CN107103073A - A kind of image indexing system - Google Patents
A kind of image indexing system Download PDFInfo
- Publication number
- CN107103073A CN107103073A CN201710267385.1A CN201710267385A CN107103073A CN 107103073 A CN107103073 A CN 107103073A CN 201710267385 A CN201710267385 A CN 201710267385A CN 107103073 A CN107103073 A CN 107103073A
- Authority
- CN
- China
- Prior art keywords
- image
- feature
- retrieval
- extraction
- output end
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/51—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/5838—Retrieval 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)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Library & Information Science (AREA)
- Software Systems (AREA)
- Processing Or Creating Images (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of image indexing system, including query image sample, the first local feature extracted in image library, anti-erroneous judgement module, the second local feature, safety control module, image retrieval and the safety of image display extracted in image library;Database is divided into multiple subdata bases by the present invention in advance by the utilization of keyword and mark, is first retrieved during retrieval in the high subdata base of the degree of correlation, reduces amount of calculation, improve arithmetic speed;When representing image based on word bag, it is proposed that weighting is represented and First look similarity, reduces time overhead;Feature based combination table diagram as when, make use of the space inclusion relation between local feature, propose to combine related local feature to strengthen its visual expression ability;This feature combination not only has good yardstick and rotational invariance, but also can carry out local geometric verification using the relative position information between each characteristic element naturally, rejects erroneous matching that may be present.
Description
Technical field
The present invention relates to a kind of image indexing system, specifically a kind of image indexing system.
Background technology
The image retrieval technologies of early stage are mainly based upon the search of keyword, and user is retouched by inputting to query image
State, corresponding retrieval result is found in searching database;With the arrival in internet and big data epoch, based on keyword
Image retrieval technologies have been not applied for the searching database of huge volumes of content, real-time extension;CBIR refers to
User provides query image, and searching algorithm calculates image feature representation, similar picture material and return are searched in search library
As a result retrieval technique is with the rise of the technologies such as mobile Internet, Internet of Things, and data volume rapidly increases in global range, big number
According to the epoch oneself through arrive;Along with the generation of big data, it plays extremely important make in modern society and economic activity
With;Information security mainly includes the content of following five aspects, i.e., need to ensure the confidentiality of information, authenticity, integrality, unauthorized
The security of copy and institute's parasitic system;The scope that information security includes in itself is very big, including how taking precautions against commercial enterprise
Secret leakage, teenager is taken precautions against to the browsing of flame, leakage of personal information etc.;Information security system under network environment
It is the key ensured information security, including computer security operating system, various security protocols, security mechanism, until safety system
System, global safety just can be threatened simply by the presence of security breaches;Information security refers to that information system is protected, not by accidental
Or the reason for malice and by destroying, change, reveal, system is continuously reliably normally run, and information service is not interrupted, finally
Realize business continuance;Large-scale data gives traditional multimedia research, and the application and research for being based especially on image are brought
New challenge and opportunity;And image erroneous judgement has had a strong impact on image retrieval.
The content of the invention
It is an object of the invention to provide a kind of image indexing system, to solve the problems mentioned in the above background technology.
To achieve the above object, the present invention provides following technical scheme:
A kind of image indexing system, including query image sample, the first local feature in extraction image library, anti-erroneous judgement mould
Block, the second local feature, safety control module, image retrieval and the safety of image display extracted in image library;The inquiry
Image pattern is the starting stage inquired about for image pattern, and preliminary observation is carried out to image;In the extraction image library
First local feature is connected to the output end of query image sample, for extracting the local feature on image, and is next step
Image retrieval is prepared;The anti-erroneous judgement module is by information classification, feature extraction, the first graphical representation based on word bag, the
One feature based combination image represents to constitute with image retrieval;Described information classification is connected to the first part extracted in image library
The output end of feature;For the first local feature extracted in image library to be carried out into classification processing;The feature extraction is arranged on
The output end of classification information, the characteristics of image in classification information is further extracted, the image retrieval after being easy to;Described
One graphical representation based on word bag is arranged on the output end of feature extraction, for by the feature of extraction, with word bag to image
It is indicated, and is easy to retrieve the image;The first feature based combination image represents to be arranged on feature extraction
Output end, the feature of extraction is combined, the image retrieval after being easy to;The second local feature in the extraction image library
The output end of anti-erroneous judgement module is arranged on, for extracting the second local feature on the basis of the first local feature, to ensure inspection
The image that rope comes out more coincide with original image;The safety control module is arranged on the second local spy extracted in image library
The output end levied, safety control module is based on by the graphical representation of information classification, feature extraction, second based on word bag and second
Combinations of features graphical representation is constituted;The output end of second local feature of the described information classification setting in image library is extracted, is used
Classified in by the extract second local feature, the retrieval after being easy to, and enable to the image and artwork of retrieval
As more coincideing;The feature of extracting is arranged on the output end of information classification, for the characteristics of image after sorted again
Extraction feature is carried out, to ensure that the image after retrieval is more safe and reliable;Described second graphical representation based on word bag
The output end of feature extraction is arranged on, to guarantee in the form of word, image is described.
It is used as further scheme of the invention:Described information secure display device receives the carrying that third-party platform is sent
The confirmation request message of two retrieval information, the second retrieval information is taken after being utilized the customer digital certificate public key encryption of user
Band is in the confirmation request message.
It is used as further scheme of the invention:Described image security display is arranged on the output end of image retrieval, is used for
Safety shows the image being retrieved, and is easy to retrieval personnel to contrast the image being retrieved with original image, in order to
Retrieve the image of needs.
It is used as further scheme of the invention:The second feature based combination image represents that being similarly disposed at feature carries
The output end taken, for the feature of extraction to be combined, the image after being easy to is retrieved so that based on word
Retrieval result is identical with image searching result.
Compared with prior art, the beneficial effects of the invention are as follows:
Database is divided into multiple subdata bases by the present invention in advance by the utilization of keyword and mark, first during retrieval
Retrieved in the high subdata base of the degree of correlation, reduce amount of calculation, improve arithmetic speed;Image is represented based on word bag
When, it is proposed that weighting is represented and First look similarity, reduces time overhead;Feature based combination table diagram as when, utilize
Space inclusion relation between local feature, proposes to combine related local feature to strengthen its visual expression energy
Power;This feature combination not only have good yardstick and rotational invariance, but also can naturally using each characteristic element it
Between relative position information carry out local geometric verification, erroneous matching that may be present is rejected, to improve the accurate of image retrieval
Rate;The precision of retrieval has been significantly increased in characteristic quantification and combinational expression.
Brief description of the drawings
Fig. 1 is the structural representation of image indexing system.
Fig. 2 is the structural representation of anti-erroneous judgement module in image indexing system.
Fig. 3 is the structural representation of safety control module in image indexing system.
Embodiment
The technical scheme of this patent is described in more detail with reference to embodiment.
Refer to the first part in Fig. 1-3, a kind of image indexing system, including query image sample, extraction image library
Feature, anti-erroneous judgement module, the second local feature, safety control module, image retrieval and the safety of image that extract in image library are aobvious
Show device;The query image sample is the starting stage inquired about for image pattern, and preliminary observation is carried out to image;It is described to carry
The first local feature in image library is taken to be connected to the output end of query image sample, for extracting the local feature on image,
And prepared for the image retrieval of next step;The anti-erroneous judgement module is by information classification, feature extraction, first based on word bag
Graphical representation, the first feature based combination image are represented and image retrieval composition;Described information classification is connected to extraction image library
In the first local feature output end;For the first local feature extracted in image library to be carried out into classification processing;The spy
The output end extracted and be arranged on classification information is levied, the characteristics of image in classification information is further extracted, the figure after being easy to
As retrieval;Described first graphical representation based on word bag is arranged on the output end of feature extraction, for by the feature of extraction, using
Word bag is indicated to image, and is easy to retrieve the image;The first feature based combination image represents to set
In the output end of feature extraction, the feature of extraction is combined, the image retrieval after being easy to;In the extraction image library
Second local feature is arranged on the output end of anti-erroneous judgement module, local special for extracting second on the basis of the first local feature
Levy, to ensure that the image being retrieved and original image more coincide;The safety control module is arranged in extraction image library
The second local feature output end, safety control module is by information classification, feature extraction, the second image table based on word bag
Show and represent composition with the second feature based combination image;Second local feature of the described information classification setting in image library is extracted
Output end, for the extract second local feature to be classified, the retrieval after being easy to, and enable to retrieval
Image more coincide with original image;The feature of extracting is arranged on the output end of information classification, for the figure after sorted
As feature carries out extraction feature again, to ensure that the image after retrieval is more safe and reliable;Described second is based on word bag
Graphical representation be arranged on the output end of feature extraction, to guarantee in the form of word, image is described.
The better embodiment to this patent is explained in detail above, but this patent is not limited to above-mentioned embodiment party
, can also be on the premise of this patent objective not be departed from formula, the knowledge that one skilled in the relevant art possesses
Make a variety of changes.
Claims (4)
1. a kind of image indexing system, including query image sample, the first local feature in extraction image library, anti-erroneous judgement mould
Block, the second local feature, safety control module, image retrieval and the safety of image display extracted in image library;Its feature exists
In the query image sample is the starting stage inquired about for image pattern, and preliminary observation is carried out to image;It is described to extract
The first local feature in image library is connected to the output end of query image sample, for extracting the local feature on image, and
Prepared for the image retrieval of next step;The anti-erroneous judgement module is by information classification, feature extraction, the first figure based on word bag
As representing, the first feature based combination image is represented and image retrieval is constituted;Described information classification is connected in extraction image library
The first local feature output end;For the first local feature extracted in image library to be carried out into classification processing;The feature
The output end for being arranged on classification information is extracted, the characteristics of image in classification information is further extracted, the image after being easy to
Retrieval;Described first graphical representation based on word bag is arranged on the output end of feature extraction, for by the feature of extraction, with list
Word bag is indicated to image, and is easy to retrieve the image;The first feature based combination image represents to be arranged on
The output end of feature extraction, the feature of extraction is combined, the image retrieval after being easy to;In the extraction image library
Two local features are arranged on the output end of anti-erroneous judgement module, local special for extracting second on the basis of the first local feature
Levy, to ensure that the image being retrieved and original image more coincide;The safety control module is arranged in extraction image library
The second local feature output end, safety control module is by information classification, feature extraction, the second image table based on word bag
Show and represent composition with the second feature based combination image;Second local feature of the described information classification setting in image library is extracted
Output end, for the extract second local feature to be classified, the retrieval after being easy to, and enable to retrieval
Image more coincide with original image;The feature of extracting is arranged on the output end of information classification, for the figure after sorted
As feature carries out extraction feature again, to ensure that the image after retrieval is more safe and reliable;Described second is based on word bag
Graphical representation be arranged on the output end of feature extraction, to guarantee in the form of word, image is described.
2. image indexing system according to claim 1, it is characterised in that described information secure display device receives the 3rd
The confirmation request message of information is retrieved in the carrying second that Fang Pingtai is sent, and the second retrieval information is utilized the number of users of user
Carried after the encryption of word CertPubKey in the confirmation request message.
3. image indexing system according to claim 1, it is characterised in that described image security display is arranged on image
The output end of retrieval, the image being retrieved is shown for safety, is easy to retrieval personnel to the image and artwork that are retrieved
As being contrasted, in order to retrieve the image of needs.
4. image indexing system according to claim 1, it is characterised in that the second feature based combination image is represented
The output end of feature extraction is similarly disposed at, for the feature of extraction to be combined, the image after being easy to is retrieved, and is made
The retrieval result being able to based on word is identical with image searching result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710267385.1A CN107103073A (en) | 2017-04-21 | 2017-04-21 | A kind of image indexing system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710267385.1A CN107103073A (en) | 2017-04-21 | 2017-04-21 | A kind of image indexing system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107103073A true CN107103073A (en) | 2017-08-29 |
Family
ID=59656389
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710267385.1A Pending CN107103073A (en) | 2017-04-21 | 2017-04-21 | A kind of image indexing system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107103073A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109697239A (en) * | 2018-11-23 | 2019-04-30 | 连尚(新昌)网络科技有限公司 | Method for generating the method for graph text information and for generating image data base |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106156349A (en) * | 2016-07-22 | 2016-11-23 | 黎海纤 | Image search method based on information security |
CN106203165A (en) * | 2016-07-01 | 2016-12-07 | 何钟柱 | The big data analysis method for supporting of information based on credible cloud computing |
CN106227818A (en) * | 2016-07-22 | 2016-12-14 | 黎海纤 | A kind of abnormal image data process and search method |
-
2017
- 2017-04-21 CN CN201710267385.1A patent/CN107103073A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106203165A (en) * | 2016-07-01 | 2016-12-07 | 何钟柱 | The big data analysis method for supporting of information based on credible cloud computing |
CN106156349A (en) * | 2016-07-22 | 2016-11-23 | 黎海纤 | Image search method based on information security |
CN106227818A (en) * | 2016-07-22 | 2016-12-14 | 黎海纤 | A kind of abnormal image data process and search method |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109697239A (en) * | 2018-11-23 | 2019-04-30 | 连尚(新昌)网络科技有限公司 | Method for generating the method for graph text information and for generating image data base |
CN109697239B (en) * | 2018-11-23 | 2022-05-31 | 南京尚网网络科技有限公司 | Method for generating teletext information |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106951411B (en) | The quick multi-key word Semantic Ranking searching method of data-privacy is protected in a kind of cloud computing | |
Zhang et al. | A domain-feature enhanced classification model for the detection of Chinese phishing e-Business websites | |
US11797620B2 (en) | Expert detection in social networks | |
US8438174B2 (en) | Automated forensic document signatures | |
CN109063509A (en) | It is a kind of that encryption method can search for based on keywords semantics sequence | |
CN108959478A (en) | Ciphertext image search method and system under a kind of cloud environment | |
Wenyin et al. | Phishing Web page detection | |
Alsmadi et al. | Online integrity and authentication checking for Quran electronic versions | |
CN102629261A (en) | Method for finding landing page from phishing page | |
CN102262670A (en) | Cross-media information retrieval system and method based on mobile visual equipment | |
Wang et al. | Who are in the darknet? Measurement and analysis of darknet person attributes | |
Paraschiv et al. | A unified graph-based approach to disinformation detection using contextual and semantic relations | |
Brenner et al. | Discovery of single-vendor marketplace operators in the Tor-Network | |
CN107103073A (en) | A kind of image indexing system | |
CN108549701A (en) | Cloud environment encrypts outsourcing data semantic extended search method and system | |
CN106156349A (en) | Image search method based on information security | |
Nattuthurai et al. | Content Analysis of Dark Net Academic Journals from 2010–2017 Using KH Coder | |
TWM523901U (en) | Search engine device for performing semantic keyword analysis | |
US11757916B1 (en) | Methods and apparatus for analyzing and scoring digital risks | |
Wang et al. | Achieve web search privacy by obfuscation | |
Huang et al. | Efficient privacy-preserving content-based image retrieval in the cloud | |
Yanagi et al. | Interactive re-ranking for cross-modal retrieval based on object-wise question answering | |
Jiang et al. | A classification model for detection of Chinese phishing e-business websites | |
Ma et al. | Preserving privacy on the searchable internet | |
Pendergrass | The intersection of human trafficking and technology |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
CB02 | Change of applicant information | ||
CB02 | Change of applicant information |
Address after: 100070, 6, building 2, building 3, Ying Ying century East Road, Fengtai District Automotive Museum, Beijing, 701-1 Applicant after: Beijing August melon Technology Co., Ltd Address before: 100070, 6, building 2, building 3, Ying Ying century East Road, Fengtai District Automotive Museum, Beijing, 701-1 Applicant before: BEIJING HENGGUAN NETWORK DATA TREATMENT CO., LTD. |
|
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170829 |