CN109002537B - Information processing system and method based on deep learning - Google Patents
Information processing system and method based on deep learning Download PDFInfo
- Publication number
- CN109002537B CN109002537B CN201810799283.9A CN201810799283A CN109002537B CN 109002537 B CN109002537 B CN 109002537B CN 201810799283 A CN201810799283 A CN 201810799283A CN 109002537 B CN109002537 B CN 109002537B
- Authority
- CN
- China
- Prior art keywords
- information
- module
- image information
- image
- feature points
- 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.)
- Active
Links
Images
Abstract
The invention belongs to the technical field of information processing, and particularly relates to an information processing system and method based on deep learning, wherein the information processing system based on deep learning comprises an input module, a processor, a behavior log library, an image query module, a memory, a keyword search module and a feedback module, and the processing method of the information processing system based on deep learning comprises the following specific steps: a) inputting a target keyword, b) establishing a search object according to the keyword and a behavior log, c) further screening the searched object result, d) feeding back the finally screened object collection to a user, and pushing or searching a target file in a mode of combining the keyword with the behavior log, so that the time can be shortened and the accuracy can be improved; through a mode of multiple screening, more useless files can be filtered, and the simplicity of the search result is kept; the image information is screened by the user subjectivity, so that the humanization is high.
Description
Technical Field
The invention relates to the technical field of information processing, in particular to an information processing system and method based on deep learning.
Background
"electronic information" is a word frequently appearing in recent years, and is a word that becomes informatics in the context of rapid development of computer technology, communication technology, and high-density storage technology and widespread use in various fields. Electronic information engineering is a subject of electronic information control and information processing by applying modern technologies such as computers, and mainly researches acquisition and processing of information, design, development, application and integration of electronic equipment and an information system.
With the deepening of the information technology, humanization is more and more embodied, target information is relatively slowly inquired in network browsing and information inquiry, sometimes target information is difficult to accurately inquire, humanization cannot be achieved in pushing of data information, so that target file inquiry time is long, and working efficiency is low.
Disclosure of Invention
The invention aims to provide an information processing system and method based on deep learning, and aims to solve the problems that the existing processing system proposed in the background technology cannot achieve humanization for pushing data information, so that the target file query time is long, and the working efficiency is low.
In order to achieve the purpose, the invention provides the following technical scheme: an information processing system based on deep learning comprises an input module, a processor, a behavior log library, an image query module, a memory, a keyword search module and a feedback module, wherein the output end of the input module is connected with the input end of the processor;
the input hardware end of the input module inputs the feature points of the target through the input module, the feature points are used as key words, and the number of the feature points is single or multiple;
the processor is used as a data processing center of the deep learning-based information processing system and is responsible for receiving information and processing data;
the behavior log library records historical network access objects and access frequency of users, and user access information recorded by the behavior log library is used as one of bases for judging user behaviors;
the image query module comprises a storage unit, an image acquisition module, a comparison unit and a processing chip, the data port of the storage unit is respectively connected with the comparison unit, the image acquisition module and the processor through data lines, the data port of the comparison unit is connected with the processing chip through a data line, the data port of the image acquisition module is connected with the memory through a data line, the image acquisition module acquires image information, the storage unit stores input feature points transmitted by the processor, the processing chip identifies the content of the feature points and the image information acquired by the image acquisition module, comparing the image information with the characteristic points through a comparison unit, judging whether the acquired image information is consistent with the required object or has common points, reserving the image information which is consistent with the required object or has common points, and filtering out the image information which does not meet the requirement;
the memory is used for storing searched data information;
the keyword searching module is accessed to the Internet, receives the input feature points sent by the processor, and asks for an object on the network according to the input feature points;
the feedback module feeds back image information meeting the image information requirement that the image information is consistent with a required object or has a common point to a user.
Preferably, the input module is a mouse, a keyboard or a touch input screen.
Preferably, the behavior log library sets a time range for storage, the time range is 3 months to 6 months, and the behavior log library provides recorded information arranged according to a time sequence.
Preferably, the order of the search objects of the keyword search module is from an address with high browsing frequency to an address with low browsing frequency.
Preferably, the processing method of the deep learning-based information processing system includes the following specific steps:
a) inputting a target keyword: according to target image information required by a user, corresponding feature points of an input image are carried out, the feature points are used as description keywords of the target image information, and the feature points are used as one of bases for searching a target image;
b) establishing a search object from the keywords and the behavior log: calling user browsing information recorded in a behavior log, searching one by one according to the frequency from high to low by taking an address with higher frequency as a main searching destination, inquiring images with higher coincidence degree described by the keywords, eliminating the images with lower coincidence degree, and storing the images with higher coincidence degree;
c) and (3) further screening on the searched object results: further screening the image information with higher contact ratio stored in the step b), screening the images with the contact ratio of more than 50%, and deleting the image information which does not meet the conditions;
d) and feeding back the finally screened object collection to a user: feeding back the image information finally reserved in the step c) to the user terminal.
Compared with the prior art, the invention has the beneficial effects that:
1) the pushing or target file searching is carried out in a mode of combining the keywords with the behavior logs, so that the time can be shortened, and the accuracy can be improved;
2) through a mode of multiple screening, more useless files can be filtered, and the simplicity of the search result is kept;
3) the image information is screened by the user subjectivity, so that the humanization is high.
Drawings
FIG. 1 is a schematic block diagram of the system of the present invention;
FIG. 2 is a functional block diagram of an image acquisition module of the present invention;
FIG. 3 is a flow chart of information processing according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1-3, the present invention provides a technical solution: an information processing system based on deep learning comprises an input module, a processor, a behavior log library, an image query module, a memory, a keyword search module and a feedback module, and is characterized in that the output end of the input module is connected with the input end of the processor, a data port of the processor is respectively connected with the behavior log library and the image query module through a data line, a data port of the image query module is connected with the memory through a data line, and the output end of the processor is respectively connected with the input ends of the processor, the keyword search module and the feedback module through data lines;
the input hardware end of the input module inputs the feature points of the target through the input module, the feature points are used as key words, and the number of the feature points is single or multiple;
the processor is used as a data processing center of the deep learning-based information processing system and is responsible for receiving information and processing data;
the behavior log library records historical network access objects and access frequency of users, and user access information recorded by the behavior log library is used as one of bases for judging user behaviors;
the image query module comprises a storage unit, an image acquisition module, a comparison unit and a processing chip, the data port of the storage unit is respectively connected with the comparison unit, the image acquisition module and the processor through data lines, the data port of the comparison unit is connected with the processing chip through a data line, the data port of the image acquisition module is connected with the memory through a data line, the image acquisition module acquires image information, the storage unit stores input feature points transmitted by the processor, the processing chip identifies the content of the feature points and the image information acquired by the image acquisition module, comparing the image information with the characteristic points through a comparison unit, judging whether the acquired image information is consistent with the required object or has common points, reserving the image information which is consistent with the required object or has common points, and filtering out the image information which does not meet the requirement;
the memory is used for storing searched data information;
the keyword searching module is accessed to the Internet, receives the input feature points sent by the processor, and asks for an object on the network according to the input feature points;
the feedback module feeds back image information meeting the image information requirement that the image information is consistent with a required object or has a common point to a user.
The input module is a mouse, a keyboard or a touch input screen and is a common input device, the behavior log library sets a stored time range, the time range is 3 months to 6 months, corresponding selection and adjustment are carried out according to specific user using behaviors, the behavior log library provides recorded information and arranges the recorded information according to a time sequence, the latest behaviors can be used as a main basis, and the sequence of search objects of the keyword search module is from an address with high browsing frequency to an address with low browsing frequency.
The processing method of the information processing system based on deep learning comprises the following specific steps:
a) inputting a target keyword: according to target image information required by a user, corresponding feature points of an input image are carried out, the feature points are used as description keywords of the target image information, and the feature points are used as one of bases for searching a target image;
b) establishing a search object from the keywords and the behavior log: calling user browsing information recorded in a behavior log, searching one by one according to the frequency from high to low by taking an address with higher frequency as a main searching destination, inquiring images with higher coincidence degree described by the keywords, eliminating the images with lower coincidence degree, and storing the images with higher coincidence degree;
s1, storing the image with high coincidence degree, and segmenting the image to obtain an initial image region set: r { R2, R1, · · rn };
s2: initializing a similarity set S phi, and initializing a possible result L phi;
s3: calculating the similarity between the regions, and adjusting the price of the regions to a similarity set S;
s4: finding two regions ri and rj corresponding to the maximum value from the similarity set, combining ri and rj into a region rt in the mode L, and performing S4: removing information related to ri and rj from the similarity set and the R, calculating the similarity between rt and an adjacent region of rt, adding the result to the similarity set S, and simultaneously adding a new region rt to the region set R;
s5: repeating the step S4 for 4 times, and storing the final result;
c) and (3) further screening on the searched object results: further screening the image information with higher contact ratio stored in the step b), screening the images with the contact ratio of more than 50%, and deleting the image information which does not meet the conditions;
d) and feeding back the finally screened object collection to a user: feeding back the image information finally reserved in the step c) to the user terminal.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (4)
1. An information processing method based on deep learning is characterized in that: the method is realized by the following system: the system comprises an input module, a processor, a behavior log library, an image query module, a memory, a keyword search module and a feedback module, wherein the output end of the input module is connected with the input end of the processor;
the input hardware end of the input module inputs the feature points of the target through the input module, the feature points are used as key words, and the number of the feature points is single or multiple;
the processor is used as a data processing center of the deep learning-based information processing system and is responsible for receiving information and processing data;
the behavior log library records historical network access objects and access frequency of users, and user access information recorded by the behavior log library is used as one of bases for judging user behaviors;
the image query module comprises a storage unit, an image acquisition module, a comparison unit and a processing chip, the data port of the storage unit is respectively connected with the comparison unit, the image acquisition module and the processor through data lines, the data port of the comparison unit is connected with the processing chip through a data line, the data port of the image acquisition module is connected with the memory through a data line, the image acquisition module acquires image information, the storage unit stores input feature points transmitted by the processor, the processing chip identifies the content of the feature points and the image information acquired by the image acquisition module, comparing the image information with the characteristic points through a comparison unit, judging whether the acquired image information is consistent with the required object or has common points, reserving the image information which is consistent with the required object or has common points, and filtering out the image information which does not meet the requirement;
the memory is used for storing searched data information;
the keyword searching module is accessed to the Internet, receives the input feature points sent by the processor, and asks for an object on the network according to the input feature points;
the feedback module feeds back image information meeting the image information requirement that the image information is consistent with a required object or has a common point to a user;
the method comprises the following specific steps:
a) inputting a target keyword: according to target image information required by a user, corresponding feature points of an input image are carried out, the feature points are used as description keywords of the target image information, and the feature points are used as one of bases for searching a target image;
b) establishing a search object from the keywords and the behavior log: calling user browsing information recorded in a behavior log, searching one by one according to the frequency from high to low by taking an address with higher frequency as a main searching destination, inquiring images with higher coincidence degree described by the keywords, eliminating the images with lower coincidence degree, and storing the images with higher coincidence degree;
c) and (3) further screening on the searched object results: further screening the image information with higher contact ratio stored in the step b), screening the images with the contact ratio of more than 50%, and deleting the image information which does not meet the conditions;
d) and feeding back the finally screened object collection to a user: feeding back the image information finally reserved in the step c) to the user terminal.
2. The information processing method based on deep learning according to claim 1, wherein: the input module is a mouse, a keyboard or a touch input screen.
3. The information processing method based on deep learning according to claim 1, wherein: the behavior log library sets a storage time range, the time range is 3 months to 6 months, and the behavior log library provides recorded information which is arranged according to a time sequence.
4. The information processing method based on deep learning according to claim 1, wherein: the sequence of the search objects of the keyword search module is from the address with high browsing frequency to the address with low browsing frequency.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810799283.9A CN109002537B (en) | 2018-07-19 | 2018-07-19 | Information processing system and method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810799283.9A CN109002537B (en) | 2018-07-19 | 2018-07-19 | Information processing system and method based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109002537A CN109002537A (en) | 2018-12-14 |
| CN109002537B true CN109002537B (en) | 2021-05-14 |
Family
ID=64596939
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810799283.9A Active CN109002537B (en) | 2018-07-19 | 2018-07-19 | Information processing system and method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109002537B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115168684B (en) * | 2022-09-05 | 2022-11-22 | 南昌工程学院 | Financial archive management method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102508901A (en) * | 2011-11-07 | 2012-06-20 | 康佳集团股份有限公司 | Content-based massive image search method and content-based massive image search system |
| CN107609152A (en) * | 2017-09-22 | 2018-01-19 | 百度在线网络技术（北京）有限公司 | Method and apparatus for expanding query formula |
| US9892538B1 (en) * | 2016-10-06 | 2018-02-13 | International Business Machines Corporation | Rebuilding images based on historical image data |
| CN108288208A (en) * | 2017-08-11 | 2018-07-17 | 腾讯科技（深圳）有限公司 | The displaying object of image content-based determines method, apparatus, medium and equipment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10026021B2 (en) * | 2016-09-27 | 2018-07-17 | Facebook, Inc. | Training image-recognition systems using a joint embedding model on online social networks |
-
2018
- 2018-07-19 CN CN201810799283.9A patent/CN109002537B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102508901A (en) * | 2011-11-07 | 2012-06-20 | 康佳集团股份有限公司 | Content-based massive image search method and content-based massive image search system |
| US9892538B1 (en) * | 2016-10-06 | 2018-02-13 | International Business Machines Corporation | Rebuilding images based on historical image data |
| CN108288208A (en) * | 2017-08-11 | 2018-07-17 | 腾讯科技（深圳）有限公司 | The displaying object of image content-based determines method, apparatus, medium and equipment |
| CN107609152A (en) * | 2017-09-22 | 2018-01-19 | 百度在线网络技术（北京）有限公司 | Method and apparatus for expanding query formula |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109002537A (en) | 2018-12-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2013329525B2 (en) | System and method for recursively traversing the internet and other sources to identify, gather, curate, adjudicate, and qualify business identity and related data | |
| US8924373B2 (en) | Query plans with parameter markers in place of object identifiers | |
| US20060253550A1 (en) | System and method for providing data for decision support | |
| CN107704202B (en) | Method and device for quickly reading and writing data | |
| US11907659B2 (en) | Item recall method and system, electronic device and readable storage medium | |
| AU2014253497B2 (en) | Entity resolution from documents | |
| CN112434027A (en) | Indexing method and device for multi-dimensional data, computer equipment and storage medium | |
| Lomet | Digital B-trees | |
| CN109002537B (en) | Information processing system and method based on deep learning | |
| CN113821630A (en) | Data clustering method and device | |
| CN107066587A (en) | A kind of efficient Mining Frequent Itemsets based on group chained list | |
| CN115858471A (en) | Service data change recording method, device, computer equipment and medium | |
| CN114238334A (en) | Heterogeneous data encoding method and device, heterogeneous data decoding method and device, computer equipment and storage medium | |
| CN110678854B (en) | Data query method and device | |
| US11663274B2 (en) | Reference-based document ranking system | |
| CN115577078B (en) | Engineering cost audit information retrieval method, system, equipment and storage medium | |
| CN110727672A (en) | Data mapping relation query method and device, electronic equipment and readable medium | |
| CN113807429B (en) | Enterprise classification method, enterprise classification device, computer equipment and storage medium | |
| CN113886702A (en) | Hybrid cloud search engine and search method based on dynamic cache mechanism | |
| CN112883232A (en) | Resource searching method, device and equipment | |
| CN116821152A (en) | Query information rewriting method, device, computer equipment and storage medium | |
| CN117827802A (en) | Hbase-based data processing system, method and device | |
| CN113849512A (en) | Operation blood relationship query method and device, electronic equipment and storage medium | |
| CN116303826A (en) | Redundancy identification method and device | |
| CN114461606A (en) | Data storage method and device, computer equipment and storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |