CN107480581A - Object identification method and device - Google Patents
Object identification method and device Download PDFInfo
- Publication number
- CN107480581A CN107480581A CN201710207520.3A CN201710207520A CN107480581A CN 107480581 A CN107480581 A CN 107480581A CN 201710207520 A CN201710207520 A CN 201710207520A CN 107480581 A CN107480581 A CN 107480581A
- Authority
- CN
- China
- Prior art keywords
- characteristic
- characteristic point
- point
- image
- identification
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
Abstract
The present invention discloses a kind of object identification method, and this method includes：The model of set of characteristic points is established according to pre-defined rule；By the image of acquisition, it is compared in the model, to identify object to be identified that described image includes.The present invention also provides a kind of object detector.According to object identification method provided by the invention and device, object identification can be automatically carried out, mitigates manpower.
Description
Technical field
The present invention relates to object identification field, in particular to a kind of object identification method and device.
Background technology
Application journey of the long-range assistance application primarily directed to remote services such as flight-line maintenance, after-sale service or remote guides
Sequence.Long-range assistance application is broadly divided into terminal and server two parts, and when terminal needs remote service, server need to be to terminal
The video of transmission carries out long-distance video processing, to be terminal remote service.
Existing remote service, the identification of the object included to image, mainly manually it is compared and services, make
It is low into efficiency of service, and take excessive manpower.
Thus, need badly and want a kind of object identification method and object detector that can be carried out automatically.
The content of the invention
A kind of object identification method and device are provided in the embodiment of the present invention, to realize automatic progress object identification.
To achieve the above object, the embodiment of the present invention provides a kind of object identification method, and this method includes：
The model of set of characteristic points is established according to pre-defined rule；
By the image of acquisition, it is compared in the model, to identify object to be identified that described image includes.
In one of the embodiments, it is described set of characteristic points is established according to pre-defined rule model the step of, specific bag
Include：
Obtain the image of multiple predetermined angulars of object to be identified；
For the image of each predetermined angular, multiple characteristic points are extracted in the object area to be identified；
The characteristic point is added to global unified index inverted list；
According to preset extraction rule, the set of the characteristic point near each characteristic point is established as close to point set.
In one of the embodiments, the characteristic point of the object to be identified includes：Color characteristic, textural characteristics, shape
One or more in feature or local feature region.
In one of the embodiments, point of the preset extraction rule for extraction apart from the characteristic point preset distance,
Or the predetermined number point of destination that extraction is nearest apart from the characteristic point.
In one of the embodiments, the image by acquisition, is compared in the model, described to identify
The step of object to be identified that image includes, specifically include：
The image of acquisition is carried out to the extraction of characteristic point；
The characteristic point for obtaining described image indexes corresponding index list in inverted list described；
All index entries of the index list are traveled through, find the most object of occurrence number as pre-identification object；
The set of characteristic points C that the pre-identification object is included is found out in the model；
All characteristic points of the set C are traveled through, for each characteristic point, are found out closest with it N number of with the collection
Other characteristic points in C are closed, and are compared with set of characteristic points corresponding with the pre-identification object is extracted in described image；
Judge in the set C, if more than being extracted in the set of characteristic points and described image near the characteristic point of predetermined number
Set of characteristic points corresponding with the pre-identification object in neighbor relationships be consistent；
If it is, confirm described image in object to be identified be pre-identification object, otherwise, it is unidentified go out object to be identified.
In one of the embodiments, the index entry is made a distinction by object classification, and the occurrence number is up to institute
It is most to state the number that the characteristic point of image occurs in the object classification.
In one of the embodiments, the half of all characteristic point quantity of the predetermined number more than or equal to set C.
According to another object of the present invention, a kind of object detector is also provided, the device includes：
Module is established, for establishing the model of set of characteristic points according to pre-defined rule；
Identification module, for by the image of acquisition, being compared in the model, to identify that what described image included treats
Identify object.
In one of the embodiments, the module of establishing includes：
First acquisition unit, the image of multiple predetermined angulars for obtaining object to be identified；
First extraction unit, for the image for each predetermined angular, extracted in the object area to be identified more
Individual characteristic point；
Adding device, for the characteristic point to be added into global unified index inverted list；
Unit is established close to point set, for according to preset extraction rule, by the collection of the characteristic point near each characteristic point
Conjunction is established as close to point set.
In one of the embodiments, the identification module includes：
Second extraction unit, for the image of acquisition to be carried out to the extraction of characteristic point；
Second acquisition unit, the characteristic point for obtaining described image index corresponding index list in inverted list described；
First Traversal Unit, for traveling through all index entries of the index list, find the most object conduct of occurrence number
Pre-identification object；
Unit is found out, the set of characteristic points C included for finding out the pre-identification object in the model；
Second Traversal Unit, for traveling through all characteristic points of the set C, for each characteristic point, find out with its distance most
Near N number of other characteristic points with the set C, and it is corresponding with the pre-identification object with being extracted in described image
Set of characteristic points compares；
Judging unit, for judging in the set C, if more than the set of characteristic points near the characteristic point of predetermined number and institute
The neighbor relationships stated in the set of characteristic points corresponding with the pre-identification object extracted in image are consistent；
Recognition unit, for the judged result according to the judging unit, if it exceeds the spy near the characteristic point of predetermined number
Sign point set meets neighbor relationships, then confirms that the object to be identified in described image is pre-identification object, otherwise, unidentified to go out to treat
Identify object.
According to another object of the present invention, then provide a kind of object identification equipment, it is characterised in that including any of the above-described
Described object detector.
In one of the embodiments, the object identification equipment is handheld device, or Intelligent Recognition terminal.
Existing object recognition technique is artificial due to needing more dependence, causes efficiency of service low, and takes excessive
Manpower.And according to object identification method provided by the invention and device, it can automatically carry out object identification, accuracy high stable
Property is good, and can reduce manpower, improves efficiency.
Brief description of the drawings
Fig. 1 is the flow and method figure of the object identification method of one embodiment of the invention；
Fig. 2 is the step S120 of embodiment illustrated in fig. 1 flow chart；
Fig. 3 is the step S140 of embodiment illustrated in fig. 1 flow chart；
Fig. 4 is the module map of the object detector of one embodiment of the invention.
Embodiment
The present invention is described in further detail with specific embodiment below in conjunction with the accompanying drawings, but not as the limit to the present invention
It is fixed.
Fig. 1 is the flow and method figure of the object identification method of one embodiment of the invention；
Fig. 2 is the step S120 of embodiment illustrated in fig. 1 flow chart；
Fig. 3 is the step S140 of embodiment illustrated in fig. 1 flow chart.
With reference to figure 1, the embodiment of the present invention provides a kind of object identification method, including：
Step 120, the model of set of characteristic points is established according to pre-defined rule；
Step 140, by the image of acquisition, it is compared in the model, to identify that described image includes to be identified
Object.
With reference to figure 2, above-mentioned steps S120 also comprises the following steps：
Step S122：Obtain the image of multiple predetermined angulars of object to be identified；
Step S124：For the image of each predetermined angular, multiple characteristic points are extracted in the object area to be identified；
Step S126：The characteristic point is added to global unified index inverted list；
Step S128：According to preset extraction rule, the set of the characteristic point near each characteristic point is established as close to point
Set.
Wherein, the characteristic point of above-mentioned object to be identified includes：Color characteristic, textural characteristics, shape facility or local special
One or more in sign point.Point of the preset extraction rule for extraction apart from the characteristic point preset distance, or extraction distance
The nearest predetermined number point of destination of the characteristic point.
Further, characteristics of image can include color characteristic, texture top grade, shape facility and local feature region etc..
Wherein local features have good stability, it is not easy to disturbed by external environment,
Image characteristics extraction is the premise of graphical analysis and image recognition, and it is to carry out the view data of higher-dimension to simplify expression most
Effective manner, from the data matrix of M × N × 3M × N × 3 of piece image, we do not see any information, so we
The key message in image, some primary elements and their relation must be extracted according to these data.
Local feature region is the local expression of characteristics of image, the local particularity that it can only have on image anyway, so
It is suitable only for matching image, the application such as retrieval.It is then unsuitable for image understanding.And the latter is complete more concerned with some
Office's feature, such as distribution of color, textural characteristics, shape of primary objects etc..Global characteristics are easily disturbed by environment, illumination,
The unfavorable factors such as rotation, noise can all influence global characteristics.Comparatively speaking, local feature region, it often correspond to one in image
A little lines intersect, and in the structure of light and shade change, the interference being subject to is also few.
And spot and angle point are two class local feature regions.Spot typically refers to have color and the other area of gray scale difference with surrounding
Domain, such as the one tree on grassland or a house.It is a region, so it is eager to excel than the ability of making an uproar of angle point, stability will
It is good.And angle point is then the cross section in image between the turning of one side object or lines.
With reference to figure 3, the above-mentioned image by acquisition, it is compared in the model, to identify that described image includes
Object to be identified the step of, specifically comprise the following steps：
Step S141：The image of acquisition is carried out to the extraction of characteristic point；
Step S142：The characteristic point for obtaining described image indexes corresponding index list in inverted list described；
Step S143：All index entries of the index list are traveled through, find the most object of occurrence number as pre-identification thing
Body；
Step S148：The set of characteristic points C that the pre-identification object is included is found out in the model；
Step S144：All characteristic points of the set C are traveled through, for each characteristic point, are found out closest with it N number of same
Other characteristic points in the set C, and with feature point set corresponding with the pre-identification object is extracted in described image
Composition and division in a proportion pair；
Step S145：Judge in the set C, if more than the set of characteristic points near the characteristic point of predetermined number and the figure
Neighbor relationships in the set of characteristic points corresponding with the pre-identification object extracted as in are consistent.
Wherein, the specific method for determining whether neighbor relationships is, by the preceding m characteristic point near some characteristic point according to
A sequence is lined up from the near to the remote, finds out the common subsequence of characteristic point sequence near two characteristic points, and it is public to calculate this
The length of subsequence, if length reaches predetermined length（For example, account for 80%）, it is believed that the neighbour of the two characteristic points is closed
System is consistent.It is actual, different predetermined lengths can be set to determine the Stringency of neighbor relationships.For example, to object identification requirement
When higher, predetermined length is improved；Opposite, the predetermined length numerical value can be reduced.Which also, can be counted, characteristic point
Neighbouring set of characteristic points meets neighbor relationships, and the set of characteristic points near which characteristic point does not conform to symbol neighbor relationships.So as to sentence
It is disconnected whether to exceed predetermined number.Optionally, the predetermined number is one of all characteristic point quantity more than or equal to set C
Half.
Step S146：If it is, confirm that the object to be identified in described image is pre-identification object；That is, work as
When meeting neighbouring relations more than the set of characteristic points near the characteristic point of predetermined number, then object is identified.
Step S147：Otherwise, it is unidentified go out object to be identified.That is, as the spy less than or equal to predetermined number
A set of characteristic points near sign point is when meeting neighbouring relations, then it is unidentified go out object to be identified.
Wherein, the index entry is made a distinction by object classification, and the occurrence number is up to the characteristic point of described image
The number occurred in the object classification is most.
Wherein, index entry can be the other list of object type.Index list can be using key point as index.Index falls to arrange
Table is with other order arrangement from key point to object type.The number that each object classification occurs is unique.Work as characteristic point
The number occurred in some index entry is most, just using category object as pre-identification object.
Fig. 4 is the module map of the object detector of one embodiment of the invention.
With reference to figure 4, a kind of object detector 200, including：Establish module 210 and identification module 230.Wherein, mould is established
Block 210 establishes the model of set of characteristic points according to pre-defined rule, and identification module 230 is used for the image of acquisition, in the model
In be compared, to identify object to be identified that described image includes.
With reference to figure 4, establishing module 210 includes：First acquisition unit 211, the first extraction unit 212, the He of adding device 213
Unit 214 is established close to point set.
Wherein, first acquisition unit 211 obtains the image of multiple predetermined angulars of object to be identified；First extraction unit
212 are directed to the image of each predetermined angular, and multiple characteristic points are extracted in the object area to be identified；Adding device
The characteristic point is added to global unified index inverted list by 213；Unit 214 is established close to point set to be advised according to predetermined extraction
Then, the set of the characteristic point near each characteristic point is established as close to point set.
Wherein, identification module 230 includes：Second extraction unit 231, second acquisition unit 232, the first Traversal Unit 233,
Find out unit 238, the second Traversal Unit 234, judging unit 235 and recognition unit 236.
Wherein, the second extraction unit 231 carries out the image of acquisition the extraction of characteristic point, and second acquisition unit 232 obtains
The characteristic point of described image corresponding index list in the index inverted list；First Traversal Unit 233 travels through the index
All index entries of list, the most object of occurrence number is found as pre-identification object；Unit 238 is found out, for from described
The set of characteristic points C that the pre-identification object is included is found out in model；Second Traversal Unit 234 travels through the institute of the set C
There is characteristic point, for each characteristic point, find out N number of other characteristic points with the set C closest with it, and and
The set of characteristic points corresponding with the pre-identification object extracted in described image compares；Judging unit 235 judges the set C
In, if more than being extracted in the set of characteristic points and described image near the characteristic point of predetermined number with the pre-identification object
Neighbor relationships in corresponding set of characteristic points are consistent；Recognition unit 236 according to the judged result of the judging unit, if
Meet neighbor relationships more than the set of characteristic points near the characteristic point of predetermined number, then confirm the object to be identified in described image
For pre-identification object, otherwise, it is unidentified go out object to be identified.
Wherein, the device 200 realizes object identification according to the above method.
The present invention also provides a kind of object identification equipment, and it includes any of the above-described kind of object detector 200.Object identification fills
It can be handheld device to put, or Intelligent Recognition terminal.
By object identification method provided by the invention, device and equipment, in various handheld devices or end can only be identified
Automatically the identification to image, and the identification to included object in image are realized on end.Its object that can be carried out automatically is known
Not, accuracy high stability is good, and can reduce manpower, improves efficiency.After target object identifies, after this is just
The continuous corresponding service of offer provides important premise, such as, according to the object identified, corresponding customer service is dialed out, or
Person transfers corresponding operation instruction and pushed, or transfers out corresponding solution.So it is achieved that remote self-help takes
Business.
Certainly, above is the preferred embodiment of the present invention.It should be pointed out that for those skilled in the art
For, on the premise of its general principles are not departed from, some improvements and modifications can also be made, these improvements and modifications
It is considered as protection scope of the present invention.
Claims (10)
- A kind of 1. object identification method, it is characterised in that including：The model of set of characteristic points is established according to pre-defined rule；By the image of acquisition, it is compared in the model, to identify object to be identified that described image includes.
- 2. object identification method according to claim 1, it is characterised in that described that feature point set is established according to pre-defined rule The step of model of conjunction, specifically include：Obtain the image of multiple predetermined angulars of object to be identified；For the image of each predetermined angular, multiple characteristic points are extracted in the object area to be identified；The characteristic point is added to global unified index inverted list；According to preset extraction rule, the set of the characteristic point near each characteristic point is established as close to point set.
- 3. object identification method according to claim 2, it is characterised in that the characteristic point of the object to be identified includes： One or more in color characteristic, textural characteristics, shape facility or local feature region.
- 4. the object identification method according to Claims 2 or 3, it is characterised in that the preset extraction rule for extraction away from From the point of the characteristic point preset distance, or the predetermined number point of destination that extraction is nearest apart from the characteristic point.
- 5. object identification method according to claim 1, it is characterised in that the image by acquisition, in the model In be compared, the step of to identify object to be identified that described image includes, specifically include：The image of acquisition is carried out to the extraction of characteristic point；The characteristic point for obtaining described image indexes corresponding index list in inverted list described；All index entries of the index list are traveled through, find the most object of occurrence number as pre-identification object；The set of characteristic points C that the pre-identification object is included is found out in the model；All characteristic points of the set C are traveled through, for each characteristic point, are found out closest with it N number of with the collection Other characteristic points in C are closed, and are compared with set of characteristic points corresponding with the pre-identification object is extracted in described image；Judge in the set C, if more than being extracted in the set of characteristic points and described image near the characteristic point of predetermined number Set of characteristic points corresponding with the pre-identification object in neighbor relationships be consistent；If it is, confirm described image in object to be identified be pre-identification object, otherwise, it is unidentified go out object to be identified.
- 6. object identification method according to claim 5, it is characterised in that the index entry carries out area by object classification Point, the number that the characteristic point that the occurrence number is up to described image occurs in the object classification is most.
- 7. object identification method according to claim 5, it is characterised in that the predetermined number is more than or equal to set C All characteristic point quantity half.
- A kind of 8. object detector, it is characterised in that including：Module is established, for establishing the model of set of characteristic points according to pre-defined rule；Identification module, for by the image of acquisition, being compared in the model, to identify that what described image included treats Identify object.
- 9. object detector according to claim 8, it is characterised in that the module of establishing includes：First acquisition unit, the image of multiple predetermined angulars for obtaining object to be identified；First extraction unit, for the image for each predetermined angular, extracted in the object area to be identified more Individual characteristic point；Adding device, for the characteristic point to be added into global unified index inverted list；Unit is established close to point set, for according to preset extraction rule, by the collection of the characteristic point near each characteristic point Conjunction is established as close to point set.
- 10. object detector according to claim 8, it is characterised in that the identification module includes：Second extraction unit, for the image of acquisition to be carried out to the extraction of characteristic point；Second acquisition unit, the characteristic point for obtaining described image index corresponding index list in inverted list described；First Traversal Unit, for traveling through all index entries of the index list, find the most object conduct of occurrence number Pre-identification object；Unit is found out, the set of characteristic points C included for finding out the pre-identification object in the model；Second Traversal Unit, for traveling through all characteristic points of the set C, for each characteristic point, find out with its distance most Near N number of other characteristic points with the set C, and it is corresponding with the pre-identification object with being extracted in described image Set of characteristic points compares；Judging unit, for judging in the set C, if more than the set of characteristic points near the characteristic point of predetermined number and institute The neighbor relationships stated in the set of characteristic points corresponding with the pre-identification object extracted in image are consistent；Recognition unit, for the judged result according to the judging unit, if it exceeds the spy near the characteristic point of predetermined number Sign point set meets neighbor relationships, then confirms that the object to be identified in described image is pre-identification object, otherwise, unidentified to go out to treat Identify object.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710207520.3A CN107480581B (en) | 2017-03-31 | 2017-03-31 | Object recognition method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710207520.3A CN107480581B (en) | 2017-03-31 | 2017-03-31 | Object recognition method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107480581A true CN107480581A (en) | 2017-12-15 |
| CN107480581B CN107480581B (en) | 2021-06-15 |
Family
ID=60594008
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710207520.3A Active CN107480581B (en) | 2017-03-31 | 2017-03-31 | Object recognition method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107480581B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060066616A1 (en) * | 2004-09-30 | 2006-03-30 | Intel Corporation | Diffuse photon map decomposition for parallelization of global illumination algorithm |
| CN101859326A (en) * | 2010-06-09 | 2010-10-13 | 南京大学 | Image searching method |
| CN103793466A (en) * | 2013-12-20 | 2014-05-14 | 深圳先进技术研究院 | Image retrieval method and image retrieval device |
| CN103996207A (en) * | 2014-04-28 | 2014-08-20 | 清华大学深圳研究生院 | Object tracking method |
| CN103999097A (en) * | 2011-07-11 | 2014-08-20 | 华为技术有限公司 | System and method for compact descriptor for visual search |
| CN104156362A (en) * | 2013-05-14 | 2014-11-19 | 视辰信息科技(上海)有限公司 | Large-scale image feature point matching method |
-
2017
- 2017-03-31 CN CN201710207520.3A patent/CN107480581B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060066616A1 (en) * | 2004-09-30 | 2006-03-30 | Intel Corporation | Diffuse photon map decomposition for parallelization of global illumination algorithm |
| CN101859326A (en) * | 2010-06-09 | 2010-10-13 | 南京大学 | Image searching method |
| CN103999097A (en) * | 2011-07-11 | 2014-08-20 | 华为技术有限公司 | System and method for compact descriptor for visual search |
| CN104156362A (en) * | 2013-05-14 | 2014-11-19 | 视辰信息科技(上海)有限公司 | Large-scale image feature point matching method |
| CN103793466A (en) * | 2013-12-20 | 2014-05-14 | 深圳先进技术研究院 | Image retrieval method and image retrieval device |
| CN103996207A (en) * | 2014-04-28 | 2014-08-20 | 清华大学深圳研究生院 | Object tracking method |
Non-Patent Citations (2)
| Title |
|---|
| SUIHUA WANG，AND ETC: "An Image Indexing and Retrieval Model Using Reasoning Services", 《2009 INTERNATIONAL CONFERENCE ON MULTIMEDIA INFORMATION NETWORKING AND SECURITY》 * |
| 熊云艳等: "有序的KD-tree在图像特征匹配上的应用", 《化工自动化及仪表》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107480581B (en) | 2021-06-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhao et al. | Rest: A reference-based framework for spatio-temporal trajectory compression | |
| CN110020433B (en) | Industrial and commercial high-management name disambiguation method based on enterprise incidence relation | |
| ATE441989T1 (en) | MOBILITY MANAGEMENT IN WIRELESS NETWORKS | |
| CN103984764B (en) | Perceive the personalized method for secret protection of query semantics based on road network | |
| CN103136627A (en) | Area real estate intelligent management system | |
| CN102789508A (en) | Distributed practical condition search engine and chat system on basis of geographical position | |
| CN108712712A (en) | Wireless Fidelity WiFi network related information display methods and device | |
| CN109710792A (en) | A kind of fast face searching system application based on index | |
| SE0201315D0 (en) | A method and system of rating in a charging system | |
| CN103207901A (en) | Method and system for obtaining internet protocol address attribution position based on search engine | |
| CN111488409A (en) | City address library construction method, retrieval method and device | |
| CN203165005U (en) | Area property intelligent management system | |
| CN110085299B (en) | Image identification dryness removal method and system and image library | |
| CN106383917A (en) | Data processing method based on user logs | |
| CN101998226A (en) | Mobile searching method and system | |
| CN107480581A (en) | Object identification method and device | |
| CN107784421A (en) | Method for allocating tasks and device | |
| WO2018232887A1 (en) | Mobile person information acquisition feedback method, server, and system | |
| CN110851547A (en) | Multi-data-source map data fusion method | |
| CN110659277A (en) | Energy internet data acquisition method and device | |
| CN107506473B (en) | Big data retrieval method based on cloud computing | |
| CN109842482A (en) | A kind of information synchronization method, system and terminal device | |
| CN112087730B (en) | Modularized cable trench system based on 5G and Beidou positioning | |
| CN112199347B (en) | Power supply service knowledge sharing platform system | |
| CN105630978A (en) | Method and device for collecting information |
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 |