CN106372068A - Method and device for image search, and terminal - Google Patents

Method and device for image search, and terminal Download PDF

Info

Publication number
CN106372068A
CN106372068A CN201510427150.5A CN201510427150A CN106372068A CN 106372068 A CN106372068 A CN 106372068A CN 201510427150 A CN201510427150 A CN 201510427150A CN 106372068 A CN106372068 A CN 106372068A
Authority
CN
China
Prior art keywords
image
user
picture
face
target
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
Application number
CN201510427150.5A
Other languages
Chinese (zh)
Inventor
邢雪峰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
ZTE Corp
Original Assignee
ZTE Corp
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by ZTE Corp filed Critical ZTE Corp
Priority to CN201510427150.5A priority Critical patent/CN106372068A/en
Publication of CN106372068A publication Critical patent/CN106372068A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G06F16/5838Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using colour

Landscapes

  • Engineering & Computer Science (AREA)
  • Library & Information Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Processing Or Creating Images (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention provides a method and device for image search, and a terminal. The method comprises the steps that image features input by a user are acquired; a target image matched with the image features is searched in a database; and the target image is displayed. Through implementation of the invention, the matched target image can be searched in the database according to the image features input by the user and can then be displayed, so that all the target images displayed by the terminal to the user are matched with the image features input by the user, and thus the user can rapidly find photos containing specific figures or sceneries. Different from the prior art that all the images in the database are directly displayed to the user, the method and device provided by the invention avoid the defect that the user can determine the target image only after checking all the images one after another, solve the problem in inconvenience of the prior art for the image search, and enhance the user's use experience.

Description

A kind of image lookup method and device, terminal
Technical field
The present invention relates to image searches field, more particularly, to a kind of image lookup method and device, terminal.
Background technology
With the raising of mobile phone terminal photographic head configuration, and the extension of its store function and portability, it is got over More to become the favor of user.User uses mobile phone photograph more and more extensive, stores therefore inside user mobile phone Picture get more and more, how to be quickly found out that oneself required picture increasingly becomes user's headache Problem.The method of traditional lookup picture is the thumbnail being generated according to picture, and user passes through slip screen Mode, not only searches slow, and with duration, and thumbnail is too little, its characteristics of image again not exclusively, to user Cause very big inconvenience.
Therefore, how to provide a kind of image lookup method solving conventional images lookup technology inconvenience problem, It is those skilled in the art's technical problem urgently to be resolved hurrily.
Content of the invention
The invention provides a kind of image lookup method and device, terminal, to solve conventional images lookup technology Inconvenient problem.
The invention provides a kind of image lookup method, comprising: obtain the characteristics of image of user input;Look into Look for the target image with Image Feature Matching in data base;Show target image.
Further, the characteristics of image obtaining user input includes: the Image Acquisition image being selected according to user Feature.
Further, the Image Acquisition characteristics of image being selected according to user: the parsing selected image of user, Extract the image features that it comprises, selected according to user or default setting is from the characteristics of image extracting In parameter, select one or more or all as characteristics of image.
Further, characteristics of image includes face information, the target with Image Feature Matching in searching data storehouse Image includes: from the photo of terminal storage, picture, video, searches and comprises face information or comprise phase Like the target picture of image features spent more than threshold value, picture, video, using Query Result as target Image.
Further, also include: select the Part of photos taken training in data base, adjust grader, utilize and divide Class device is classified to pictures all in data base.
Further, also include after the classification: recognition of face is carried out to all images comprising face, obtains Take the face information of each image, represent and store face information.
The invention provides a kind of image searches device, comprising: acquisition module, for obtaining user input Characteristics of image;Searching modul, for the target image with Image Feature Matching in searching data storehouse;Show Module, for showing target image.
Further, acquisition module is used for the Image Acquisition characteristics of image selecting according to user.
Further, acquisition module is used for parsing the selected image of user, extracts the characteristics of image that it comprises Parameter, selects according to user or default setting is from the image features extracting, and selects one or many Individual or all as characteristics of image.
Further, characteristics of image includes face information, and searching modul is used for from the photo of terminal storage, figure In piece, video, search and comprise face information or comprise the mesh that similarity is more than the image features of threshold value Mark photo, picture, video, using Query Result as target image.
Further, also include sort module, divide for selecting the Part of photos taken in data base to train, adjusting Pictures all in data base are classified by class device using grader.
Further, also include parsing module, for after the classification, entering to all images comprising face Row recognition of face, obtains the face information of each image, marks and store face information.
The invention provides a kind of terminal, it includes the image of present invention offer and searches device.
Beneficial effects of the present invention:
The invention provides a kind of image lookup method, according to the characteristics of image of user input, in data base Search the target image of coupling, and show, such terminal display to the target image of user is all and user institute The Image Feature Matching of input, and then user just can quickly find and to comprise particular persons or scenery Photo, rather than directly image all of in data base is all showed user as prior art, it is to avoid User just can determine that the shortcoming of target image after checking one by one, solves conventional images lookup technology not Convenient problem, enhances the experience of user.
Brief description
Fig. 1 searches the structural representation of device for the image that first embodiment of the invention provides;
The flow chart of the image lookup method that Fig. 2 provides for second embodiment of the invention;
The flow chart of the image lookup method that Fig. 3 provides for third embodiment of the invention.
Specific embodiment
Now combined by specific embodiment and by way of accompanying drawing, the present invention is made with further annotation explanation.
First embodiment:
Fig. 1 searches the structural representation of device for the image that first embodiment of the invention provides, as shown in Figure 1, In the present embodiment, the image that the present invention provides is searched device 1 and is included:
Acquisition module 11, for obtaining the characteristics of image of user input;
Searching modul 12, for the target image with Image Feature Matching in searching data storehouse;
Display module 13, for showing target image.
In certain embodiments, the image that the acquisition module 11 in above-described embodiment is used for according to user selects obtains Take characteristics of image.
In certain embodiments, the acquisition module 11 in above-described embodiment is used for parsing the selected image of user, Extract the image features that it comprises, selected according to user or default setting is from the characteristics of image extracting In parameter, select one or more or all as characteristics of image.
In certain embodiments, the characteristics of image in above-described embodiment includes face information, and searching modul 12 is used Comprise face information or comprise similarity to be more than in, from the photo of terminal storage, picture, video, searching The target picture of the image features of threshold value, picture, video, using Query Result as target image.
In certain embodiments, as shown in figure 1, image in above-described embodiment searches device 1 also includes point Generic module 14, for selecting the Part of photos taken in data base to train, adjust grader, using grader pair In data base, all pictures are classified.
In certain embodiments, as shown in figure 1, the image in above-described embodiment searches device 1 also includes solution Analysis module 15, for after the classification, all images comprising face are carried out with recognition of face, obtains each The face information of image, marks and stores face information.
Corresponding, the invention provides a kind of terminal, it includes the image of present invention offer and searches device 1.
Second embodiment:
The flow chart of the image lookup method that Fig. 2 provides for second embodiment of the invention, as shown in Figure 2, In the present embodiment, the image lookup method that the present invention provides comprises the following steps:
S201: obtain the characteristics of image of user input;
S202: the target image with Image Feature Matching in searching data storehouse;
S203: show target image.
In certain embodiments, the characteristics of image of the acquisition user input in above-described embodiment includes: according to The Image Acquisition characteristics of image that family selects.
In certain embodiments, the Image Acquisition characteristics of image being selected according to user in above-described embodiment: solution The analysis selected image of user, extracts the image features that it comprises, and is selected according to user or acquiescence setting Put from the image features extracting, select one or more or all as characteristics of image.
In certain embodiments, the characteristics of image in above-described embodiment includes face information, in searching data storehouse Include with the target image of Image Feature Matching: from the photo of terminal storage, picture, video, search bag Containing face information or comprise similarity and be more than the target picture of image features of threshold value, picture, video, Using Query Result as target image.
In certain embodiments, the method in above-described embodiment also includes: selects the Part of photos taken in data base Pictures all in data base are classified by training, adjustment grader using grader.
In certain embodiments, the method in above-described embodiment also includes after the classification: comprises people to all The image of face carries out recognition of face, obtains the face information of each image, represents and store face information.
In above-mentioned all embodiments, involved data base includes but is not limited to: the figure of tag memory storage Database as data base, user's storage image beyond the clouds.
In above-mentioned all embodiments, involved characteristics of image includes but is not limited to: face information, wind Scape information, background information, camera information etc..
In conjunction with concrete application scene, the present invention is done with further annotation explanation.
3rd embodiment:
The present embodiment with image as photo, image information as face information as a example illustrate, Fig. 3 be this The flow chart of the image lookup method that bright 3rd embodiment provides, from the figure 3, it may be seen that in the present embodiment, this The image lookup method that invention provides comprises the following steps:
S301: set up training storehouse.
Select photo from data base, the photo of face and the photo of no face need to be included, set up and train storehouse. Is classified in training storehouse, be divided into training set and test set, using training set, grader be trained, make With test set, grader is adjusted, draws an optimal grader.
In actual applications, Face datection training can be instructed using now ripe a variety of methods Practice, now than faster and maturation five positions usually according to face for the algorithm (two eyes, nose, Two corners of the mouths) navigate to face information, then according to the hog/lbp characteristic vector conduct of these information retrievals The information of face., from hundreds of to tens of thousands of, dimension is higher for the digit of characteristic vector, and precision is higher, but meter Calculation amount is also bigger, for mobile phone, typically the dimension of characteristic vector is navigated to hundred.After training, The form of optimum faceform, characteristic information text is saved, forms feature database.
Preserve the Face Image Database training out and feature database, if any ripe feature database, this step can To omit.
S302: photo is carried out with classification parsing.
Mobile terminal storage face characteristic storehouse information, opens background program, carries out traversal to local picture and sweep Retouch, first Face datection is carried out to every pictures, find out all of face information in picture, i.e. face characteristic Five points position in the picture, then hog, lbp are converted into face according to the positional information of face Characteristic vector, a characteristic vector corresponds to a face information, and all of characteristic vector is saved, In view of scanning be one-off scanning and scan speed slow it is contemplated that open background scanning mode enter OK.
S303: terminal shows the face information of photo to user.
When user requires to look up all photo related to someone, user's choosing selects to comprise its certain photo (conjunction Shadow or independent photo), terminal is labeled the face information of this photo detecting, and presents to user, Mark form have multiple, can select on face on picture draw rectangle frame be labeled, literary composition may also be employed The mode of word is labeled, and the position of mark should facilitate user to be operated.
S304: obtain the face information of user input.
User selects a head portrait according to the information of mark, wants to search the picture related to this head portrait, after execution Continuous lookup action, the face information of the head portrait that user selects is as the search criterion inputting.
S305: terminal is searched and is shown relevant picture.
Terminal, according to the input of user, preserves search criterion, according to the face information of preservation, carries out similar Degree calculate it is considered to be feature be all multi-C vector, the mode of Similarity Measure can have multiple, can using than The method of the cosine value vectorial compared with two, that is, two vectorial cosine values are bigger, then its similarity is higher, Can be using comparing two vectorial distances, distance is nearer, then its similarity is higher, determines all similarities Higher picture, as relevant picture.
Terminal can show photo by way of thumbnail, while showing photo, by the people of each photo Face information also shows on its corresponding thumbnail, selects again for user, these users just can find Photo that is all and specifying user-dependent personnel.
The present embodiment is scanned to local picture according to face recognition algorithms, obtains face information, and people Face information is converted to the easy eigenvector information calculating and preserving, and these information are disposably preserved Come.And the face detecting is presented to user by certain form, user wants to search the figure of someone correlation Piece, then may be selected face and make a look up, terminal passes through Similarity Measure, returns the high picture of similarity to use Family.This invention can make user when mobile phone storage picture is more, can be quickly found out across file User wants the picture finding, and improves the experience of user.
In summary, by the enforcement of the present invention, at least there is following beneficial effect:
The invention provides a kind of image lookup method, according to the characteristics of image of user input, in data base Search the target image of coupling, and show, such terminal display to the target image of user is all and user institute The Image Feature Matching of input, and then user just can quickly find and to comprise particular persons or scenery Photo, rather than directly image all of in data base is all showed user as prior art, it is to avoid User just can determine that the shortcoming of target image after checking one by one, solves conventional images lookup technology not Convenient problem, enhances the experience of user.
The above is only the specific embodiment of the present invention, not the present invention done with any pro forma restriction, Every technical spirit according to the present invention embodiment of above is made arbitrarily simply modification, equivalent variations, In conjunction with or modify, all still fall within the protection domain of technical solution of the present invention.

Claims (13)

1. a kind of image lookup method is it is characterised in that include:
Obtain the characteristics of image of user input;
Target image with described image characteristic matching in searching data storehouse;
Show described target image.
2. image lookup method as claimed in claim 1 is it is characterised in that described acquisition user input Characteristics of image include: according to user select Image Acquisition described image feature.
3. image lookup method as claimed in claim 2 it is characterised in that described according to user's selection Image Acquisition described image feature: the parsing selected image of user, extract the characteristics of image ginseng that it comprises Number, selects according to user or default setting is from the image features extracting, and selects one or more Or all as described image feature.
4. the image lookup method as described in any one of claims 1 to 3 is it is characterised in that described figure As feature includes face information, include with the target image of described image characteristic matching in described searching data storehouse: From the photo of terminal storage, picture, video, search and comprise described face information or to comprise similarity big In the target picture of the image features of threshold value, picture, video, using Query Result as described target figure Picture.
5. image lookup method as claimed in claim 4 is it is characterised in that also include: selects described Part of photos taken training in data base, adjustment grader, using described grader to all in described data base Picture is classified.
6. image lookup method as claimed in claim 5 is it is characterised in that also include after the classification: All images comprising face are carried out with recognition of face, obtains the face information of each image, represent and store institute State face information.
7. a kind of image searches device it is characterised in that including:
Acquisition module, for obtaining the characteristics of image of user input;
Searching modul, for the target image with described image characteristic matching in searching data storehouse;
Display module, for showing described target image.
8. image as claimed in claim 7 searches device it is characterised in that described acquisition module is used for The Image Acquisition described image feature being selected according to user.
9. image as claimed in claim 8 searches device it is characterised in that described acquisition module is used for The parsing selected image of user, extracts the image features that it comprises, and is selected according to user or gives tacit consent to Setting, from the image features extracting, selects one or more or all as described image feature.
10. image as described in any one of claim 7 to 9 searches device it is characterised in that described figure As feature includes face information, described searching modul is used for from the photo of terminal storage, picture, video, Search comprise described face information or comprise similarity be more than threshold value the target picture of image features, Picture, video, using Query Result as described target image.
11. images as claimed in claim 10 search device it is characterised in that also including sort module, For selecting the Part of photos taken in described data base to train, adjust grader, using described grader to described In data base, all pictures are classified.
12. images as claimed in claim 11 search device it is characterised in that also including parsing module, For after the classification, all images comprising face are carried out with recognition of face, obtain the face letter of each image Breath, marks and stores described face information.
A kind of 13. terminals are it is characterised in that include the image as described in any one of claim 7 to 12 Search device.
CN201510427150.5A 2015-07-20 2015-07-20 Method and device for image search, and terminal Pending CN106372068A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510427150.5A CN106372068A (en) 2015-07-20 2015-07-20 Method and device for image search, and terminal

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510427150.5A CN106372068A (en) 2015-07-20 2015-07-20 Method and device for image search, and terminal

Publications (1)

Publication Number Publication Date
CN106372068A true CN106372068A (en) 2017-02-01

Family

ID=57880521

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510427150.5A Pending CN106372068A (en) 2015-07-20 2015-07-20 Method and device for image search, and terminal

Country Status (1)

Country Link
CN (1) CN106372068A (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107330859A (en) * 2017-06-30 2017-11-07 广东欧珀移动通信有限公司 A kind of image processing method, device, storage medium and terminal
CN107392138A (en) * 2017-07-18 2017-11-24 上海与德科技有限公司 A kind of display methods and device
CN108055400A (en) * 2017-12-20 2018-05-18 维沃移动通信有限公司 A kind of wallpaper switching method and mobile terminal
CN108304799A (en) * 2018-01-30 2018-07-20 广州市君望机器人自动化有限公司 A kind of face tracking methods
CN109359570A (en) * 2018-09-30 2019-02-19 广东***职业学院 A kind of method of video personage identification
CN109543579A (en) * 2018-11-14 2019-03-29 杭州登虹科技有限公司 Recognition methods, device and the storage medium of target object in a kind of image
WO2019075644A1 (en) * 2017-10-17 2019-04-25 深圳传音通讯有限公司 Portrait photograph searching method, and terminal
WO2019109246A1 (en) * 2017-12-05 2019-06-13 深圳前海达闼云端智能科技有限公司 Picture lookup method and device
CN111209425A (en) * 2020-01-06 2020-05-29 闻泰通讯股份有限公司 Image searching method and device, electronic equipment and computer readable storage medium
CN112101072A (en) * 2019-06-18 2020-12-18 北京陌陌信息技术有限公司 Face matching method, device, equipment and medium

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1716253A (en) * 2004-07-02 2006-01-04 佳能株式会社 Method and apparatus for retrieving data
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
CN103824051A (en) * 2014-02-17 2014-05-28 北京旷视科技有限公司 Local region matching-based face search method
CN104239336A (en) * 2013-06-19 2014-12-24 华为技术有限公司 Image screening method, device and terminal
US20150032719A1 (en) * 2008-06-05 2015-01-29 Enpulz, L.L.C. Search system employing multiple image based search processing approaches
CN104572905A (en) * 2014-12-26 2015-04-29 小米科技有限责任公司 Photo index creation method, photo searching method and devices

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1716253A (en) * 2004-07-02 2006-01-04 佳能株式会社 Method and apparatus for retrieving data
US20150032719A1 (en) * 2008-06-05 2015-01-29 Enpulz, L.L.C. Search system employing multiple image based search processing approaches
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
CN104239336A (en) * 2013-06-19 2014-12-24 华为技术有限公司 Image screening method, device and terminal
CN103824051A (en) * 2014-02-17 2014-05-28 北京旷视科技有限公司 Local region matching-based face search method
CN104572905A (en) * 2014-12-26 2015-04-29 小米科技有限责任公司 Photo index creation method, photo searching method and devices

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107330859A (en) * 2017-06-30 2017-11-07 广东欧珀移动通信有限公司 A kind of image processing method, device, storage medium and terminal
CN107330859B (en) * 2017-06-30 2021-06-15 Oppo广东移动通信有限公司 Image processing method and device, storage medium and terminal
CN107392138A (en) * 2017-07-18 2017-11-24 上海与德科技有限公司 A kind of display methods and device
WO2019075644A1 (en) * 2017-10-17 2019-04-25 深圳传音通讯有限公司 Portrait photograph searching method, and terminal
WO2019109246A1 (en) * 2017-12-05 2019-06-13 深圳前海达闼云端智能科技有限公司 Picture lookup method and device
CN108055400A (en) * 2017-12-20 2018-05-18 维沃移动通信有限公司 A kind of wallpaper switching method and mobile terminal
CN108304799A (en) * 2018-01-30 2018-07-20 广州市君望机器人自动化有限公司 A kind of face tracking methods
CN109359570A (en) * 2018-09-30 2019-02-19 广东***职业学院 A kind of method of video personage identification
CN109543579A (en) * 2018-11-14 2019-03-29 杭州登虹科技有限公司 Recognition methods, device and the storage medium of target object in a kind of image
CN112101072A (en) * 2019-06-18 2020-12-18 北京陌陌信息技术有限公司 Face matching method, device, equipment and medium
CN111209425A (en) * 2020-01-06 2020-05-29 闻泰通讯股份有限公司 Image searching method and device, electronic equipment and computer readable storage medium

Similar Documents

Publication Publication Date Title
CN106372068A (en) Method and device for image search, and terminal
CN109284729B (en) Method, device and medium for acquiring face recognition model training data based on video
US7711145B2 (en) Finding images with multiple people or objects
JP6966057B2 (en) Account information acquisition method, terminal, server, and system
JP5612310B2 (en) User interface for face recognition
Chen et al. Building book inventories using smartphones
US11704357B2 (en) Shape-based graphics search
US10594930B2 (en) Image enhancement and repair using sample data from other images
JP2009514107A (en) Determining a specific person from an aggregate
JP2007249588A (en) Face image registration device, method, program, and storage medium
TW201222288A (en) Image retrieving system and method and computer program product thereof
TWI586160B (en) Real time object scanning using a mobile phone and cloud-based visual search engine
TW201339867A (en) Video file search system and method
JP2010267180A (en) Image processing apparatus, electronic device, and program
WO2016145844A1 (en) Picture sorting method and corresponding picture storage and display device
KR101307325B1 (en) System for dual-searching image using region of interest set
KR101867497B1 (en) Picture processing method and electronic device
US11222208B2 (en) Portrait image evaluation based on aesthetics
US20150261752A1 (en) Personalized criteria-based media organization
Santos et al. RECOGNIZING AND EXPLORING AZULEJOS ON HISTORIC BUILDINGS’FACADES BY COMBINING COMPUTER VISION AND GEOLOCATION IN MOBILE AUGMENTED REALITY APPLICATIONS
Farhat et al. Intelligent portrait composition assistance: Integrating deep-learned models and photography idea retrieval
CN104199950B (en) A kind of scientific paper searching method based on image similarity Rapid matching
JPH10124655A (en) Device for preparing digital album and digital album device
US11283945B2 (en) Image processing apparatus, image processing method, program, and recording medium
CN114707055A (en) Photographing posture recommendation method integrating image content and feature extraction

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
SE01 Entry into force of request for 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: 20170201