CN107545306A - A kind of big data analysis platform based on cloud computing - Google Patents
A kind of big data analysis platform based on cloud computing Download PDFInfo
- Publication number
- CN107545306A CN107545306A CN201710542792.9A CN201710542792A CN107545306A CN 107545306 A CN107545306 A CN 107545306A CN 201710542792 A CN201710542792 A CN 201710542792A CN 107545306 A CN107545306 A CN 107545306A
- Authority
- CN
- China
- Prior art keywords
- target resource
- big data
- data analysis
- analysis platform
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention discloses a kind of big data analysis platform based on cloud computing, the big data analysis platform includes：Source obtaining module and training module；Source obtaining module, sent for obtaining target resource from cloud computing environment, and by the target resource of acquisition to training module；Training module, using the convolutional neural networks algorithm of target detection Faster RCNN based on region, the target resource is trained, exported the result of training as analysis result.The big data analysis platform provided using the embodiment of the present invention, can quickly, efficiently, the data resource of system for cloud computing Environments amount is analyzed and processed in time, for equipment safety, stably, efficiently run orderly guarantee is provided.
Description
Technical field
The present invention relates to communication technical field, more particularly to a kind of big data analysis platform based on cloud computing.
Background technology
With the development of mobile Internet and intelligent terminal technology, mobile intelligent terminal caused data in application process
Resource also shows explosive growth.Need to propose higher requirement to the computing capability of service end, for example, mobile cloud computing
（MCC, Mobile Cloud Computing）, i.e. mobile intelligent terminal passes through amount of calculation and the intensive task of energy consumption mobile mutual
Networking is unloaded to cloud computing center, by cloud computing center load capacity calculation function, to meet the calculating service of mobile intelligent terminal
Demand.
But still it can not meet to cloud meter along with the explosive growth of network data resource, existing computational methods
The data resource efficient analysis processing of magnanimity in network environment is calculated, therefore, a more preferable big data analysis platform is needed badly, with fast
Speed, efficiently, the data resource of magnanimity in cloud computing environment is analyzed and processed in time.
The content of the invention
, can quickly, efficiently, in time the embodiment of the invention discloses a kind of big data analysis platform based on cloud computing
The data resource of system for cloud computing Environments amount is analyzed and processed, for equipment safety, stably, efficiently operation be provided with
The guarantee of sequence.
In order to achieve the above object, should the embodiments of the invention provide a kind of big data analysis platform based on cloud computing
Big data analysis platform, including：Source obtaining module and training module；
The source obtaining module, for obtaining target resource from cloud computing environment, and by the target resource of acquisition send to
The training module；
The training module, using the convolutional neural networks algorithm of target detection Faster-RCNN based on region, to the target
Resource is trained, and is exported the result of training as analysis result.
Optionally, the big data analysis platform, in addition to：Grader；
Grader, for the training result for the target resource obtained according to the training module, and features training
Storehouse, the target resource is classified.
Optionally, the big data analysis platform, in addition to：Memory module；
The memory module, the target resource obtained for storing the source obtaining module.
Optionally, the big data analysis platform, in addition to：Display module；
The display module, for pushing the analysis result, so as to obtain the analysis result that user obtains the target resource.
Optionally, it is described to utilize the convolutional neural networks algorithm of target detection Faster-RCNN based on region, to the mesh
Mark resource is trained, and is exported the result of training as analysis result, including：
Using parse network Parse-Net to the target resource carry out global characteristics extraction, using the global characteristics of extraction as
The analysis result output of the target resource.
Big data analysis platform provided in an embodiment of the present invention based on cloud computing includes：Source obtaining module and training mould
Block；Source obtaining module, sent for obtaining target resource from cloud computing environment, and by the target resource of acquisition to training mould
Block；Training module, using the convolutional neural networks algorithm of target detection Faster-RCNN based on region, to the target resource
It is trained, is exported the result of training as analysis result.The big data analysis platform provided using the embodiment of the present invention,
Can quickly, efficiently, the data resource of system for cloud computing Environments amount is analyzed and processed in time, be equipment safety,
Stable, the efficiently orderly guarantee of operation offer.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can be with
Other accompanying drawings are obtained according to these accompanying drawings.
Fig. 1 is the structural representation of the big data analysis platform provided in an embodiment of the present invention based on cloud computing.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of protection of the invention.
Below by specific embodiment, the present invention is described in detail.
Fig. 1 be the big data analysis platform provided in an embodiment of the present invention based on cloud computing structural representation, the big number
According to analysis platform, including：Source obtaining module 100 and training module 200；
The source obtaining module 100, sent for obtaining target resource from cloud computing environment, and by the target resource of acquisition
To the training module 200；
The training module 200, using the convolutional neural networks algorithm of target detection Faster-RCNN based on region, to described
Target resource is trained, and is exported the result of training as analysis result.
This area skill personnel are full convolutional neural networks respectively, it is understood that Faster-RCNN is made up of two parts
（Region Proposal Network, abbreviation RPN）With fast convolution neutral net algorithm of target detection Fast-RCNN.
So that target resource is image as an example, RPN input can be the picture of arbitrary size, and RPN enters to the image of input
Row feature extraction, feature extraction can use VGG Net to carry out convolution, pondization operation, obtain the characteristic pattern on input picture
Spectrum, is divided into multiple rectangular windows, the central point of each rectangular window according still further to the convolution kernel of default size by characteristic spectrum
As a datum mark, 9 various sizes of pre-selection frames are chosen around this datum mark, for each pre-selection frame, rear
Face connects, and meets a two grader softmax and then connects a bounding box again（bounding box）Recurrence device, return device
Export and can determine the region of corresponding pre-selection frame, this 9 various sizes of pre-selection frame structures for 4 coordinate points, this 4 coordinate points
Into target preselected area collection that may be present.
The image that Fast-RCNN inputs above-mentioned RPN, and the target preselected area that may be present obtained by RPN
Collection is handled the image to obtain a convolution characteristic pattern first, then concentrates each region for preselected area as input
ROI（Region Of Interest）Pond layer extracts the characteristic vector of a regular length from convolution characteristic pattern, each
Characteristic vector is admitted to full articulamentum, finally gives the described classification of target resource.The class for target resource that will be obtained
It is not defined as the analysis result of the target resource.
In order to prevent the undue protrusion of target resource local feature, the analysis result interference to target resource is in turn resulted in,
Using the convolutional neural networks algorithm of target detection Faster-RCNN based on region, the target resource is trained, will be instructed
Experienced result exports as analysis result, can include：
Using parse network Parse-Net to the target resource carry out global characteristics extraction, using the global characteristics of extraction as
The analysis result output of the target resource.
Extraction global characteristics are carried out to the characteristic spectrum that Faster-RCNN is exported using network ParseNet is parsed, specifically
Method is to extract new feature, i.e. global characteristics using max pooling operations（Global feature）, this global characteristics
Together be normalized with the features of Faster-RCNN convolutional layers and operate and cascade up, be spliced into one it is new comprising the overall situation
The union feature of information and local detail（Combined feature）, this new feature is sent into behind Faster-RCNN
ROI ponds layer makees further detection.Global characteristics extraction, Neng Gougeng are carried out to foregoing characteristic spectrum by using ParseNet
Add the classification for accurately obtaining target resource.Improve the degree of accuracy to big data analysis.
In addition, in order that user understands the analysis result for visually obtaining the target resource, in offer of the embodiment of the present invention
A kind of implementation in, above-mentioned big data analysis platform, it is characterised in that the big data analysis platform, in addition to：It is aobvious
Show module；
The display module, for pushing the analysis result, so as to obtain the analysis result that user obtains the target resource.
In another embodiment of the invention, in order to classify to target resource, above-mentioned big data analysis platform, also
Including：Grader；
Grader, for the training result for the target resource obtained according to the training module, and features training
Storehouse, the target resource is classified.
Understandable to be, grader can be, but not limited to be SVMs (Support Vector Machine, letter
Claim SVM) grader, as from the foregoing, after analyzing target resource, the class for the target resource can be obtained
Other information, therefore, grader can be according to the classification informations of the target resource, and binding characteristic trains feature existing in storehouse, right
Target resource is classified.
In order to ensure the loss situation of target resource, above-mentioned big data analysis platform, in addition to：Memory module；
The memory module, the target resource obtained for storing the source obtaining module.
As fully visible.Big data analysis platform provided in an embodiment of the present invention based on cloud computing includes：Resource acquisition mould
Block and training module；Source obtaining module, sent out for obtaining target resource from cloud computing environment, and by the target resource of acquisition
Deliver to training module；Training module, using the convolutional neural networks algorithm of target detection Faster-RCNN based on region, to institute
State target resource to be trained, exported the result of training as analysis result.The big number provided using the embodiment of the present invention
According to analysis platform, can quickly, efficiently, the data resource of system for cloud computing Environments amount is analyzed and processed in time,
For equipment safety, stably, efficiently operation orderly guarantee is provided.
It should be noted that each embodiment in this specification is described by the way of related, each embodiment it
Between identical similar part mutually referring to what each embodiment stressed is the difference with other embodiment.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the scope of the present invention.It is all
Any modification, equivalent substitution and improvements made within the spirit and principles in the present invention etc., are all contained in protection scope of the present invention
It is interior.
Claims (5)
- A kind of 1. big data analysis platform based on cloud computing, it is characterised in that the big data analysis platform, including：Resource Acquisition module and training module；The source obtaining module, for obtaining target resource from cloud computing environment, and by the target resource of acquisition send to The training module；The training module, using the convolutional neural networks algorithm of target detection Faster-RCNN based on region, to the target Resource is trained, and is exported the result of training as analysis result.
- 2. big data analysis platform according to claim 1, it is characterised in that the big data analysis platform, in addition to：Point Class device；Grader, for the training result for the target resource obtained according to the training module, and features training Storehouse, the target resource is classified.
- 3. big data analysis platform according to claim 1, it is characterised in that the big data analysis platform, in addition to： Memory module；The memory module, the target resource obtained for storing the source obtaining module.
- 4. big data analysis platform according to claim 1, it is characterised in that the big data analysis platform, in addition to： Display module；The display module, for pushing the analysis result, so as to obtain the analysis result that user obtains the target resource.
- 5. the big data analysis platform according to claim 1-4, it is characterised in that described to utilize the convolution god based on region Through network objectives detection algorithm Faster-RCNN, the target resource is trained, using the result of training as analysis result Output, including：Using parse network Parse-Net to the target resource carry out global characteristics extraction, using the global characteristics of extraction as The analysis result output of the target resource.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710542792.9A CN107545306A (en) | 2017-07-05 | 2017-07-05 | A kind of big data analysis platform based on cloud computing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710542792.9A CN107545306A (en) | 2017-07-05 | 2017-07-05 | A kind of big data analysis platform based on cloud computing |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107545306A true CN107545306A (en) | 2018-01-05 |
Family
ID=60971096
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710542792.9A Pending CN107545306A (en) | 2017-07-05 | 2017-07-05 | A kind of big data analysis platform based on cloud computing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107545306A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110220275A (en) * | 2019-05-06 | 2019-09-10 | 珠海格力电器股份有限公司 | Control the method and the apparatus of air conditioning of apparatus of air conditioning output temperature |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015134328A1 (en) * | 2014-03-07 | 2015-09-11 | Qualcomm Incorporated | Photo management |
| CN105138977A (en) * | 2015-08-18 | 2015-12-09 | 成都鼎智汇科技有限公司 | Face identification method under big data environment |
| CN106548192A (en) * | 2016-09-23 | 2017-03-29 | 北京市商汤科技开发有限公司 | Based on the image processing method of neutral net, device and electronic equipment |
| CN106844614A (en) * | 2017-01-18 | 2017-06-13 | 天津中科智能识别产业技术研究院有限公司 | A kind of floor plan functional area system for rapidly identifying |
| CN106910206A (en) * | 2017-01-17 | 2017-06-30 | 邢惠钧 | A kind of method for tracking target and device |
-
2017
- 2017-07-05 CN CN201710542792.9A patent/CN107545306A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015134328A1 (en) * | 2014-03-07 | 2015-09-11 | Qualcomm Incorporated | Photo management |
| CN105138977A (en) * | 2015-08-18 | 2015-12-09 | 成都鼎智汇科技有限公司 | Face identification method under big data environment |
| CN106548192A (en) * | 2016-09-23 | 2017-03-29 | 北京市商汤科技开发有限公司 | Based on the image processing method of neutral net, device and electronic equipment |
| CN106910206A (en) * | 2017-01-17 | 2017-06-30 | 邢惠钧 | A kind of method for tracking target and device |
| CN106844614A (en) * | 2017-01-18 | 2017-06-13 | 天津中科智能识别产业技术研究院有限公司 | A kind of floor plan functional area system for rapidly identifying |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110220275A (en) * | 2019-05-06 | 2019-09-10 | 珠海格力电器股份有限公司 | Control the method and the apparatus of air conditioning of apparatus of air conditioning output temperature |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hossain et al. | Improving consumer satisfaction in smart cities using edge computing and caching: A case study of date fruits classification | |
| CN113902897B (en) | Training of target detection model, target detection method, device, equipment and medium | |
| CN112801164A (en) | Training method, device and equipment of target detection model and storage medium | |
| Alidoost et al. | A CNN-based approach for automatic building detection and recognition of roof types using a single aerial image | |
| CN109446889A (en) | Object tracking method and device based on twin matching network | |
| CN113011282A (en) | Graph data processing method and device, electronic equipment and computer storage medium | |
| CN109934181A (en) | Text recognition method, device, equipment and computer-readable medium | |
| Kim et al. | Edge-network-assisted real-time object detection framework for autonomous driving | |
| CN109472199A (en) | A kind of method and device of image co-registration classification | |
| CN108427971A (en) | The method and system of tobacco leaf grading based on mobile terminal | |
| CN115620081A (en) | Training method of target detection model, target detection method and device | |
| CN115359308A (en) | Model training method, apparatus, device, storage medium, and program for identifying difficult cases | |
| CN107545306A (en) | A kind of big data analysis platform based on cloud computing | |
| CN113378768A (en) | Garbage can state identification method, device, equipment and storage medium | |
| CN108960246A (en) | A kind of binary conversion treatment device and method for image recognition | |
| CN107368923A (en) | Sight spot temperature Forecasting Methodology and device | |
| CN112819157A (en) | Neural network training method and device and intelligent driving control method and device | |
| CN104867129A (en) | Light field image segmentation method | |
| CN112560936A (en) | Model parallel training method, device, equipment, storage medium and program product | |
| CN115861755A (en) | Feature fusion method and device, electronic equipment and automatic driving vehicle | |
| CN110377776B (en) | Method and device for generating point cloud data | |
| CN111415168A (en) | Transaction warning method and device | |
| Xie et al. | Yolov4-mobilenetv2-dw-lcarm: A real-time ship detection network | |
| CN113362227B (en) | Image processing method, device, electronic equipment and storage medium | |
| CN113344121B (en) | Method for training a sign classification model and sign classification |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180105 |