CN105550222A - Distributed storage-based image service system and method - Google Patents
Distributed storage-based image service system and method Download PDFInfo
- Publication number
- CN105550222A CN105550222A CN201510886086.7A CN201510886086A CN105550222A CN 105550222 A CN105550222 A CN 105550222A CN 201510886086 A CN201510886086 A CN 201510886086A CN 105550222 A CN105550222 A CN 105550222A
- Authority
- CN
- China
- Prior art keywords
- picture
- feature
- module
- distributed storage
- hdfs
- 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.)
- Granted
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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Library & Information Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a distributed storage-based image service system. The system comprises a picture reception module, a feature analysis module, a feature library, a picture cache server, an HDFS, a feature matching module, a cache management module, a picture output module, a feedback processing module, an offline file optimization module and an image optimization module, wherein the cache management module is used for carrying out distributed caching on the picture searching result so that the picture searching result can be directly used next time; the picture output module is used for outputting the picture searching result; the feedback processing module is used for carrying out similar picture merging according to the picture searching result and restoring the pictures into the HDFS so as to optimize the picture storage and improve the picture similarity in the same large files; the offline file optimization module is used for carrying out offline partition de-weighting and integrating processing on the large files in the HDFS so as to obtain new large files and indexes; and the image optimization module is used for carrying out image analysis and processing on the pictures stored in the HDFS so as to optimize the image storage. The invention furthermore discloses a distributed storage-based image service method which is realized through the distributed storage-based image service system.
Description
Technical field
The present invention relates to a kind of cloud computing and large data fields, particularly relate to a kind of images serve system and method based on distributed storage.
Background technology
The readwrite performance of HDFS to small documents is poor, and current normal picture is nearly all small documents, if need to store mass picture in HDFS, need small documents to be merged into large files, set up index, restore HDFS.When using image retrieval service, the result that image retrieval obtains is often a N number of picture set, in the worst cases, obtain N number of picture, if each picture is in different large files, then needs to open and reads N number of large files.The set of N number of picture, owing to being drawn by characteristic similarity coupling, is feature similarity in fact.The picture of feature similarity, in retrieval afterwards, occurs possibly again together.But such situation is not optimized now, be stored in picture in HDFS and likely have repetition, or the immediate picture of characteristic similarity is stored in different large files, when needing to extract these similar pictures after retrieving, need to open multiple large files to read, reduce efficiency.
Summary of the invention
In order to solve the problem, the invention provides a kind of images serve system and method based on distributed storage that can improve picture retrieval efficiency and accuracy.
A kind of images serve system based on distributed storage of the present invention, comprise the picture receiver module receiving the picture that needs store and needs are retrieved, the picture received described picture receiver module extracts the characteristics analysis module of picture feature according to image characteristics extraction algorithm, the picture feature of the picture that the needs described characteristics analysis module extracted store carries out the feature database stored, the picture that being received by described picture receiver module needs to store extracts stored in buffer memory and according to described characteristics analysis module and needs the picture feature similarity of the picture of storage to carry out to picture the image cache server merged of classifying, for storing the HDFS of the large files be made up of the picture of described image cache block merging, picture feature and the feature in described feature database of the picture of the needs retrieval of described characteristics analysis module being extracted carry out the characteristic matching module of mating, the caching management module of distributed caching in order to directly using next time is carried out to the result for retrieval that retrieving image produces, export the picture output module of picture result for retrieval, according to picture retrieval result carry out similar pictures merging treatment and again stored in described HDFS to optimize the feedback processing modules that picture-storage improves the picture analogies degree in same large files, carry out off-line partition duplicate removal to the large files in described HDFS and integrate process optimizing module with the off-line files obtaining new large files and index, graphical analysis is carried out to the picture stored in described HDFS and processes the image optimization module stored with optimized image.
A kind of images serve method based on distributed storage of the present invention, by realizing picture-storage management, picture retrieval management, cache management and optimization process based on the images serve system of distributed storage.
Wherein, described picture-storage management comprises the steps:
S11, received the picture needing to store by the described picture receiver module based in the images serve system of distributed storage, enter step s2;
S12, by the described feature extracting the picture received in described step s1 based on the characteristics analysis module in the images serve system of distributed storage, simultaneously by the picture that receives stored in described based on the image cache server in the images serve system of distributed storage, enter step s3 simultaneously;
S13, by the picture feature extracted in described step s2 stored in described based on the feature database in the images serve system of distributed storage, enter step s4;
S14, the picture feature extracted in described step s2 is sent to image cache server, according to picture feature, merging treatment is carried out to similar pictures by described image cache server, by the large files that is made up of the picture after merging treatment stored in HDFS, enter step s5;
S15, set up the index of picture in large files and the mapping relations of picture feature and picture-storage address by described image cache server.
Described picture retrieval management comprises the steps:
The picture of s21, the retrieval of input needs, receives picture by described picture receiver module, enters step s22;
S22, according to image characteristics extraction algorithm, signature analysis is carried out to the picture to be retrieved received in described step s21 by described characteristics analysis module, extract the picture feature of picture to be retrieved, enter step s23;
S23, the picture feature of the picture to be retrieved analyzed in described step s22 to be mated with the feature in described feature database based on the characteristics analysis module in the images serve system of distributed storage by described, in feature database, mate similar feature according to the picture feature of picture to be retrieved, enter step s24;
S24, find the mapping relations of picture feature and picture address according to the feature of mating in described step s23, find all large files relevant in HDFS according to these mapping relations, enter step s25;
S25, the picture index in large files corresponding according to picture feature, obtain, with the concrete picture of the characteristic matching of mating in described step s23, entering step s26;
S26, judge that whether in the buffer the concrete picture that to obtain in described step s25, if so, enters step s27; If not, enter step s28;
S27, diameter extract picture from buffer memory to be shown as result for retrieval based on the picture output module in the images serve system of distributed storage by described; Enter step s210;
S28, from HDFS, extract the picture that obtains in described step s25 and shown as result for retrieval by described picture output module, entering step s29;
S29, according to showing that result is by the described caching management module renewal buffer memory based in the images serve system of distributed storage, enters step s210;
S210, picture retrieval terminate.
Described cache management comprises the steps:
S31, acquisition picture retrieval result, enter step s32;
S32, according to characteristic similarity, merging treatment is carried out to picture retrieval result, enter step s33;
S33, after merging treatment picture composition file stored in HDFS, enter step s34;
S34, according to the change upgrading picture-storage in HDFS in described step s33 and bring, upgrade the index of picture in large files and the mapping relations of picture feature and picture-storage address.
Described optimization process carries out off-line partition duplicate removal and integrate processing to obtain the file optimization of new large files and index and to re-start image analysis processing to improve the picture optimization of picture-storage performance and picture feature similarity based on the image optimization module in the images serve system of distributed storage to the picture in the large files stored in described HDFS by described to the large files in HDFS based on the off-line files optimization module in the images serve system of distributed storage by described.
Adopt the images serve system and method based on distributed storage of the present invention, by three similar pictures consolidation strategies, namely be optimized by image cache server, off-line files processing module, the storage of feedback processing modules to similar pictures, improve the characteristic similarity of picture, to improve the accuracy and efficiency of picture retrieval.Meanwhile, with the image data in HDFS, carry out the training of convolutional neural networks, the network obtained can go to extract picture feature; Cluster or other Analysis Service is carried out by MapReduce or Spark; Image procossing is carried out to Query Result, optimizes picture-storage greatly.
Accompanying drawing explanation
Fig. 1 is the framework composition schematic diagram of the images serve system based on distributed storage of the present invention;
Fig. 2 is the images serve method main flow schematic diagram based on distributed storage of the present invention;
Fig. 3 is of the present invention based on the picture-storage management process schematic diagram in the images serve method of distributed storage;
Fig. 4 is of the present invention based on the picture retrieval management process schematic diagram in the images serve method of distributed storage;
Fig. 5 is of the present invention based on the cache management schematic flow sheet in the images serve method of distributed storage.
Embodiment
For a better understanding of the present invention, the present invention is described in detail below in conjunction with accompanying drawing.
As shown in Figure 1, a kind of images serve system based on distributed storage of the present invention, comprise the picture receiver module receiving the picture that needs store and needs are retrieved, the picture received described picture receiver module extracts the characteristics analysis module of picture feature according to image characteristics extraction algorithm, the picture feature of the picture that the needs described characteristics analysis module extracted store carries out the feature database stored, the picture that being received by described picture receiver module needs to store extracts stored in buffer memory and according to described characteristics analysis module and needs the picture feature similarity of the picture of storage to carry out to picture the image cache server merged of classifying, for storing the HDFS of the large files be made up of the picture of described image cache block merging, picture feature and the feature in described feature database of the picture of the needs retrieval of described characteristics analysis module being extracted carry out the characteristic matching module of mating, the caching management module of distributed caching in order to directly using next time is carried out to the result for retrieval that retrieving image produces, export the picture output module of picture result for retrieval, according to picture retrieval result carry out similar pictures merging treatment and again stored in described HDFS to optimize the feedback processing modules that picture-storage improves the picture analogies degree in same large files, carry out off-line partition duplicate removal to the large files in described HDFS and integrate process optimizing module with the off-line files obtaining new large files and index, graphical analysis is carried out to the picture stored in described HDFS and processes the image optimization module stored with optimized image.
Preferably, the large files that described image cache server forms after merging for similar pictures sets up the index of picture and large files; Described image cache server sets up the mapping relations of picture feature and picture-storage address according to picture feature.
Preferably, the picture feature of each picture of described storage and picture-storage address one_to_one corresponding.
Preferably, feature in the picture feature of the picture of the needs retrieval that described characteristic matching module is extracted according to described characteristics analysis module and described feature database compares, find the most similar N number of feature (N>=1), according to the one-to-one relationship of feature and picture-storage address, find picture address corresponding to described N number of feature to find corresponding large files, and according to the index of picture in large files, take out concrete N number of picture.
Preferably, described image optimization module carries out graphical analysis and process to the picture in described HDFS by MapReduce.
Preferably, described image optimization module carries out graphical analysis and process to the picture in described HDFS by Spark.
Preferably, described image optimization module carries out convolutional neural networks training, cluster analysis, deblurring, denoising and mist elimination process to the picture in described HDFS.
A kind of images serve method based on distributed storage of the present invention, by realizing picture-storage management, picture retrieval management, cache management and optimization process based on the images serve system of distributed storage; Wherein, described picture-storage management comprises the steps:
S11, received the picture needing to store by the described picture receiver module based in the images serve system of distributed storage, enter step s2;
S12, by the described feature extracting the picture received in described step s1 based on the characteristics analysis module in the images serve system of distributed storage, simultaneously by the picture that receives stored in described based on the image cache server in the images serve system of distributed storage, enter step s3 simultaneously;
S13, by the picture feature extracted in described step s2 stored in described based on the feature database in the images serve system of distributed storage, enter step s4;
S14, the picture feature extracted in described step s2 is sent to image cache server, according to picture feature, merging treatment is carried out to similar pictures by described image cache server, by the large files that is made up of the picture after merging treatment stored in HDFS, enter step s5;
S15, set up the index of picture in large files and the mapping relations of picture feature and picture-storage address by described image cache server.
Described picture retrieval management comprises the steps:
The picture of s21, the retrieval of input needs, receives picture by described picture receiver module, enters step s22;
S22, according to image characteristics extraction algorithm, signature analysis is carried out to the picture to be retrieved received in described step s21 by described characteristics analysis module, extract the picture feature of picture to be retrieved, enter step s23;
S23, the picture feature of the picture to be retrieved analyzed in described step s22 to be mated with the feature in described feature database based on the characteristics analysis module in the images serve system of distributed storage by described, in feature database, mate similar feature according to the picture feature of picture to be retrieved, enter step s24;
S24, find the mapping relations of picture feature and picture address according to the feature of mating in described step s23, find all large files relevant in HDFS according to these mapping relations, enter step s25;
S25, the picture index in large files corresponding according to picture feature, obtain, with the concrete picture of the characteristic matching of mating in described step s23, entering step s26;
S26, judge that whether in the buffer the concrete picture that to obtain in described step s25, if so, enters step s27; If not, enter step s28;
S27, diameter extract picture from buffer memory to be shown as result for retrieval based on the picture output module in the images serve system of distributed storage by described; Enter step s210;
S28, from HDFS, extract the picture that obtains in described step s25 and shown as result for retrieval by described picture output module, entering step s29;
S29, according to showing that result is by the described caching management module renewal buffer memory based in the images serve system of distributed storage, enters step s210;
S210, picture retrieval terminate.
Described cache management comprises the steps:
S31, acquisition picture retrieval result, enter step s32;
S32, according to characteristic similarity, merging treatment is carried out to picture retrieval result, enter step s33;
S33, after merging treatment picture composition file stored in HDFS, enter step s34;
S34, according to the change upgrading picture-storage in HDFS in described step s33 and bring, upgrade the index of picture in large files and the mapping relations of picture feature and picture-storage address.
Described optimization process carries out off-line partition duplicate removal and integrate processing to obtain the file optimization of new large files and index and to re-start image analysis processing to improve the picture optimization of picture-storage performance and picture feature similarity based on the image optimization module in the images serve system of distributed storage to the picture in the large files stored in described HDFS by described to the large files in HDFS based on the off-line files optimization module in the images serve system of distributed storage by described.
Preferably, the picture feature of each picture of described storage and picture-storage address one_to_one corresponding.
Preferably, feature in the picture feature of the picture of the needs retrieval of being extracted according to described characteristics analysis module by described characteristic matching module and described feature database is compared, find the most similar N number of feature (N>=1), the most similar N number of feature found and the one-to-one relationship of picture-storage address, find picture address corresponding to described N number of feature to find corresponding large files, and according to the index of picture in large files, take out concrete N number of picture.
Preferably, by described image optimization module, by MapReduce technology, graphical analysis and process are carried out to the picture in described HDFS.
Preferably, by described image optimization module, by Spark technology, graphical analysis and process are carried out to the picture in described HDFS.
Preferably, described image optimization module carries out convolutional neural networks training, cluster analysis, deblurring, denoising and mist elimination process to the picture in described HDFS.
Adopt the images serve system and method based on distributed storage of the present invention, by three similar pictures consolidation strategies, namely be optimized by image cache server, off-line files processing module, the storage of feedback processing modules to similar pictures, improve the characteristic similarity of the picture be stored in same large files, to improve the accuracy and efficiency of picture retrieval.Meanwhile, utilize the image data in HDFS, carry out the training of convolutional neural networks, the network obtained can go to extract picture feature; Cluster or other Analysis Service is carried out by MapReduce or Spark; Image procossing is carried out to Query Result, optimizes picture-storage greatly.Adopt system and method for the present invention, effectively improve the performance of images serve system, improve the accuracy of service and picture retrieval
The above; be only the present invention's preferably embodiment; but protection scope of the present invention is not limited thereto; anyly be familiar with those skilled in the art in the technical scope that the present invention discloses; be equal to according to technical scheme of the present invention and inventive concept thereof and replace or change, all should be encompassed within protection scope of the present invention.
Claims (13)
1. the images serve system based on distributed storage, it is characterized in that, the described images serve system based on distributed storage comprises the picture receiver module receiving the picture that needs store and needs are retrieved, the picture received described picture receiver module extracts the characteristics analysis module of picture feature according to image characteristics extraction algorithm, the picture feature of the picture that the needs described characteristics analysis module extracted store carries out the feature database stored, the picture that being received by described picture receiver module needs to store extracts stored in buffer memory and according to described characteristics analysis module and needs the picture feature similarity of the picture of storage to carry out to picture the image cache server merged of classifying, for storing the HDFS of the large files be made up of the picture of described image cache block merging, picture feature and the feature in described feature database of the picture of the needs retrieval of described characteristics analysis module being extracted carry out the characteristic matching module of mating, the caching management module of distributed caching in order to directly using next time is carried out to the result for retrieval that retrieving image produces, export the picture output module of picture result for retrieval, according to picture retrieval result carry out similar pictures merging treatment and again stored in described HDFS to optimize the feedback processing modules that picture-storage improves the picture analogies degree in same large files, carry out off-line partition duplicate removal to the large files in described HDFS and integrate process optimizing module with the off-line files obtaining new large files and index, graphical analysis is carried out to the picture stored in described HDFS and processes the image optimization module stored with optimized image.
2. the images serve system based on distributed storage according to claim 1, is characterized in that, after described image cache server merges for similar pictures, the large files of composition sets up the index of picture and large files; Described image cache server sets up the mapping relations of picture feature and picture-storage address according to picture feature.
3. the images serve system based on distributed storage according to claim 2, is characterized in that, the picture feature of each picture of described storage and picture-storage address one_to_one corresponding.
4. the images serve system based on distributed storage according to claim 1, it is characterized in that, feature in the picture feature of the picture of the needs retrieval that described characteristic matching module is extracted according to described characteristics analysis module and described feature database compares, find the most similar N number of feature (N>=1), according to the one-to-one relationship of feature and picture-storage address, find picture address corresponding to described N number of feature to find corresponding large files, and according to the index of picture in large files, take out concrete N number of picture.
5. according to claim 1 the images serve system based on distributed storage, it is characterized in that, described image optimization module carries out graphical analysis and process to the picture in described HDFS by MapReduce.
6. the images serve system based on distributed storage according to claim 1, is characterized in that, described image optimization module carries out graphical analysis and process to the picture in described HDFS by Spark.
7. the images serve system based on distributed storage according to claim 5 or 6, is characterized in that, described image optimization module carries out convolutional neural networks training, cluster analysis, deblurring, denoising and mist elimination process to the picture in described HDFS.
8. the images serve method based on distributed storage, it is characterized in that, the described images serve method based on distributed storage is by realizing picture-storage management, picture retrieval management, cache management and optimization process based on the images serve system of distributed storage;
Described picture-storage management comprises the steps:
S11, received the picture needing to store by the described picture receiver module based in the images serve system of distributed storage, enter step s2;
S12, by the described feature extracting the picture received in described step s1 based on the characteristics analysis module in the images serve system of distributed storage, simultaneously by the picture that receives stored in described based on the image cache server in the images serve system of distributed storage, enter step s3 simultaneously;
S13, by the picture feature extracted in described step s2 stored in described based on the feature database in the images serve system of distributed storage, enter step s4;
S14, the picture feature extracted in described step s2 is sent to image cache server, according to picture feature, merging treatment is carried out to similar pictures by described image cache server, by the large files that is made up of the picture after merging treatment stored in HDFS, enter step s5;
S15, set up the index of picture in large files and the mapping relations of picture feature and picture-storage address by described image cache server;
Described picture retrieval management comprises the steps:
The picture of s21, the retrieval of input needs, receives picture by described picture receiver module, enters step s22;
S22, according to image characteristics extraction algorithm, signature analysis is carried out to the picture to be retrieved received in described step s21 by described characteristics analysis module, extract the picture feature of picture to be retrieved, enter step s23;
S23, the picture feature of the picture to be retrieved analyzed in described step s22 to be mated with the feature in described feature database based on the characteristics analysis module in the images serve system of distributed storage by described, in feature database, mate similar feature according to the picture feature of picture to be retrieved, enter step s24;
S24, find the mapping relations of picture feature and picture address according to the feature of mating in described step s23, find all large files relevant in HDFS according to these mapping relations, enter step s25;
S25, the picture index in large files corresponding according to picture feature, obtain, with the concrete picture of the characteristic matching of mating in described step s23, entering step s26;
S26, judge that whether in the buffer the concrete picture that to obtain in described step s25, if so, enters step s27; If not, enter step s28;
S27, diameter extract picture from buffer memory to be shown as result for retrieval based on the picture output module in the images serve system of distributed storage by described; Enter step s210;
S28, from HDFS, extract the picture that obtains in described step s25 and shown as result for retrieval by described picture output module, entering step s29;
S29, according to showing that result is by the described caching management module renewal buffer memory based in the images serve system of distributed storage, enters step s210;
S210, picture retrieval terminate;
Described cache management comprises the steps:
S31, acquisition picture retrieval result, enter step s32;
S32, according to characteristic similarity, merging treatment is carried out to picture retrieval result, enter step s33;
S33, after merging treatment picture composition file stored in HDFS, enter step s34;
S34, according to the change upgrading picture-storage in HDFS in described step s33 and bring, upgrade the index of picture in large files and the mapping relations of picture feature and picture-storage address;
Described optimization process carries out off-line partition duplicate removal and integrate processing to obtain the file optimization of new large files and index and to re-start image analysis processing to improve the picture optimization of picture-storage performance and picture feature similarity based on the image optimization module in the images serve system of distributed storage to the picture in the large files stored in described HDFS by described to the large files in HDFS based on the off-line files optimization module in the images serve system of distributed storage by described.
9. the images serve method based on distributed storage according to claim 8, is characterized in that, the picture feature of each picture of described storage and picture-storage address one_to_one corresponding.
10. the images serve method based on distributed storage according to claim 8, it is characterized in that, feature in the picture feature of the picture of the needs retrieval of being extracted according to described characteristics analysis module by described characteristic matching module and described feature database is compared, find the most similar N number of feature (N>=1), the most similar N number of feature found and the one-to-one relationship of picture-storage address, find picture address corresponding to described N number of feature to find corresponding large files, and according to the index of picture in large files, take out concrete N number of picture.
The 11. images serve methods based on distributed storage according to claim 8, is characterized in that, carry out graphical analysis and process by described image optimization module to the picture in described HDFS by MapReduce.
The 12. images serve methods based on distributed storage according to claim 8, is characterized in that, carry out graphical analysis and process by described image optimization module to the picture in described HDFS by Spark.
13. images serve methods based on distributed storage according to claim 11 or 12, it is characterized in that, described image optimization module carries out convolutional neural networks training, cluster analysis, deblurring, denoising and mist elimination process to the picture in described HDFS.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510886086.7A CN105550222B (en) | 2015-12-07 | 2015-12-07 | A kind of images serve system and method based on distributed storage |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510886086.7A CN105550222B (en) | 2015-12-07 | 2015-12-07 | A kind of images serve system and method based on distributed storage |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105550222A true CN105550222A (en) | 2016-05-04 |
| CN105550222B CN105550222B (en) | 2019-04-05 |
Family
ID=55829411
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510886086.7A Active CN105550222B (en) | 2015-12-07 | 2015-12-07 | A kind of images serve system and method based on distributed storage |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105550222B (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106383891A (en) * | 2016-09-22 | 2017-02-08 | 重庆理工大学 | Deep hash-based medical image distributed retrieval method |
| CN106874219A (en) * | 2016-12-23 | 2017-06-20 | 深圳云天励飞技术有限公司 | A kind of data dispatching method of convolutional neural networks, system and computer equipment |
| CN107315809A (en) * | 2017-06-27 | 2017-11-03 | 山东浪潮通软信息科技有限公司 | A kind of group's news data preprocess method based on Spark platforms |
| CN107506362A (en) * | 2016-11-23 | 2017-12-22 | 上海大学 | Image classification based on customer group optimization imitates brain storage method |
| CN108038847A (en) * | 2017-12-05 | 2018-05-15 | 国网内蒙古东部电力有限公司 | Transformer inspection digital image recognition and fault detection system based on deep learning |
| CN108664495A (en) * | 2017-03-28 | 2018-10-16 | 广东神马搜索科技有限公司 | A kind of search method, device and the equipment of multi-segment index parallel search and point counting |
| CN108776758A (en) * | 2018-04-13 | 2018-11-09 | 西安电子科技大学 | The block level data De-weight method of dynamic ownership management is supported in a kind of storage of mist |
| CN108846142A (en) * | 2018-07-12 | 2018-11-20 | 南方电网调峰调频发电有限公司 | A kind of Text Clustering Method, device, equipment and readable storage medium storing program for executing |
| CN110019870A (en) * | 2017-12-29 | 2019-07-16 | 浙江宇视科技有限公司 | The image search method and system of image cluster based on memory |
| CN111242321A (en) * | 2019-04-18 | 2020-06-05 | 中科寒武纪科技股份有限公司 | Data processing method and related product |
| CN111597375A (en) * | 2020-05-19 | 2020-08-28 | 清华大学 | Picture retrieval method based on similar picture group representative feature vector and related equipment |
| CN112200740A (en) * | 2020-10-08 | 2021-01-08 | 华中科技大学 | Image blocking and de-duplication method and system based on image edge detection |
| CN115129921A (en) * | 2022-06-30 | 2022-09-30 | 重庆紫光华山智安科技有限公司 | Picture retrieval method and device, electronic equipment and computer-readable storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090172050A1 (en) * | 2008-01-02 | 2009-07-02 | Sandisk Il Ltd. | Dual representation of stored digital content |
| CN102436491A (en) * | 2011-11-08 | 2012-05-02 | 张三明 | System and method used for searching huge amount of pictures and based on BigBase |
| CN103678665A (en) * | 2013-12-24 | 2014-03-26 | 焦点科技股份有限公司 | Heterogeneous large data integration method and system based on data warehouses |
| CN104199899A (en) * | 2014-08-26 | 2014-12-10 | 浪潮(北京)电子信息产业有限公司 | Method and device for storing massive pictures based on Hbase |
| US20150125133A1 (en) * | 2013-11-06 | 2015-05-07 | Konkuk University Industrial Cooperation Corp. | Method for transcoding multimedia, and hadoop-based multimedia transcoding system for performing the method |
-
2015
- 2015-12-07 CN CN201510886086.7A patent/CN105550222B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090172050A1 (en) * | 2008-01-02 | 2009-07-02 | Sandisk Il Ltd. | Dual representation of stored digital content |
| CN102436491A (en) * | 2011-11-08 | 2012-05-02 | 张三明 | System and method used for searching huge amount of pictures and based on BigBase |
| US20150125133A1 (en) * | 2013-11-06 | 2015-05-07 | Konkuk University Industrial Cooperation Corp. | Method for transcoding multimedia, and hadoop-based multimedia transcoding system for performing the method |
| CN103678665A (en) * | 2013-12-24 | 2014-03-26 | 焦点科技股份有限公司 | Heterogeneous large data integration method and system based on data warehouses |
| CN104199899A (en) * | 2014-08-26 | 2014-12-10 | 浪潮(北京)电子信息产业有限公司 | Method and device for storing massive pictures based on Hbase |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106383891B (en) * | 2016-09-22 | 2019-04-23 | 重庆理工大学 | A kind of medical image distributed search method based on depth Hash |
| CN106383891A (en) * | 2016-09-22 | 2017-02-08 | 重庆理工大学 | Deep hash-based medical image distributed retrieval method |
| CN107506362A (en) * | 2016-11-23 | 2017-12-22 | 上海大学 | Image classification based on customer group optimization imitates brain storage method |
| CN107506362B (en) * | 2016-11-23 | 2021-02-23 | 上海大学 | Image classification brain-imitation storage method based on user group optimization |
| CN106874219A (en) * | 2016-12-23 | 2017-06-20 | 深圳云天励飞技术有限公司 | A kind of data dispatching method of convolutional neural networks, system and computer equipment |
| CN108664495A (en) * | 2017-03-28 | 2018-10-16 | 广东神马搜索科技有限公司 | A kind of search method, device and the equipment of multi-segment index parallel search and point counting |
| CN107315809A (en) * | 2017-06-27 | 2017-11-03 | 山东浪潮通软信息科技有限公司 | A kind of group's news data preprocess method based on Spark platforms |
| CN108038847A (en) * | 2017-12-05 | 2018-05-15 | 国网内蒙古东部电力有限公司 | Transformer inspection digital image recognition and fault detection system based on deep learning |
| CN110019870A (en) * | 2017-12-29 | 2019-07-16 | 浙江宇视科技有限公司 | The image search method and system of image cluster based on memory |
| CN108776758A (en) * | 2018-04-13 | 2018-11-09 | 西安电子科技大学 | The block level data De-weight method of dynamic ownership management is supported in a kind of storage of mist |
| CN108776758B (en) * | 2018-04-13 | 2021-08-17 | 西安电子科技大学 | Block-level data deduplication method supporting dynamic ownership management in fog storage |
| CN108846142A (en) * | 2018-07-12 | 2018-11-20 | 南方电网调峰调频发电有限公司 | A kind of Text Clustering Method, device, equipment and readable storage medium storing program for executing |
| CN111242321A (en) * | 2019-04-18 | 2020-06-05 | 中科寒武纪科技股份有限公司 | Data processing method and related product |
| US11762690B2 (en) | 2019-04-18 | 2023-09-19 | Cambricon Technologies Corporation Limited | Data processing method and related products |
| CN111242321B (en) * | 2019-04-18 | 2023-09-26 | 中科寒武纪科技股份有限公司 | Data processing method and related product |
| CN111597375A (en) * | 2020-05-19 | 2020-08-28 | 清华大学 | Picture retrieval method based on similar picture group representative feature vector and related equipment |
| CN112200740A (en) * | 2020-10-08 | 2021-01-08 | 华中科技大学 | Image blocking and de-duplication method and system based on image edge detection |
| CN115129921A (en) * | 2022-06-30 | 2022-09-30 | 重庆紫光华山智安科技有限公司 | Picture retrieval method and device, electronic equipment and computer-readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105550222B (en) | 2019-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105550222A (en) | Distributed storage-based image service system and method | |
| CN111522986B (en) | Image retrieval method, device, equipment and medium | |
| CN109034993B (en) | Account checking method, account checking equipment, account checking system and computer readable storage medium | |
| US20210233319A1 (en) | Context-aware tagging for augmented reality environments | |
| CN107832406B (en) | Method, device, equipment and storage medium for removing duplicate entries of mass log data | |
| CN103324552A (en) | Two-stage single-instance data de-duplication backup method | |
| CN102880854B (en) | Distributed processing and Hash mapping-based outdoor massive object identification method and system | |
| CN110489475B (en) | Multi-source heterogeneous data processing method, system and related device | |
| CN110311953B (en) | Media data uploading and storing system and method | |
| US10733454B2 (en) | Transformation of video streams | |
| CN107317838A (en) | A kind of astronomical metadata archiving method and system based on stream data processing framework | |
| US10783127B2 (en) | Componentized data storage | |
| CN111625552B (en) | Data collection method, device, equipment and readable storage medium | |
| US9519543B2 (en) | Method and apparatus for image search using feature point | |
| KR102633433B1 (en) | Method and device for classifying face image, electronic device and storage medium | |
| KR101329102B1 (en) | augmented reality - image retrieval system using layout descriptor and image feature. | |
| US20120163475A1 (en) | Fast matching system for digital video | |
| US11509669B2 (en) | Network data timeline | |
| CN105227386A (en) | For method, the Apparatus and system of point population statistics online user number | |
| CN114398520A (en) | Data retrieval method, system, device, electronic equipment and storage medium | |
| CN107590233B (en) | File management method and device | |
| CN103778258A (en) | Method for sending and receiving data of database, client terminal and server | |
| CN106557503A (en) | A kind of method and system of image retrieval | |
| CN106708876B (en) | Similar video retrieval method and system based on Lucene | |
| Ou et al. | Communication-efficient multi-view keyframe extraction in distributed video sensors |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |