment of the present invention 1.
Fig. 2 is the schematic diagram of the mobile phone of the embodiment of thepresent invention 3.
Embodiment
Below in conjunction with accompanying drawing, provide preferred embodiment of the present invention, to describe technical scheme of the present invention in detail, but therefore do not limit the present invention among described scope of embodiments.
Embodiment 1
As shown in Figure 1, the information retrieval method of the destination object of the present embodiment comprises the following steps:
S1, photographic subjects object photo, and location obtains current location;
S2, from picture shared data bank, choose all or part of picture that the distance between geographic position and current location is less than the first default distance threshold as co-located picture, the picture wherein having in this picture shared data bank has label and geographic position;
S21, from co-located picture, choose picture that the distance between geographic position and current location is less than default second distance threshold value as picture concerned, and remove picture concerned from co-located picture, wherein second distance threshold value is less than the first distance threshold;
S3, adopt gabor small wave converting method to extract global feature and the local feature of this photo;
S4, adopt gabor small wave converting method to extract global feature and the local feature of co-located picture, and filter out co-located picture that global feature is identical with this photo with local feature as picture concerned;
S5, obtain the relatively many some labels of multiplicity in the label of all picture concerned;
S6, these some labels are carried out to network retrieval as keyword, and the information that obtains of output retrieval.
Wherein, second distance threshold value is 100 meters, and the first distance threshold is 500 meters.Like this, as step S21in when the determination result is NO, can think that the position of picture concerned and this destination object is almost identical.Now, have greater probability picture concerned for object and this photo in the destination object taken be identical.S2in all or part of picture be: in the situation that meet the picture number that distance between geographic position and current location is less than the condition of the first default distance threshold, be greater than a reference value (in the present embodiment, this reference value is 100), choose meet this condition 100 pictures as co-located picture; Otherwise, choose meet this condition all pictures as co-located picture.
In the present embodiment, the quantity of these some labels is 4, finally utilizes 4 keywords to carry out network retrieval, and network retrieval utilizes google search engine to carry out.
This picture shared data bank in the present embodiment is the picture database of the shared website Flickr of photo and Picasa.At this destination object, for user, be unknown, user cannot initiatively provide in the situation of search key, the search method of the present embodiment can utilize network picture resource automatically effectively to retrieve according to positional information and photo, thereby obtains the information of this unknown destination object.
The difference of the information retrieval method of the present embodiment and embodiment 1 is only:
S2by S2areplace S2afor: adopt k-means clustering algorithm by the whole pictures that have in this picture shared data bank according to geographic Location Classification, and using geographic position and the immediate class picture of current location as co-located picture.That is to say, if in the multiclass picture that adopts k-means clustering algorithm to obtain, mean value and the current location in the geographic position of a certain class picture are the most approaching, just using this class picture as co-located picture.
The step of the k-means clustering algorithm adopting in the present embodiment is as follows:
Step 1, select k object (i.e. picture in this picture shared data bank) as the barycenter (barycenter is herein for geographic position corresponding to picture, the centroid position calculating) of initial bunch at random;
The distance of the barycenter ofstep 2, calculating object and each bunch, is divided into object apart from its nearest bunch;
If the barycenter of step 4 bunch no longer changes, return to division result, otherwise return tostep 2.
As shown in Figure 2, the information searching device of the present embodiment is a mobile phone, and this mobile phone comprises a camera module 1, alocating module 2, a primaryimportance screening module 3, a second place screening module 4, characteristics ofimage screening module 5 and a keyword retrieval module 6.
This camera module is for the photo of photographic subjects object, and this locating module obtains current location for location.This primary importance screening module, for choosing all or part of picture that the distance between geographic position and current location is less than the first default distance threshold from picture shared data bank as co-located picture, then enable this second place screening module, the picture wherein having in this picture shared data bank has label and geographic position.
This second place screening module for the picture choosing the distance between geographic position and current location from co-located picture and be less than default second distance threshold value as picture concerned, and remove picture concerned from co-located picture, then enable this characteristics of image screening module, wherein second distance threshold value is less than the first distance threshold.This characteristics of image screening module is for adopting gabor small wave converting method to extract global feature and the local feature of this photo and co-located picture, then filters out picture that global feature is identical with this photo with local feature as picture concerned.
This keyword retrieval module, for obtaining the relatively many some labels of label multiplicity of all picture concerned, and carries out network retrieval using these some labels as keyword, the information that output retrieval obtains.
Embodiment 4
The present embodiment is compared with the mobile phone ofembodiment 3, and difference is only:
This primary importance screening module is replaced by a cluster screening module, this cluster screening module for whole pictures of adopting k-means clustering algorithm this picture shared data bank being had according to geographic Location Classification, and using geographic position and the immediate class picture of current location as co-located picture.
Although more than described the specific embodiment of the present invention, it will be understood by those of skill in the art that these only illustrate, protection scope of the present invention is limited by appended claims.Those skilled in the art is not deviating under the prerequisite of principle of the present invention and essence, can make various changes or modifications to these embodiments, but these changes and modification all fall into protection scope of the present invention.
Claims (9)
1. an information retrieval method for destination object, is characterized in that, comprises the following steps:
S1, photographic subjects object photo, and location obtains current location;
S2, from picture shared data bank, choose all or part of picture that the distance between geographic position and current location is less than the first default distance threshold as co-located picture, the picture wherein having in this picture shared data bank has label and geographic position;
S3, from this photo, extract some characteristics of image;
S4, from co-located picture, filter out there are these some characteristics of image picture as picture concerned;
S5, obtain the relatively many some labels of multiplicity in the label of all picture concerned;
S6, these some labels are carried out to network retrieval as keyword, and the information that obtains of output retrieval.
2. information retrieval method as claimed in claim 1, is characterized in that, S2and S3between further comprising the steps of:
S21, from co-located picture, choose picture that the distance between geographic position and current location is less than default second distance threshold value as picture concerned, and remove picture concerned from co-located picture, wherein second distance threshold value is less than the first distance threshold.
3. information retrieval method as claimed in claim 1 or 2, is characterized in that, S2by S2areplace S2afor: adopt k-means clustering algorithm by the whole pictures that have in this picture shared data bank according to geographic Location Classification, and using geographic position and the immediate class picture of current location as co-located picture.
4. information retrieval method as claimed in claim 1 or 2, is characterized in that, S3for: adopt gabor small wave converting method to extract global feature and the local feature of this photo;
S4for: adopt gabor small wave converting method to extract global feature and the local feature of co-located picture, and filter out co-located picture that global feature is identical with this photo with local feature as picture concerned.
5. an information searching device, is characterized in that, comprising:
One camera module, for the photo of photographic subjects object;
One locating module, obtains current location for location;
One primary importance screening module, for choosing all or part of picture that the distance between geographic position and current location is less than the first default distance threshold from picture shared data bank as co-located picture, the picture wherein having in this picture shared data bank has label and geographic position;
One characteristics of image screening module, for extracting some characteristics of image from this photo, then from co-located picture, filter out there are these some characteristics of image picture as picture concerned;
One keyword retrieval module, for obtaining the relatively many some labels of label multiplicity of all picture concerned, and carries out network retrieval using these some labels as keyword, the information that output retrieval obtains.
6. information searching device as claimed in claim 5, it is characterized in that, this primary importance screening module is enabled a second place screening module after choosing co-located picture, this second place screening module for the picture choosing the distance between geographic position and current location from co-located picture and be less than default second distance threshold value as picture concerned, and remove picture concerned from co-located picture, then enable this characteristics of image screening module, wherein second distance threshold value is less than the first distance threshold.
7. the information searching device as described in claim 5 or 6, it is characterized in that, this primary importance screening module is replaced by a cluster screening module, this cluster screening module for whole pictures of adopting k-means clustering algorithm this picture shared data bank being had according to geographic Location Classification, and using geographic position and the immediate class picture of current location as co-located picture.
8. the information searching device as described in claim 5 or 6, it is characterized in that, this characteristics of image screening module is for adopting gabor small wave converting method to extract global feature and the local feature of this photo and co-located picture, then filters out picture that global feature is identical with this photo with local feature as picture concerned.
9. the information searching device as described in claim 5 or 6, is characterized in that, this information searching device is mobile phone.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310753628.4ACN103699679A (en) | 2013-12-31 | 2013-12-31 | Method and equipment for retrieving information of target objects |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310753628.4ACN103699679A (en) | 2013-12-31 | 2013-12-31 | Method and equipment for retrieving information of target objects |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103699679Atrue CN103699679A (en) | 2014-04-02 |
Family
ID=50361207
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310753628.4APendingCN103699679A (en) | 2013-12-31 | 2013-12-31 | Method and equipment for retrieving information of target objects |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103699679A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103995871A (en)* | 2014-05-22 | 2014-08-20 | 东北大学 | Approximate massive address data nearest neighbor query method |
| CN104933165A (en)* | 2015-06-29 | 2015-09-23 | 努比亚技术有限公司 | Data screening strategy generation method and device |
| CN106294558A (en)* | 2016-07-26 | 2017-01-04 | 北京小米移动软件有限公司 | Object positioning method and device |
| CN106354744A (en)* | 2015-07-16 | 2017-01-25 | 三星电子株式会社 | Method for sharing content information and electronic device thereof |
| WO2017028573A1 (en)* | 2015-08-20 | 2017-02-23 | 惠州Tcl移动通信有限公司 | Method and system for processing picture information based on mobile terminal |
| CN107122499A (en)* | 2017-06-09 | 2017-09-01 | 苏州唯亚信息科技股份有限公司 | Suitable for the method for quickly querying of R ＆ D Database |
| CN107577790A (en)* | 2017-09-18 | 2018-01-12 | 北京金山安全软件有限公司 | Image searching method and device |
| CN110188230A (en)* | 2019-05-30 | 2019-08-30 | 中煤航测遥感集团有限公司 | Semantic-based image search method and device |
| CN110764726A (en)* | 2019-10-18 | 2020-02-07 | 网易（杭州）网络有限公司 | Target object determination method and device, terminal device and storage medium |
| CN111881322A (en)* | 2020-09-28 | 2020-11-03 | 成都睿沿科技有限公司 | Target searching method and device, electronic equipment and storage medium |
| CN114064953A (en)* | 2021-11-26 | 2022-02-18 | 北京度友信息技术有限公司 | Image search method, device, electronic device and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101241504A (en)* | 2008-01-23 | 2008-08-13 | 武汉大学 | A content-based intelligent search method for remote sensing image data |
| CN101340661A (en)* | 2008-08-14 | 2009-01-07 | 北京中星微电子有限公司 | Guide control implementing mobile apparatus and server, guide control method |
| CN101847154A (en)* | 2010-02-26 | 2010-09-29 | 宇龙计算机通信科技(深圳)有限公司 | Method and system for inquiring information and method for mobile terminal to inquire information |
| CN101924992A (en)* | 2010-07-30 | 2010-12-22 | 中国电信股份有限公司 | Method, system and equipment for acquiring scene information through mobile terminal |
| US20110314049A1 (en)* | 2010-06-22 | 2011-12-22 | Xerox Corporation | Photography assistant and method for assisting a user in photographing landmarks and scenes |
| CN102411582A (en)* | 2010-09-21 | 2012-04-11 | 腾讯科技（深圳）有限公司 | Picture searching method and device and client |
| CN102831242A (en)* | 2012-09-10 | 2012-12-19 | 东莞宇龙通信科技有限公司 | Method and device for searching picture information |
| US20120323914A1 (en)* | 2004-11-29 | 2012-12-20 | Ariel Inventions, Llc | Storing and retrieving associated information with a digital image |
| CN103136228A (en)* | 2011-11-25 | 2013-06-05 | 阿里巴巴集团控股有限公司 | Image search method and image search device |
- 2013
- 2013-12-31CNCN201310753628.4Apatent/CN103699679A/enactivePending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120323914A1 (en)* | 2004-11-29 | 2012-12-20 | Ariel Inventions, Llc | Storing and retrieving associated information with a digital image |
| CN101241504A (en)* | 2008-01-23 | 2008-08-13 | 武汉大学 | A content-based intelligent search method for remote sensing image data |
| CN101340661A (en)* | 2008-08-14 | 2009-01-07 | 北京中星微电子有限公司 | Guide control implementing mobile apparatus and server, guide control method |
| CN101847154A (en)* | 2010-02-26 | 2010-09-29 | 宇龙计算机通信科技(深圳)有限公司 | Method and system for inquiring information and method for mobile terminal to inquire information |
| US20110314049A1 (en)* | 2010-06-22 | 2011-12-22 | Xerox Corporation | Photography assistant and method for assisting a user in photographing landmarks and scenes |
| CN101924992A (en)* | 2010-07-30 | 2010-12-22 | 中国电信股份有限公司 | Method, system and equipment for acquiring scene information through mobile terminal |
| CN102411582A (en)* | 2010-09-21 | 2012-04-11 | 腾讯科技（深圳）有限公司 | Picture searching method and device and client |
| CN103136228A (en)* | 2011-11-25 | 2013-06-05 | 阿里巴巴集团控股有限公司 | Image search method and image search device |
| CN102831242A (en)* | 2012-09-10 | 2012-12-19 | 东莞宇龙通信科技有限公司 | Method and device for searching picture information |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103995871B (en)* | 2014-05-22 | 2017-04-12 | 东北大学 | Approximate massive address data nearest neighbor query method |
| CN103995871A (en)* | 2014-05-22 | 2014-08-20 | 东北大学 | Approximate massive address data nearest neighbor query method |
| CN104933165A (en)* | 2015-06-29 | 2015-09-23 | 努比亚技术有限公司 | Data screening strategy generation method and device |
| CN104933165B (en)* | 2015-06-29 | 2018-06-05 | 努比亚技术有限公司 | Data screening strategy-generating method and device |
| CN106354744A (en)* | 2015-07-16 | 2017-01-25 | 三星电子株式会社 | Method for sharing content information and electronic device thereof |
| WO2017028573A1 (en)* | 2015-08-20 | 2017-02-23 | 惠州Tcl移动通信有限公司 | Method and system for processing picture information based on mobile terminal |
| US20170278267A1 (en)* | 2015-08-20 | 2017-09-28 | HuiZhou TLC Mobile Communication Co., Ltd. | Picture information processing method and system based on mobile terminal |
| CN106294558A (en)* | 2016-07-26 | 2017-01-04 | 北京小米移动软件有限公司 | Object positioning method and device |
| CN107122499A (en)* | 2017-06-09 | 2017-09-01 | 苏州唯亚信息科技股份有限公司 | Suitable for the method for quickly querying of R ＆ D Database |
| CN107577790A (en)* | 2017-09-18 | 2018-01-12 | 北京金山安全软件有限公司 | Image searching method and device |
| CN110188230A (en)* | 2019-05-30 | 2019-08-30 | 中煤航测遥感集团有限公司 | Semantic-based image search method and device |
| CN110764726A (en)* | 2019-10-18 | 2020-02-07 | 网易（杭州）网络有限公司 | Target object determination method and device, terminal device and storage medium |
| CN110764726B (en)* | 2019-10-18 | 2023-08-22 | 网易（杭州）网络有限公司 | Target object determination method and device, terminal equipment and storage medium |
| CN111881322A (en)* | 2020-09-28 | 2020-11-03 | 成都睿沿科技有限公司 | Target searching method and device, electronic equipment and storage medium |
| CN114064953A (en)* | 2021-11-26 | 2022-02-18 | 北京度友信息技术有限公司 | Image search method, device, electronic device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103699679A (en) | Method and equipment for retrieving information of target objects | |
| US20180365489A1 (en) | Automatically organizing images | |
| US8831352B2 (en) | Event determination from photos | |
| JP5170961B2 (en) | Image processing system, image processing apparatus and method, program, and recording medium | |
| US20180108182A1 (en) | Context-aware tagging for augmented reality environments | |
| CN104331509A (en) | Picture managing method and device | |
| CN104426841A (en) | Method for arranging background image, and correlation server and system | |
| JP2007206919A (en) | Display control device, method, program and storage medium | |
| US10380267B2 (en) | System and method for tagging multimedia content elements | |
| US9665773B2 (en) | Searching for events by attendants | |
| US20240037142A1 (en) | Systems and methods for filtering of computer vision generated tags using natural language processing | |
| KR101782590B1 (en) | Method for Providing and Recommending Related Tag Using Image Analysis | |
| CN105159976A (en) | Image file processing method and system | |
| US11347739B2 (en) | Performing a chained search function | |
| JP6109118B2 (en) | Image processing apparatus and method, information processing apparatus and method, and program | |
| US10885619B2 (en) | Context-based imagery selection | |
| JP5289211B2 (en) | Image search system, image search program, and server device | |
| Chen et al. | Context-aware discriminative vocabulary learning for mobile landmark recognition | |
| CN102253935A (en) | System and method for classifying photographs | |
| Constantinou et al. | Spatial keyframe extraction of mobile videos for efficient object detection at the edge | |
| Lee et al. | Effective car video retrieval using feature matching in a mobile video cloud | |
| Zhong et al. | i-Memory: An intelligence Android-based photo management system | |
| Vo et al. | WhereAmI: Energy efficient positioning using partial textual signatures | |
| Cheng et al. | OSNI: Searching for Needles in a Haystack of Social Network Data. | |
| Doong | Predicting the popularity of internet memes with Hilbert-Huang spectrum |
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:20140402 |