CN104142922A - Online mobile image searching and mining classification method - Google Patents
Online mobile image searching and mining classification method Download PDFInfo
- Publication number
- CN104142922A CN104142922A CN201310161565.3A CN201310161565A CN104142922A CN 104142922 A CN104142922 A CN 104142922A CN 201310161565 A CN201310161565 A CN 201310161565A CN 104142922 A CN104142922 A CN 104142922A
- Authority
- CN
- China
- Prior art keywords
- image
- feature
- semantic
- sorting technique
- excavation
- 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
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Library & Information Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to an online mobile image searching and mining classification method and belongs to the computer application field. The online mobile image searching and mining classification method is characterized in that image indexing-based photographing searching and mining functions are developed by means of the voice, image and video obtaining agility and convenience of a mobile terminal; the global feature of the image is extracted automatically through classifying different features of the image. The online mobile image searching and mining classification method enables a computer image indexing system to better conform to the thinking habits of people, solves the difficulty in searching through characters, provides more convenient information service for users and has important development significance.
Description
Technical field
The sorting technique that the present invention relates to a kind of mobile image on-line search and excavation, belongs to computer application field.
Background technology
Image is as a kind of rich connotation, show multimedia messages intuitively, enjoy people's favor, increasing business activity, affairs transaction and Informational Expression comprise view data, particularly popular shopping online, substantially taking image as principal mode to user's presenting information, how in mass image data, effectively to retrieve the hot issue that becomes current research, early stage image retrieval is the retrieval based on image labeling text, but artificial mark is wasted time and energy, and also there is subjectivity and inexactness in mark, the directly accuracy of impact retrieval.
Summary of the invention
Object of the present invention is achieved through the following technical solutions:
A sorting technique for mobile image on-line search and excavation, is a kind of combination switch, capacitor etc. organically to be carried out to integrated technical products, belongs to computer application field.It is characterized in that: utilize mobile terminal have the information such as the voice of obtaining, image, video flexibly, feature easily, develop take pictures search and data mining duty based on image retrieval, by the classification to the each category feature of image, automatically extract image overall feature, make searching computer system more meet the mankind's thinking habit, solve the difficulty that user searches for word, for user provides information service more easily, there is far-reaching exploitation meaning.
Further, above-mentioned a kind of mobile image on-line search and the sorting technique of excavation, it is characterized in that: described sorting technique is to show image affix to comprise semantic at interior more high-rise content information, make searching computer system more meet the mankind's thinking habit, linguistic indexing of pictures and retrieval become the closely-related study hotspot with CBIR, utilize image overall feature and have generated a series of semantic markers taking certain degree of confidence as every width image.
Further, above-mentioned a kind of mobile image on-line search and the sorting technique of excavation, it is characterized in that: described semantic tagger refers to that being labeled as training set by craft adds markup information, and use svm classifier device to classify, by by the method for the visual signature such as color histogram and edge histogram and self-defining semantic feature combination construction feature vector, use fuzzy k-NN sorting algorithm to classify.
Further, above-mentioned a kind of mobile image on-line search and the sorting technique of excavation, it is characterized in that: described classification refers to by the latent semantic analysis model of two probability obtains respectively the vision mode of training image and the information of text modality, and the theme that is two kinds of different modalities by every width image representation distributes.
Further, above-mentioned a kind of mobile image on-line search and the sorting technique of excavation, it is characterized in that: described sorting technique adopts Optimality Criteria based on minimal error rate and the thought of statistical classification, be that every width image extracts a feature set in the training classifier stage, utilize the feature set learning semantic concept of many learn-by-example algorithms from multiple image, thereby set up probability model for each semantic concept.
Embodiment
A sorting technique for mobile image on-line search and excavation, is a kind of combination switch, capacitor etc. organically to be carried out to integrated technical products, belongs to computer application field.It is characterized in that: utilize mobile terminal have the information such as the voice of obtaining, image, video flexibly, feature easily, develop take pictures search and data mining duty based on image retrieval, by the classification to the each category feature of image, automatically extract image overall feature, make searching computer system more meet the mankind's thinking habit, solve the difficulty that user searches for word, for user provides information service more easily, there is far-reaching exploitation meaning.
Mobile picture search based on cloud computing and digging system framework comprise two major parts: high in the clouds and mobile terminal.View data is generally stored in the Web database of WWW, and image data acquisition, view data storage are mainly realized in high in the clouds, image cuts apart and assemblage characteristic optimization, image index structure and user interest model structure etc.First high in the clouds need to obtain view data from the data source of InterM according to mobile picture search, then utilizes the data storage and management technology of cloud computing to realize distributed storage and the management of view data.
Image is cut apart and assemblage characteristic is optimized module model and move the characteristic set of the Images Classification data of picture search needs, then image is carried out to regional area uncertainty and cuts apart and adopt the best cluster feature of intensified learning acquisition particular category image to combine; Index construct module is to cut apart on the basis of optimizing with assemblage characteristic at image, utilizes respectively the duplicate key of the semantic design of graphics picture of image block and global image, improves accuracy and the dirigibility of index, the new image data of obtaining is realized to increment index simultaneously and builds; User interest model is added up and is learnt automatically to build in the image that user interest structure module is mainly associated according to the recent access activity of mobile subscriber and the behavior of processing image thereof, and along with the variation of user interest, realize the renewal of user interest model, keep the timeliness n of interest model.
Image search is uploaded in mobile terminal by taking pictures, show the Search Results that high in the clouds is returned, and complete the various operations (for example: purchase, collection, preservation etc.) to returning results.In the time that mobile subscriber searches for, on the basis that accurately mate at characteristics of image in high in the clouds, carry out personalization according to user interest model and arrange, can also carry out personalized recommendation service based on user interest model simultaneously.
This method is by analyzing basic thought, the feature of current image characteristic extracting method, the Images Classification data characteristics set that is suitable for mobile picture search application is set up in research, on this basis, further the image local area uncertainty of research based on cloud computing cut apart and Feature Combination optimization method, obtains the best cluster feature combination of particular category image.
The parallel clustering method of research based on image local area physical features, on this basis, the semantic similarity community network of design of graphics picture, and the semantic indexing of the extensive community analysis method design of graphics picture of research based on Map/Reduce.The physical features of combining image and semantic feature index building, make the index of large nuber of images more accurate.
The physical features of combining image and semantic feature, the user personalized interest model building method of research based on cloud computing, and research adopts the parallel update strategy of the user interest model of the interest attenuation factor and increment updating technology, the efficiency that improves user interest model structure and upgrade.
Design and Implement mobile picture search and digging system based on cloud computing, shopping search field by this system applies to fashion class commodity, in the time that user sees the commodity of liking, take pictures and just can search for mobile terminal, solve user and can not or be reluctant the difficulty of searching for word, for user provides the decision information of doing shopping accurately and effectively, the shopping that strengthens user is experienced.
Claims (5)
1. a sorting technique for mobile image on-line search and excavation, belongs to computer application field.Be characterized in: utilize mobile terminal have the information such as the voice of obtaining, image, video flexibly, feature easily, develop take pictures search and data mining duty based on image retrieval, by the classification to the each category feature of image, automatically extract image overall feature, make searching computer system more meet the mankind's thinking habit, solve the difficulty that user searches for word, for user provides information service more easily, there is far-reaching exploitation meaning.
2. the sorting technique of a kind of mobile image on-line search according to claim 1 and excavation, it is characterized in that: described sorting technique is to show image affix to comprise semantic at interior more high-rise content information, make searching computer system more meet the mankind's thinking habit, linguistic indexing of pictures and retrieval become the closely-related study hotspot with CBIR, utilize image overall feature and have generated a series of semantic markers taking certain degree of confidence as every width image.
3. the sorting technique of a kind of mobile image on-line search according to claim 2 and excavation, it is characterized in that: described semantic tagger refers to that being labeled as training set by craft adds markup information, and use svm classifier device to classify, by by the method for the visual signature such as color histogram and edge histogram and self-defining semantic feature combination construction feature vector, use fuzzy k-NN sorting algorithm to classify.
4. the sorting technique of a kind of mobile image on-line search according to claim 3 and excavation, it is characterized in that: described classification refers to by the latent semantic analysis model of two probability obtains respectively the vision mode of training image and the information of text modality, and the theme that is two kinds of different modalities by every width image representation distributes.
5. the sorting technique of a kind of mobile image on-line search according to claim 3 and excavation, it is characterized in that: described sorting technique adopts Optimality Criteria based on minimal error rate and the thought of statistical classification, be that every width image extracts a feature set in the training classifier stage, utilize the feature set learning semantic concept of many learn-by-example algorithms from multiple image, thereby set up probability model for each semantic concept.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310161565.3A CN104142922A (en) | 2013-05-06 | 2013-05-06 | Online mobile image searching and mining classification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310161565.3A CN104142922A (en) | 2013-05-06 | 2013-05-06 | Online mobile image searching and mining classification method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104142922A true CN104142922A (en) | 2014-11-12 |
Family
ID=51852098
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310161565.3A Pending CN104142922A (en) | 2013-05-06 | 2013-05-06 | Online mobile image searching and mining classification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104142922A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105956631A (en) * | 2016-05-19 | 2016-09-21 | 南京大学 | On-line progressive image classification method facing electronic image base |
| CN108647264A (en) * | 2018-04-28 | 2018-10-12 | 北京邮电大学 | A kind of image automatic annotation method and device based on support vector machines |
| CN109165293A (en) * | 2018-08-08 | 2019-01-08 | 上海宝尊电子商务有限公司 | A kind of expert data mask method and program towards fashion world |
| CN109784267A (en) * | 2019-01-10 | 2019-05-21 | 济南浪潮高新科技投资发展有限公司 | A kind of mobile terminal multi-source fusion image, semantic content generation system and method |
| CN115934661A (en) * | 2023-03-02 | 2023-04-07 | 浪潮电子信息产业股份有限公司 | Graph neural network compression method and device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101000623A (en) * | 2007-01-08 | 2007-07-18 | 深圳市宜搜科技发展有限公司 | Method for image identification search by mobile phone photographing and device using the method |
| US20110200260A1 (en) * | 2010-02-16 | 2011-08-18 | Imprezzeo Pty Limited | Image retrieval using texture data |
| CN102810160A (en) * | 2012-06-06 | 2012-12-05 | 北京京东世纪贸易有限公司 | Method and device for searching images |
-
2013
- 2013-05-06 CN CN201310161565.3A patent/CN104142922A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101000623A (en) * | 2007-01-08 | 2007-07-18 | 深圳市宜搜科技发展有限公司 | Method for image identification search by mobile phone photographing and device using the method |
| US20110200260A1 (en) * | 2010-02-16 | 2011-08-18 | Imprezzeo Pty Limited | Image retrieval using texture data |
| CN102810160A (en) * | 2012-06-06 | 2012-12-05 | 北京京东世纪贸易有限公司 | Method and device for searching images |
Non-Patent Citations (2)
| Title |
|---|
| 何竞: "图像语义标注中的块-全局特征提取方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 李志欣 等: "融合语义主题的图像自动标注", 《软件学报》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105956631A (en) * | 2016-05-19 | 2016-09-21 | 南京大学 | On-line progressive image classification method facing electronic image base |
| CN108647264A (en) * | 2018-04-28 | 2018-10-12 | 北京邮电大学 | A kind of image automatic annotation method and device based on support vector machines |
| CN108647264B (en) * | 2018-04-28 | 2020-10-13 | 北京邮电大学 | Automatic image annotation method and device based on support vector machine |
| CN109165293A (en) * | 2018-08-08 | 2019-01-08 | 上海宝尊电子商务有限公司 | A kind of expert data mask method and program towards fashion world |
| CN109784267A (en) * | 2019-01-10 | 2019-05-21 | 济南浪潮高新科技投资发展有限公司 | A kind of mobile terminal multi-source fusion image, semantic content generation system and method |
| CN109784267B (en) * | 2019-01-10 | 2021-10-15 | 山东浪潮科学研究院有限公司 | Mobile terminal multi-source fusion image semantic content generation system and method |
| CN115934661A (en) * | 2023-03-02 | 2023-04-07 | 浪潮电子信息产业股份有限公司 | Graph neural network compression method and device, electronic equipment and storage medium |
| CN115934661B (en) * | 2023-03-02 | 2023-07-14 | 浪潮电子信息产业股份有限公司 | Method and device for compressing graphic neural network, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kou et al. | Social network search based on semantic analysis and learning | |
| CN103678335B (en) | The method of method, apparatus and the commodity navigation of commodity sign label | |
| Xu et al. | Topic based context-aware travel recommendation method exploiting geotagged photos | |
| CN105824959B (en) | Public opinion monitoring method and system | |
| CN104239501B (en) | Mass video semantic annotation method based on Spark | |
| Shen et al. | Attraction recommendation: Towards personalized tourism via collective intelligence | |
| CN105718579A (en) | Information push method based on internet-surfing log mining and user activity recognition | |
| CN113158023B (en) | Public digital life accurate classification service method based on mixed recommendation algorithm | |
| Zhao et al. | Scene classification via latent Dirichlet allocation using a hybrid generative/discriminative strategy for high spatial resolution remote sensing imagery | |
| CN103678670A (en) | Micro-blog hot word and hot topic mining system and method | |
| CN104142922A (en) | Online mobile image searching and mining classification method | |
| CN112559907A (en) | Basic data retrieval and integrated display method based on spatio-temporal label spatio-temporal correlation | |
| CN103778206A (en) | Method for providing network service resources | |
| Ye et al. | A web services classification method based on GCN | |
| CN103761286B (en) | A kind of Service Source search method based on user interest | |
| Van Canneyt et al. | Categorizing events using spatio-temporal and user features from Flickr | |
| Rangel et al. | Lextomap: lexical-based topological mapping | |
| Huang et al. | Tag refinement of micro-videos by learning from multiple data sources | |
| Chiang | Interactive tool for image annotation using a semi-supervised and hierarchical approach | |
| CN111223014B (en) | Method and system for online generation of subdivision scene teaching courses from a large number of subdivision teaching contents | |
| CN106777395A (en) | A kind of topic based on community's text data finds system | |
| Yu et al. | A Multi-Directional Search technique for image annotation propagation | |
| Liu | A survey on social image mining | |
| CN104142921A (en) | Image feature optimal-combination method based on cloud computing | |
| Kawewong et al. | A speeded-up online incremental vision-based loop-closure detection for long-term SLAM |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20141112 |