CN107730357A - A kind of view-based access control model dictionary realizes the method and system of image quick-searching - Google Patents
A kind of view-based access control model dictionary realizes the method and system of image quick-searching Download PDFInfo
- Publication number
- CN107730357A CN107730357A CN201710978931.2A CN201710978931A CN107730357A CN 107730357 A CN107730357 A CN 107730357A CN 201710978931 A CN201710978931 A CN 201710978931A CN 107730357 A CN107730357 A CN 107730357A
- Authority
- CN
- China
- Prior art keywords
- image
- visual
- sample image
- sample
- characteristic vector
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0623—Item investigation
- G06Q30/0625—Directed, with specific intent or strategy
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Abstract
The invention discloses the method and system that a kind of view-based access control model dictionary realizes image quick-searching, and for solving the problems, such as that corresponding goods can not be quickly found out by the image of commodity in store, this method includes：S1：Sample image data collection is received, extracts all visual signatures of the sample image, and pass through all visual signatures of the sample image ID name identifications sample image；S2：The visual signature is quantized into sample image characteristic vector to generate visual dictionary storehouse；S3：The visual signature of image to be retrieved is extracted, and is quantized into image feature vector to be retrieved；S4：The characteristic vector of the image to be retrieved and the sample image characteristic vector in visual dictionary storehouse are contrasted, the sample image that output matches with the retrieval image.Using the present invention, corresponding commodity can be quickly found out by establishing visual dictionary storehouse, reduce the time that user finds.
Description
Technical field
The present invention relates to a kind of field of image search, more particularly to a kind of view-based access control model dictionary to realize image quick-searching
Method and system
Background technology
Under current fast-developing Internet technology and the background in multimedia messages epoch, people are for multimedia messages
Utilization obtained sufficient displaying, and the image of article is only known when people buy some article in store, but always needed
The display position of corresponding goods can just be found by running to and fro, and both wasted the time during searching and be also possible to buy
To the commodity admired.
With the fast development of multimedia technology, vision research and database technology based on computer technology it is wide
General use so that image retrieval technologies are fast-developing, by itself extracting the feature of image from image, with image itself
Feature retrieve to obtain similar image, the characteristic information of the commodity image in store is first extracted, resettles visual word
Allusion quotation storehouse, when people are buying commodity, can directly retrieves corresponding commodity letter by inputting the image of desired commodity
Breath, can save the substantial amounts of hunting time of buyer.
Patent No. CN102208038A patent provides a kind of image classification method of view-based access control model dictionary, including with
Lower step：S1：Extract the joint local feature of training image data set；S2：By based on mobile average and region Hash method
Clustering algorithm enters row vector vector quantization to the joint local feature, so as to select cluster centre number, to form visual dictionary；
S3：The character representation of image is generated according to the visual dictionary, to establish Image Classifier；S4：According to described image grader
The image classified in the training image data set.The invention passes through the image of visual dictionary generation by forming visual dictionary
Feature establishes Image Classifier and image is classified, but the invention is unable to reach the purpose of retrieval image, is needed in user
Corresponding target image can not be retrieved by image.
The content of the invention
The technical problem to be solved in the present invention purpose is that providing a kind of view-based access control model dictionary realizes image quick-searching
Method and system, to solve the problems, such as that user can not quickly search out corresponding goods positional information when buying commodity.
To achieve these goals, the technical solution adopted by the present invention is：
A kind of method that view-based access control model dictionary realizes image quick-searching, including step：
S1：Sample image data collection is received, extracts all visual signatures of the sample image, and pass through the sample graph
As all visual signatures of the ID name identifications sample image；
S2：The visual signature is quantized into sample image characteristic vector to generate visual dictionary storehouse；
S3：The visual signature of image to be retrieved is extracted, and is quantized into image feature vector to be retrieved；
S4：The characteristic vector of the image to be retrieved and the sample image characteristic vector in visual dictionary storehouse are measured
Contrast, the sample image that output matches with the retrieval image.
Further, the step S2 also includes step：
The visual signature set of the sample image extracted is clustered to form cluster according to hierarchical clustering method
Cluster simultaneously enters line splitting to clustering cluster, when the clustering cluster reaches default division number, completes cluster.
Further, the step S2 also includes step：
In the clustering cluster that the maps feature vectors that the sample image extracts are formed to cluster, the sample graph is obtained
The visual representation vocabulary of picture；
Further, in addition to step：
Extract the visual vocabulary of ID titles and described image corresponding to the sample image to represent, generation visual dictionary storehouse.
Further, the step S3 also includes step：
Calculate the Euclidean distance of the characteristic vector and the sample image characteristic vector in visual dictionary storehouse of image to be retrieved；
The Euclidean distance is obtained less than sample image ID titles corresponding to the characteristic vector of predetermined threshold value, and by the sample
This image exports.
The system that a kind of view-based access control model dictionary realizes image quick-searching, including：
Mark module：For receiving image sample data collection, the visual signature of the sample image is extracted, and by described
All visual signatures of the sample image ID name identifications sample image；
Generation module：For the visual signature to be quantized into sample image characteristic vector to generate visual dictionary storehouse；
Quantization modules：For extracting the visual signature of image to be retrieved, and it is quantized into image feature vector to be retrieved；
Contrast output module：For the sample image in the characteristic vector of the image to be retrieved and visual dictionary storehouse is special
Sign vector is contrasted, the sample image that output matches with the retrieval image.
Further, the generation module also includes：
Cluster cell：For the visual signature set of the sample image extracted to be gathered according to hierarchical clustering method
Class, when the clustering cluster reaches default division number, completes cluster to form clustering cluster and enter line splitting to clustering cluster.
Further, the generation module also includes：
Map unit：For the clustering cluster for forming the maps feature vectors that described image is extracted to cluster, obtain
The visual representation vocabulary of the sample image；
Further, in addition to：
Generation unit：Represented for extracting the visual vocabulary of ID titles and described image corresponding to the sample image, it is raw
Into visual dictionary storehouse.
Further, the contrast output module also includes：
Computing unit：For calculating the characteristic vector of image to be retrieved and the sample image characteristic vector in visual dictionary storehouse
Euclidean distance；
Output unit：For obtaining the Euclidean distance less than sample image ID names corresponding to the characteristic vector of predetermined threshold value
Claim, and the sample image is exported.
Using the present invention, when user buys commodity in store, the image retrieval of commodity to corresponding commodity can be passed through
Information, so as to quickly find corresponding commodity, improve the efficiency of purchase simultaneously and find business when saving purchase commodity
The plenty of time that product are spent.
Brief description of the drawings
Fig. 1 is the method flow diagram that a kind of view-based access control model dictionary that embodiment one provides realizes image quick-searching；
Fig. 2 is the system construction drawing that a kind of view-based access control model dictionary that embodiment one provides realizes image quick-searching；
Fig. 3 is the method flow diagram that a kind of view-based access control model dictionary that embodiment two provides realizes image quick-searching；
Fig. 4 is the system construction drawing that a kind of view-based access control model dictionary that embodiment two provides realizes image quick-searching.
Embodiment
It is the specific embodiment of the present invention and with reference to accompanying drawing below, technical scheme is further described,
But the present invention is not limited to these embodiments.
In the present invention, by gathering the sample graph image set of commodity in store, the characteristics of image of sample graph image set is extracted, is established
Visual dictionary storehouse, the image for needing to buy commodity can be directly inputted when user is buying commodity, by retrieval so as to quickly
Obtain the commodity image similar to purchase commodity, and its merchandise news.
Embodiment one
A kind of method that view-based access control model dictionary realizes image quick-searching is present embodiments provided, as shown in figure 1, including
Step：
S11：Sample image data collection is received, extracts all visual signatures of the sample image, and pass through the sample
All visual signatures of the image ID name identifications sample image；
S12：The visual signature set of the sample image extracted is carried out into cluster formation according to hierarchical clustering method to gather
Class cluster, when the clustering cluster reaches default division number, complete cluster；
S13：In the clustering cluster that the maps feature vectors that sample image extracts are formed to cluster, the sample graph is obtained
The visual representation vocabulary of picture；
S14：Extract the visual vocabulary of ID titles and described image corresponding to the sample image to represent, generate visual dictionary
Storehouse；
S15：The visual signature of image to be retrieved is extracted, and is quantized into image feature vector to be retrieved；
S16：By the sample image characteristic vector degree of progress in the characteristic vector of the image to be retrieved and visual dictionary storehouse
Amount contrast, the sample image that output matches with the retrieval image.
The image of commodity in store is acquired, generates sample image data collection, the image extracted in sample image is special
Sign, establishes visual dictionary storehouse, when user requires to look up corresponding goods, need to only input corresponding goods image, after retrieval
So as to export the commodity image and information of similar degree matching.
In the present embodiment, for step S11 to receive sample image data collection, all visions for extracting the sample image are special
Sign, and pass through all visual signatures of the sample image ID name identifications sample image.
Specifically,
Store gathers the image of all commodity as sample image data collection, and the vision that image is extracted by algorithm is special
Sign, a while ID title is distributed for each secondary sample image, ID titles be in image data base every width figure think it is unique
Identity, carry out single mapping association by all visual signatures in ID titles image corresponding with ID titles.
In the present embodiment, step S12 is the visual signature set for the sample image that will be extracted according to hierarchical clustering
Method carries out cluster and forms clustering cluster, when the clustering cluster reaches default division number, completion cluster.
Specifically,
After mapping all visual signatures of the image by image ID titles, the visual signature extracted is quantized into spy
Sign vector, is clustered by hierarchical clustering method to all visual signatures, forms clustering cluster, clustering cluster is constantly divided
Split, reach default number of plies threshold value when the characteristic vector quantity in clustering cluster is less than predetermined threshold value or is layered the number of plies, stop division,
Complete cluster.
In the present embodiment, step S13 is the clustering cluster for forming the maps feature vectors that sample image extracts to cluster
On, obtain the visual representation vocabulary of the sample image.
Specifically,
After the completion of all clusters, then the cluster that the maps feature vectors extracted from sample image are formed to cluster
On cluster, represented so as to form the visual vocabulary of every piece image.
In the present embodiment, step S14 is the visual vocabulary of ID titles and described image corresponding to the extraction sample image
Represent, generation visual dictionary storehouse.
Specifically, by a certain characteristic vector, visual representation vocabulary and file corresponding to this feature vector can be got
Name ID, can be with the similar picture of search and output by filename ID.
In the present embodiment, step S15 is the visual signature for extracting image to be retrieved, and be quantized into characteristics of image to be retrieved to
Amount.
The commodity image to be retrieved provided by user, is quantized into characteristic vector, with visual dictionary by its visual signature
Characteristic vector in storehouse carries out measurement comparison.
In the present embodiment, step S16 is by the sample graph in the characteristic vector of the image to be retrieved and visual dictionary storehouse
As characteristic vector carries out measurement contrast, the sample image that output matches with the retrieval image.
Specifically,
Characteristic vector in the characteristic vector of image to be retrieved and visual dictionary storehouse is subjected to similarity mode, compares two
The Euclidean distance of vector, when less than pre-determined distance threshold value, extracts image corresponding to the similar characteristic vector, by corresponding diagram
As output, the similar commodity that user needs are obtained.
The present embodiment also provides the system that a kind of view-based access control model dictionary realizes image quick-searching, as shown in Fig. 2 bag
Include：
Mark module 21：For receiving sample image data collection, all visual signatures of the sample image are extracted, and are led to
Cross all visual signatures of the sample image ID name identifications sample image；
Generation module 22：For the visual signature to be quantized into sample image characteristic vector to generate visual dictionary storehouse；
Quantization modules 23：For extracting the visual signature of image to be retrieved, and it is quantized into image feature vector to be retrieved；
Contrast output module 24：For by the sample image in the characteristic vector of the image to be retrieved and visual dictionary storehouse
Characteristic vector carries out measurement contrast, the sample image that output matches with the retrieval image.
Wherein generation module 22 also includes：Cluster cell 221, map unit 222, generation unit 223.
In the present embodiment, the visual signature set for the sample image that cluster cell 221 is used to extract is according to layer
Secondary clustering procedure carries out cluster and forms clustering cluster, when the clustering cluster reaches default division number, completes cluster.
Specifically,
All characteristic vector set are clustered, and the clustering cluster formed to cluster enters line splitting, into clustering cluster
Characteristic vector quantity when being less than predetermined threshold value or when the division number of plies reaches the default layering number of plies, complete cluster.
In the present embodiment, map unit 222 is used for form the maps feature vectors that sample image extracts to cluster
In clustering cluster, the visual representation vocabulary of the sample image is obtained.
In the present embodiment, generation unit 223 is used to extract regarding for ID titles corresponding to the sample image and described image
Feel lexical representation, generation visual dictionary storehouse.
By establishing visual dictionary storehouse, the commodity image that user can be bought by needs, retrieval obtains similar to its
Merchandise news, save user and stroll the time that store is spent, while can also provide the user the reference information of a variety of similar commodity,
Both the purchasing demand of the user met or the efficiently buying pattern that provides users with the convenient.
Embodiment two
A kind of method that view-based access control model dictionary realizes image quick-searching is present embodiments provided, as shown in figure 3, including
Step：
S31：Sample image data collection is received, extracts all visual signatures of the sample image, and pass through the sample
All visual signatures of the image ID name identifications sample image；
S32：The visual signature is quantized into sample image characteristic vector to generate visual dictionary storehouse；
S33：The visual signature of image to be retrieved is extracted, and is quantized into image feature vector to be retrieved；
S34：Calculate the characteristic vector of image to be retrieved and the sample image characteristic vector in visual dictionary storehouse it is European away from
From；
S35：The Euclidean distance is obtained less than sample image ID titles corresponding to the characteristic vector of predetermined threshold value, and by institute
State sample image output.
Euclidean distance is the distance definition of a generally use, refers to the actual distance between two points in m-dimensional space, or
The natural length of person's vector.Euclidean distance in two and three dimensions space is exactly the actual range between 2 points.
It is with the difference of embodiment one, the present embodiment also includes step S34, S35.
Step S34 is the Europe of the characteristic vector and the sample image characteristic vector in visual dictionary storehouse that calculate image to be retrieved
Formula distance.
Specifically,
Similarity mode is carried out by calculating the Euclidean distance between characteristic vector, when the Euclidean distance between vector is small
When predetermined threshold value, the characteristic vector in the visual dictionary storehouse is obtained.
Step S35 is to obtain the Euclidean distance to be less than sample image ID titles corresponding to the characteristic vector of predetermined threshold value,
And the sample image is exported.
Specifically,
Characteristic vector in visual dictionary storehouse corresponds to the image ID of store commodity, is retrieved by image ID and obtains corresponding goods
Image, and will be exported with the close image of image similarity to be retrieved, it is shown to user.
The present embodiment also includes the system that a kind of view-based access control model dictionary realizes image quick-searching, as shown in figure 4, bag
Include：
Mark module 41：For receiving sample image data collection, all visual signatures of the sample image are extracted, and are led to
Cross all visual signatures of the sample image ID name identifications sample image；
Generation module 42：The visual signature is quantized into sample image characteristic vector to generate visual dictionary storehouse；
Quantization modules 43：The visual signature of image to be retrieved is extracted, and is quantized into image feature vector to be retrieved；
Contrast output module 44：The Euclidean distance is obtained less than sample image ID corresponding to the characteristic vector of predetermined threshold value
Title, and the sample image is exported.
In the present embodiment, contrast output module 44 also includes：Computing unit 441, output unit 442.
In the present embodiment, computing unit 441 is used for the characteristic vector for calculating image to be retrieved and the sample in visual dictionary storehouse
The Euclidean distance of this image feature vector.
In the present embodiment, output unit 442 is corresponding less than the characteristic vector of predetermined threshold value for obtaining the Euclidean distance
Sample image ID titles, and the sample image is exported.
Specifically, by the way that characteristic vector to be retrieved characteristic vector corresponding with visual dictionary storehouse is carried out into matching pair
Than calculating the Euclidean distance between two vectors, when less than predetermined threshold value, obtaining image corresponding to the characteristic vector of the matching
ID, and image corresponding to image ID is exported.
By retrieval of visual dictionary, after obtaining the image of the commodity similar with the commodity image of user's input, by class
As merchandise news corresponding to commodity image be sent to user, user can directly obtain the information of similar clause, without
Searching of going shopping is gone, has great convenience for the user shopping process, improves the purchase experiences of user.
Specific embodiment described herein is only to spirit explanation for example of the invention.Technology belonging to the present invention is led
The technical staff in domain can be made various modifications or supplement to described specific embodiment or be replaced using similar mode
Generation, but without departing from the spiritual of the present invention or surmount scope defined in appended claims.
Claims (10)
1. a kind of method that view-based access control model dictionary realizes image quick-searching, it is characterised in that including step：
S1：Sample image data collection is received, extracts all visual signatures of the sample image, and pass through the sample image ID
All visual signatures of the name identification sample image；
S2：The visual signature is quantized into sample image characteristic vector to generate visual dictionary storehouse；
S3：The visual signature of image to be retrieved is extracted, and is quantized into image feature vector to be retrieved；
S4：Sample image characteristic vector in the characteristic vector of the image to be retrieved and visual dictionary storehouse is subjected to measurement pair
Than the sample image that output matches with the retrieval image.
2. the method that a kind of view-based access control model dictionary according to claim 1 realizes image quick-searching, it is characterised in that
The step S2 also includes step：
The visual signature set of the sample image extracted is clustered to form clustering cluster simultaneously according to hierarchical clustering method
Line splitting is entered to clustering cluster, when the clustering cluster reaches default division number, completes cluster.
3. the method that a kind of view-based access control model dictionary according to claim 1,2 realizes image quick-searching, its feature exist
In the step S2 also includes step：
In the clustering cluster that the maps feature vectors that the sample image extracts are formed to cluster, the sample image is obtained
Visual representation vocabulary.
4. the method that a kind of view-based access control model dictionary according to claim 3 realizes image quick-searching, it is characterised in that
Also include step：
Extract the visual vocabulary of ID titles and described image corresponding to the sample image to represent, generation visual dictionary storehouse.
5. the method that a kind of view-based access control model dictionary according to claim 1 realizes image quick-searching, it is characterised in that
The step S3 also includes step：
Calculate the Euclidean distance of the characteristic vector and the sample image characteristic vector in visual dictionary storehouse of image to be retrieved；
The Euclidean distance is obtained less than sample image ID titles corresponding to the characteristic vector of predetermined threshold value, and by the sample graph
As output.
A kind of 6. system that view-based access control model dictionary realizes image quick-searching, it is characterised in that including：
Mark module：For receiving image sample data collection, the visual signature of the sample image is extracted, and pass through the sample
All visual signatures of the image ID name identifications sample image；
Generation module：For the visual signature to be quantized into sample image characteristic vector to generate visual dictionary storehouse；
Quantization modules：For extracting the visual signature of image to be retrieved, and it is quantized into image feature vector to be retrieved；
Contrast output module：For by the sample image feature in the characteristic vector of the image to be retrieved and visual dictionary storehouse to
Amount is contrasted, the sample image that output matches with the retrieval image.
7. the system that a kind of view-based access control model dictionary according to claim 6 realizes image quick-searching, it is characterised in that
The generation module also includes：
Cluster cell：For by the visual signature set of the sample image extracted according to hierarchical clustering method clustered with
Form clustering cluster and line splitting is entered to clustering cluster, when the clustering cluster reaches default division number, complete cluster.
8. the system that a kind of view-based access control model dictionary according to claim 6,7 realizes image quick-searching, its feature exist
In the generation module also includes：
Map unit：For the clustering cluster for forming the maps feature vectors that described image is extracted to cluster, described in acquisition
The visual representation vocabulary of sample image.
9. the system that a kind of view-based access control model dictionary according to the claims 8 realizes image quick-searching, its feature
It is, in addition to：
Generation unit：Represented for extracting the visual vocabulary of ID titles and described image corresponding to the sample image, generation regards
Feel dictionary.
10. the system that a kind of view-based access control model dictionary according to claim 6 realizes image quick-searching, its feature exist
In the contrast output module also includes：
Computing unit：For the Europe of the characteristic vector and the sample image characteristic vector in visual dictionary storehouse that calculate image to be retrieved
Formula distance；
Output unit：It is less than sample image ID titles corresponding to the characteristic vector of predetermined threshold value for obtaining the Euclidean distance,
And the sample image is exported.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710978931.2A CN107730357A (en) | 2017-10-19 | 2017-10-19 | A kind of view-based access control model dictionary realizes the method and system of image quick-searching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710978931.2A CN107730357A (en) | 2017-10-19 | 2017-10-19 | A kind of view-based access control model dictionary realizes the method and system of image quick-searching |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107730357A true CN107730357A (en) | 2018-02-23 |
Family
ID=61212079
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710978931.2A Pending CN107730357A (en) | 2017-10-19 | 2017-10-19 | A kind of view-based access control model dictionary realizes the method and system of image quick-searching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107730357A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112269894A (en) * | 2020-12-03 | 2021-01-26 | 北京沃东天骏信息技术有限公司 | Article pool generation method, image search method, device, electronic equipment and medium |
| CN114356852A (en) * | 2022-03-21 | 2022-04-15 | 展讯通信(天津)有限公司 | File retrieval method, electronic equipment and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103020111A (en) * | 2012-10-29 | 2013-04-03 | 苏州大学 | Image retrieval method based on vocabulary tree level semantic model |
| CN103207879A (en) * | 2012-01-17 | 2013-07-17 | 阿里巴巴集团控股有限公司 | Method and equipment for generating image index |
| CN104036012A (en) * | 2014-06-24 | 2014-09-10 | 中国科学院计算技术研究所 | Dictionary learning method, visual word bag characteristic extracting method and retrieval system |
| CN104820718A (en) * | 2015-05-22 | 2015-08-05 | 哈尔滨工业大学 | Image classification and searching method based on geographic position characteristics and overall situation vision characteristics |
| CN106156374A (en) * | 2016-09-13 | 2016-11-23 | 华侨大学 | A kind of view-based access control model dictionary optimizes and the image search method of query expansion |
| CN106649490A (en) * | 2016-10-08 | 2017-05-10 | 中国人民解放军理工大学 | Depth feature-based image retrieval method and apparatus |
| CN106886553A (en) * | 2016-12-27 | 2017-06-23 | 浙江宇视科技有限公司 | A kind of image search method and server |
| CN107066459A (en) * | 2016-08-30 | 2017-08-18 | 广东百华科技股份有限公司 | A kind of efficient image search method |
-
2017
- 2017-10-19 CN CN201710978931.2A patent/CN107730357A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103207879A (en) * | 2012-01-17 | 2013-07-17 | 阿里巴巴集团控股有限公司 | Method and equipment for generating image index |
| CN103020111A (en) * | 2012-10-29 | 2013-04-03 | 苏州大学 | Image retrieval method based on vocabulary tree level semantic model |
| CN104036012A (en) * | 2014-06-24 | 2014-09-10 | 中国科学院计算技术研究所 | Dictionary learning method, visual word bag characteristic extracting method and retrieval system |
| CN104820718A (en) * | 2015-05-22 | 2015-08-05 | 哈尔滨工业大学 | Image classification and searching method based on geographic position characteristics and overall situation vision characteristics |
| CN107066459A (en) * | 2016-08-30 | 2017-08-18 | 广东百华科技股份有限公司 | A kind of efficient image search method |
| CN106156374A (en) * | 2016-09-13 | 2016-11-23 | 华侨大学 | A kind of view-based access control model dictionary optimizes and the image search method of query expansion |
| CN106649490A (en) * | 2016-10-08 | 2017-05-10 | 中国人民解放军理工大学 | Depth feature-based image retrieval method and apparatus |
| CN106886553A (en) * | 2016-12-27 | 2017-06-23 | 浙江宇视科技有限公司 | A kind of image search method and server |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112269894A (en) * | 2020-12-03 | 2021-01-26 | 北京沃东天骏信息技术有限公司 | Article pool generation method, image search method, device, electronic equipment and medium |
| CN112269894B (en) * | 2020-12-03 | 2021-07-13 | 北京沃东天骏信息技术有限公司 | Article pool generation method, image search method, device, electronic equipment and medium |
| CN114356852A (en) * | 2022-03-21 | 2022-04-15 | 展讯通信(天津)有限公司 | File retrieval method, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jagadeesh et al. | Large scale visual recommendations from street fashion images | |
| US10747826B2 (en) | Interactive clothes searching in online stores | |
| CN103678335B (en) | The method of method, apparatus and the commodity navigation of commodity sign label | |
| CN102254043B (en) | Semantic mapping-based clothing image retrieving method | |
| CN107748754B (en) | Knowledge graph perfecting method and device | |
| KR102075833B1 (en) | Curation method and system for recommending of art contents | |
| CN103617230B (en) | Method and system for advertisement recommendation based microblog | |
| CN102609523B (en) | The collaborative filtering recommending method classified based on taxonomy of goods and user | |
| CN107679960B (en) | Personalized clothing recommendation method based on clothing image and label text bimodal content analysis | |
| KR20230087622A (en) | Methods and apparatus for detecting, filtering, and identifying objects in streaming video | |
| CN101937549B (en) | Picture guidance system for network shopping guidance | |
| CN104298749A (en) | Commodity retrieval method based on image visual and textual semantic integration | |
| CN101950400B (en) | Picture retrieving method of network shopping guiding method | |
| US20180322208A1 (en) | System and method for searching for products in catalogs | |
| CN109213921A (en) | A kind of searching method and device of merchandise news | |
| CN103049513A (en) | Multi-visual-feature fusion method of commodity images of clothing, shoes and bags | |
| Gao et al. | Fashion clothes matching scheme based on Siamese Network and AutoEncoder | |
| CN103577472B (en) | Personal information acquisition, presumption, the classification of commodity, search method and system | |
| Kotouza et al. | Towards fashion recommendation: an AI system for clothing data retrieval and analysis | |
| CN108932647A (en) | A kind of method and apparatus for predicting its model of similar article and training | |
| CN105844302A (en) | Depth-learning-based method for automatically calculating commodity trend indexes | |
| CN107730357A (en) | A kind of view-based access control model dictionary realizes the method and system of image quick-searching | |
| CN107622071B (en) | Clothes image retrieval system and method under non-source-retrieval condition through indirect correlation feedback | |
| US20230022712A1 (en) | Method, apparatus, and computer program for recommending fashion product | |
| CN112508114A (en) | Intelligent clothing recommendation system and method |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20201125 Address after: Room 10242, No. 260, Jiangshu Road, Xixing street, Binjiang District, Hangzhou City, Zhejiang Province Applicant after: Hangzhou Jiji Intellectual Property Operation Co.,Ltd. Address before: 201616 Shanghai city Songjiang District Sixian Road No. 3666 Applicant before: Phicomm (Shanghai) Co.,Ltd. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180223 |