CN101996212A - Image search system and method - Google Patents
Image search system and method Download PDFInfo
- Publication number
- CN101996212A CN101996212A CN2009101694108A CN200910169410A CN101996212A CN 101996212 A CN101996212 A CN 101996212A CN 2009101694108 A CN2009101694108 A CN 2009101694108A CN 200910169410 A CN200910169410 A CN 200910169410A CN 101996212 A CN101996212 A CN 101996212A
- Authority
- CN
- China
- Prior art keywords
- picture
- rgb value
- search condition
- search
- value range
- 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 discloses an image search system and an image search method. A search condition is analyzed to be an RGB range value, fuzzy search is performed according to the RGB range value so as to acquire an image which accords with the RGB range value, the color structure of the image per se is used as the search condition, and the efficiency for searching the matched image is improved.
Description
Technical field
A kind of image searching system and method thereof are meant a kind of system and method thereof of carrying out picture searching with rgb value especially.
Background technology
Search is a kind of instrument that is commonly used to look for data, and the user can directly key in search condition by data search, and search meets or near the data of search condition in specified database, and the result is presented on user's computing machine.
Arriving along with digital Age, the widespread use of digital camera, thereby produce a large amount of pictures and be stored in electronic installation or the various internal memory, along with the picture number that stores constantly increases, when the user seeks the picture of wanting according to demand, just need a kind of can the convenience to search out the function that meets picture fast.In general, user's search pictures all is to search for picture name, for example, when the user wants to look for the picture that content is a sky, user's input " sky " is searched for, but if picture name is not with " sky " when naming, it is the picture of sky that the user can't find content, and the user can only browse search one by one, can't directly search image content, efficient is extremely low, even the picture that possibly can't find the user to need.Present technology can only be searched for picture name as way of search, if the user wishes to seek the picture that content is the particular color structure, then can't obtain the picture that needs, therefore cause is that title with picture is searched in the past, often causes the problem of the picture that search wants less than the user.
In sum, to have existed since the midium or long term be to search for the title of picture to prior art always in the past as can be known, often causes the problem of the picture that search wants less than the user, therefore is necessary to propose improved technological means, solves this problem.
Summary of the invention
Because prior art exists was to search for the title of picture in the past, often caused the problem of the picture that search wants less than the user, and the present invention discloses a kind of image searching system and method thereof then, wherein:
Image searching system disclosed in this invention comprises picture database, receiver module, parsing module, retrieval module and display module at least, and wherein picture database is in order to store plural pictures; Receiver module is in order to receive search condition; Parsing module is to be corresponding at least one RGB value range in order to resolve search condition; Retrieval module is to picture database picture to be carried out fuzzy search with image recognition in order to foundation RGB value range, to be met the picture of RGB value range simultaneously; Display module is the picture that meets in order to demonstration.
Image searching method disclosed in this invention, its step comprise at least sets up picture database, and picture database is in order to store plural pictures; Receive search condition; Resolving search condition is corresponding at least one RGB value range; To picture database, picture is carried out fuzzy search with image recognition according to the RGB value range, to be met the picture of RGB value range simultaneously; Show those pictures that meet.
System and method for disclosed in this invention as above, and after the difference between the prior art is that the present invention is the RGB value range by the parsing search condition, carry out fuzzy search according to the RGB value range, to obtain meeting the picture of RGB value range, can reach with picture intrinsic colour structure as search condition, improve and search the efficient that meets picture.
Description of drawings
Fig. 1 is an image searching system block scheme of the present invention;
Fig. 2 is an image searching method process flow diagram of the present invention;
Fig. 3 and Fig. 4 are the image searching system first embodiment synoptic diagram of the present invention;
Fig. 5 is the image searching system second embodiment synoptic diagram of the present invention.
[primary clustering symbol description]
100 image searching systems
110 picture databases
120 receiver modules
130 parsing modules
140 retrieval modules
150 display modules
300 search pictures query word input interfaces
310 query words input block
The 400 RGB tables of comparisons
410 color words
420 RGB value ranges
500 inquiry picture input interfaces
510 inquiry picture viewing area pieces
520 navigation mechanism
530 input modules
540 inquiry pictures
Embodiment
Below will cooperate graphic and embodiment describes embodiments of the present invention in detail, whereby to the present invention how the application technology means implementation procedure that solves technical matters and reach technique effect can fully understand and implement according to this.
See also Fig. 1, Fig. 1 is an image searching system block scheme of the present invention, and image searching system 100 comprises at least: picture database 110, receiver module 120, parsing module 130, retrieval module 140 and display module 150.
It is corresponding RGB value range that parsing module 130 is responsible for resolving search condition, and just after receiver module 120 received search condition, parsing module 130 can be resolved search condition to obtain corresponding RGB value range.It should be noted that, when if the search condition that receiver module 120 is received is query word, parsing module 130 can be according to the RGB table of comparisons parsing query word of tabling look-up, just parsing module 130 can correspond to query word and carry out the comparison of tabling look-up of literal and RGB value range in the RGB table of comparisons, to obtain corresponding RGB value range, the form of the described RGB table of comparisons for example is but is not defined as " yellow: (230~255,230~255,0~70) ".For instance, when if the search condition that received of receiver module 130 is query word " yellow ", parsing module 130 parsings of search condition " yellow " can being tabled look-up to the RGB table of comparisons, the RGB value range that can obtain correspondence is (230~255,230~255,0~70).
It should be noted that if the search condition that receiver module 120 received is that parsing module 130 can be resolved the inquiry picture when inquiring about picture with image recognition, to obtain corresponding RGB value range.The described parsing with image recognition inquired about the RGB value range that picture obtains correspondence, be meant and inquire about picture obtains inquiring about each pixel of picture with image recognition rgb value earlier, the rgb value of all pixels that will obtain is resolved again, for example be to blur comparison with the rgb value statistical induction of all pixels and to RGB value range deck watch, can obtain corresponding RGB value range in the mode of statistical induction.For instance, when if the search condition that receiver module 120 received is inquiry picture " A ", parsing module 130 will be inquired about picture " A " and carry out the rgb value that each pixel of inquiry picture " A " is resolved in image recognition, conclude by statistics and to RGB value range deck watch blur compare after, can obtain corresponding RGB value range and be (0~70,40~255,245~255) and (245~255,245~255,245~255).
For instance, be (230~255,230~255 if parsing module 130 is resolved the RGB value range that obtains, 0~70), retrieval module 140 carries out fuzzy search with picture with image recognition according to RGB value range (230~255,230~255,0~70) in picture database 110, if the rgb value in the picture database 110 drops on RGB value range (230~255,230~255,0~70) picture can be met RGB value range (230~255 simultaneously, 230~255,0~70) picture.
Move by image searching system, to resolve after search condition is the RGB value range, carry out fuzzy search according to the RGB value range, to obtain meeting the picture of RGB value range, can solve was that title with picture is searched in the past, often caused the problem of the picture that search wants less than the user.
The inventive method flow process of below will arranging in pairs or groups embodiment that develops simultaneously specifies the present invention, and so it is not in order to limiting the present invention, and those skilled in the art can do modification to following embodiment slightly according to spirit of the present invention, but it still belongs to scope of the present invention.
Figure 2 shows that image searching method process flow diagram of the present invention, Fig. 3 to Fig. 4 is the image searching system first embodiment synoptic diagram of the present invention.Below will cooperate Fig. 2 to Fig. 4 that the image searching system operating procedure of the present invention and first embodiment are described simultaneously.First embodiment of the invention be with the expression color query word as search condition.
See also Fig. 4, when image searching system 100 receives query word " yellow ", after " green ", image searching system 100 can be resolved the query word " yellow " of search condition, " green ", image searching system 100 can be tabled look-up according to the RGB table of comparisons 400 and be resolved query word " yellow ", " green ", have color words 410 and RGB value range 420 in the RGB table of comparisons 400, color words 410 for example is " yellow ", " green ", RGB value range 420 for example is (230~255,230~255,0~70), (0~70,200~255,0~70), just image searching system 100 can be with query word " yellow ", whether " green " tables look-up to the RGB table of comparisons 400 has and query word " yellow ", the color words 410 " yellow " that " green " conforms to, " green ", to obtain the corresponding RGB value range of query word " yellow " 420 for (230~255,230~255,0~70), the RGB value range 420 that query word " green " is corresponding is (0~70,230~255,0~70) (step 203).
When image searching system 100 obtains RGB value range 420 (230~255,230~255,0~70), (0~70,230~255,0~70) after, image searching system 100 can read all pictures the rgb value of picture to the picture database in the image recognition mode, and according to RGB value range 420 (230~255,230~255,0~70), (0~70,230~255,0~70) carries out fuzzy search, so-called fuzzy search is exactly that the rgb value that has more than 80% of the rgb value when picture drops on RGB value range 420 (230~255,230~255,0~70), (0~70,230~255,0~70) in the time of in, represents that then picture meets RGB value range 420 (230~255,230~255,0~70), (0~70,230~255,0~70), therefore met RGB value range 420 (230~255 simultaneously, 230~255,0~70), the picture of (0~70,230~255,0~70) for example is picture " 0302 ", " 1694 ", " 2137 " (step 204).Last image searching system 100 can show picture " 0302 ", " 1694 ", " 2137 " (step 205) that meets at the picture searching display interface.
Figure 2 shows that image searching method process flow diagram of the present invention, Fig. 5 is the image searching system second embodiment synoptic diagram of the present invention.Below will cooperate Fig. 2 and Fig. 5 that the image searching system operating procedure of the present invention and second embodiment are described simultaneously.Second embodiment of the invention is as search condition with the inquiry picture.
After image searching system 100 receives inquiry picture 540, image searching system 100 can be resolved inquiry picture 540 with image recognition, just image searching system 100 can carry out inquiry picture 540 obtaining inquiring about after the image recognition rgb value of picture 540, just obtain inquiring about the rgb value of picture 540 each pixel with image recognition inquiry picture 540, the rgb value of all pixels that will obtain is resolved again, blur comparison in the mode of statistical induction with the rgb value statistical induction of all pixels and to RGB value range deck watch, can obtain the RGB value range after resolving for example is (230~255,0~70,0~70) (step 203).
When image searching system 100 obtains RGB value range (230~255,0~70,0~70) after, image searching system 100 can read all pictures the rgb value of picture to the picture database in the image recognition mode, and according to RGB value range (230~255,0~70,0~70) carries out fuzzy search, so-called fuzzy search is exactly that the rgb value that has more than 80% of the rgb value when picture drops on RGB value range (230~255,0~70,0~70) in the time of in, represents that then picture meets RGB value range (230~255,0~70,0~70), therefore the picture that is met RGB value range (230~255,0~70,0~70) simultaneously for example is picture " 0302 ", " 1694 ", " 2137 " (step 204).Last image searching system 100 can show picture " 0302 ", " 1694 ", " 2137 " (step 205) that meets at the picture searching display interface.
In sum, difference between the present invention and the prior art is to have and resolves after search condition is the RGB value range as can be known, carry out fuzzy search according to the RGB value range, technological means with the picture that obtains meeting the RGB value range, can solve prior art existing by this technological means was to search for the title of picture in the past, often cause the problem of the picture that search wants less than the user, and then reach with picture intrinsic colour structure as search condition, improve and search the efficient that meets picture.
Though embodiment disclosed in this invention as above, described content is not in order to direct qualification scope of patent protection of the present invention.Any the technical staff in the technical field of the invention under the prerequisite that does not break away from spirit and scope disclosed in this invention, can do a little change what implement in form and on the details.Scope of patent protection of the present invention, that still must be defined with appended claim scope is as the criterion.
Claims (8)
1. an image searching system is characterized in that, comprises at least:
One picture database is in order to store plural pictures;
One receiver module is in order to receive a search condition;
One parsing module is corresponding at least one RGB value range in order to resolve this search condition;
One retrieval module is in order to carry out fuzzy search with image recognition according to this RGB value range, to be met the picture of those RGB value ranges simultaneously to picture database; And
One display module, the picture that meets in order to demonstration.
2. image searching system as claimed in claim 1 is characterized in that, this search condition is at least one query word of expression color, and this parsing module is to resolve those query words according to a RGB table of comparisons.
3. image searching system as claimed in claim 1 is characterized in that, this search condition is an inquiry picture, and this parsing module is to resolve this inquiry picture with image recognition.
4. image searching system as claimed in claim 1 is characterized in that, those pictures of this picture database are to obtain from network download.
5. an image searching method is characterized in that, comprises the following step at least:
Set up a picture database, this picture database is in order to store plural pictures;
Receive a search condition;
Resolve this search condition and be corresponding at least one RGB value range;
To this picture database, those pictures are carried out fuzzy search with image recognition according to those RGB value ranges, to be met those pictures of those RGB value ranges simultaneously; And
Show those pictures that meet.
6. image searching method as claimed in claim 5, it is characterized in that, this search condition is at least one query word of expression color, and resolving this search condition is to table look-up according to a RGB table of comparisons to resolve those query words for the step of those corresponding RGB value ranges.
7. image searching method as claimed in claim 5 is characterized in that, this search condition is an inquiry picture, and resolving this search condition is to resolve this inquiry picture with image recognition for the step of those corresponding RGB value ranges.
8. image searching method as claimed in claim 5 is characterized in that, those pictures of this picture database are to obtain from network download.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009101694108A CN101996212A (en) | 2009-08-31 | 2009-08-31 | Image search system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009101694108A CN101996212A (en) | 2009-08-31 | 2009-08-31 | Image search system and method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101996212A true CN101996212A (en) | 2011-03-30 |
Family
ID=43786374
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2009101694108A Pending CN101996212A (en) | 2009-08-31 | 2009-08-31 | Image search system and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101996212A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104298786A (en) * | 2014-11-12 | 2015-01-21 | 广州出益信息科技有限公司 | Image retrieval method and device |
CN105354275A (en) * | 2015-10-29 | 2016-02-24 | 努比亚技术有限公司 | Information processing method and apparatus, and terminal |
CN106610987A (en) * | 2015-10-22 | 2017-05-03 | 杭州海康威视数字技术股份有限公司 | Video image retrieval method, device and system |
-
2009
- 2009-08-31 CN CN2009101694108A patent/CN101996212A/en active Pending
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104298786A (en) * | 2014-11-12 | 2015-01-21 | 广州出益信息科技有限公司 | Image retrieval method and device |
CN104298786B (en) * | 2014-11-12 | 2018-07-10 | 广州出益信息科技有限公司 | A kind of image search method and device |
CN106610987A (en) * | 2015-10-22 | 2017-05-03 | 杭州海康威视数字技术股份有限公司 | Video image retrieval method, device and system |
CN106610987B (en) * | 2015-10-22 | 2020-02-21 | 杭州海康威视数字技术股份有限公司 | Video image retrieval method, device and system |
CN105354275A (en) * | 2015-10-29 | 2016-02-24 | 努比亚技术有限公司 | Information processing method and apparatus, and terminal |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20070050337A1 (en) | Method and system for dynamically processing ambiguous, reduced text search queries and highlighting results thereof | |
US8307391B2 (en) | Information processing apparatus and method | |
CN102236423A (en) | Automatic character supplementation method, device and input method system | |
US20030126126A1 (en) | Apparatus and method for searching multimedia object | |
US20070061321A1 (en) | Method and system for processing ambiguous, multi-term search queries | |
CN105653700A (en) | Video search method and system | |
US7199729B2 (en) | Character code conversion methods and systems | |
CN100502493C (en) | Digital television receiver program searching method | |
CN106384108A (en) | Text content retrieval method, word interpreting device and mobile terminal | |
CN101853297A (en) | Method for fast obtaining expected image in electronic equipment | |
CN101676911A (en) | Contact person searching system and method | |
CN101996212A (en) | Image search system and method | |
CN103186587A (en) | Method for quickly translating English word of book through mobile phone | |
US20030174833A1 (en) | Dialing method with phone number retrieval using partial pattern matching | |
CN102055932A (en) | Method for searching television program and television set using same | |
CN1538329A (en) | Searching method of calalogue of stored items and its device | |
CN101261645B (en) | Method and apparatus for obtaining multiple layer information | |
CN100538704C (en) | The interactive inverse-checking method of electronic dictionary fast skipping and device thereof | |
CN101938551A (en) | Fuzzy inquiry method and device for phonebook of mobile terminal | |
US8156133B2 (en) | Modifying an electronic graphics file to be searchable according to annotation information | |
CN106850923B (en) | contact information display method, system and equipment | |
US20080208793A1 (en) | Image file searching method and electronic device | |
CN100442275C (en) | Method and system for indentifying Chinese address data | |
CN101425057A (en) | English word enquiring system and method thereof | |
CN110362703B (en) | Picture arranging method and system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20110330 |