CN110708567A - Distributed self-optimization video real-time analysis framework based on active learning - Google Patents
Distributed self-optimization video real-time analysis framework based on active learning Download PDFInfo
- Publication number
- CN110708567A CN110708567A CN201910303275.5A CN201910303275A CN110708567A CN 110708567 A CN110708567 A CN 110708567A CN 201910303275 A CN201910303275 A CN 201910303275A CN 110708567 A CN110708567 A CN 110708567A
- Authority
- CN
- China
- Prior art keywords
- video
- analysis
- distributed
- data
- optimization
- 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
- 238000005457 optimization Methods 0.000 title claims abstract description 16
- 238000010223 real-time analysis Methods 0.000 title claims abstract description 15
- 238000004458 analytical method Methods 0.000 claims abstract description 30
- 238000013136 deep learning model Methods 0.000 claims abstract description 13
- 238000012549 training Methods 0.000 claims abstract description 12
- 230000007246 mechanism Effects 0.000 claims abstract description 5
- 238000000034 method Methods 0.000 claims description 12
- 238000013135 deep learning Methods 0.000 claims description 10
- 238000003860 storage Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 4
- 230000004927 fusion Effects 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 claims description 3
- 238000010191 image analysis Methods 0.000 claims description 2
- 238000012937 correction Methods 0.000 abstract description 2
- 238000009826 distribution Methods 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/234—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs
- H04N21/23418—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/56—Provisioning of proxy services
- H04L67/568—Storing data temporarily at an intermediate stage, e.g. caching
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/231—Content storage operation, e.g. caching movies for short term storage, replicating data over plural servers, prioritizing data for deletion
- H04N21/23106—Content storage operation, e.g. caching movies for short term storage, replicating data over plural servers, prioritizing data for deletion involving caching operations
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/02—Protocols based on web technology, e.g. hypertext transfer protocol [HTTP]
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a distributed self-optimization video real-time analysis framework based on active learning, wherein a Storm framework performs parallel training on distributed stored video data, a deep learning model is fused to form the distributed parallel video analysis framework, and the optimal distribution of computing resources is realized under the Storm framework by matching with a multi-terminal parallel service mechanism. And performing iterative training on the original model after correction by using the data of the error analysis, and updating the deep learning model in the video analysis frame, so that self-optimization of the frame is realized, and the accuracy and the robustness of the model are improved.
Description
Technical Field
The invention relates to the field of massive video analysis, data distributed computation and storage and deep learning, in particular to a distributed self-optimization video real-time analysis framework based on active learning.
Background
The rapid development of the deep learning technology provides a new idea for solving a plurality of problems in production. The convolutional neural network is powerful in that the multilayer structure of the convolutional neural network can automatically learn features, and can learn features of multiple layers: the sensing domain of the shallower convolutional layer is smaller, and the characteristics of some local regions are learned; deeper convolutional layers have larger perceptual domains and can learn more abstract features. These abstract features are less sensitive to the size, position, orientation, etc. of the object, thereby contributing to an improvement in recognition performance. The method has strong adaptability to factors such as geometric transformation, deformation and illumination of the target, and effectively overcomes the recognition resistance caused by variable target appearance. The method can automatically extract and analyze the features according to the data input into the network, and has higher universality generalization capability. The closest techniques to the present invention are:
(1) and deep learning: deep learning provides a method for enabling a computer to automatically learn mode characteristics, and the characteristic learning is integrated into the process of establishing a model, so that incompleteness caused by artificial design characteristics is reduced. Some machine learning applications taking deep learning as a core reach recognition or classification performance exceeding that of the existing algorithm under the application scene meeting specific conditions. However, in an application scenario where a limited amount of data is provided, the deep learning algorithm cannot perform an unbiased estimation on the regularity of the data. To achieve good accuracy, large data supports are required.
(2) Storm calculation framework: storm is an open-source distributed real-time computing system, can simply and reliably process a large number of data streams, and has a plurality of use scenes: such as real-time analysis, online machine learning, continuous computing, distributed RPC, ETL, and the like. Storm supports horizontal expansion, has high fault tolerance, ensures that each message can be processed, has high processing speed, and can process millions of messages per second by each node in a small cluster. Storm deployment and operation and maintenance are convenient, and more importantly, any programming language can be used for developing application.
In the background of massive video data, the conventional deep learning video analysis is difficult to achieve the effect of real-time analysis and bear the load of massive video data. In order to realize real-time analysis of videos under the condition of parallel input of massive multi-source video data and actively optimize a video analysis model by utilizing the input videos, the invention provides a distributed self-optimization video real-time analysis framework based on active learning.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides a distributed self-optimization video real-time analysis framework based on active learning, which is combined with a Storm stream processing framework and a deep learning image analysis model, realizes real-time analysis of videos under the condition of parallel input of mass multi-source video data, and realizes self-optimization of the deep learning model by using verified analysis results.
The technical scheme of the invention is as follows:
the method comprises the following steps that (1) distributed training of image data is achieved by adopting a Storm framework based on a Hadoop distributed storage system;
performing model fusion on the obtained multiple deep learning models, deploying the models to a distributed parallel video analysis framework, and realizing optimal allocation of computing resources by matching with a multi-terminal parallel service mechanism;
step (3), aiming at multi-source data input, selecting an optimal computing node in a distributed parallel video analysis frame autonomously;
step (4), the result of the video analysis is transmitted to the front end for displaying, and the result and the video image file are cached to a temporary storage server;
step (5), correcting the data in the temporary storage server, particularly the data with analysis errors, and then performing iterative training;
and (6) updating the neural network model in a video analysis frame to realize one-time free iteration.
The invention has the beneficial effects that:
(1) the method utilizes the advantage of Storm streaming processing to realize the distributed parallel training of the deep learning model and improve the efficiency of the deep learning model training;
(2) the real-time video analysis under the condition of parallel input of mass multi-source video data is realized;
(3) aiming at the difficult problem of large updating calculation amount of the deep learning model, the self-optimization of the video analysis model is realized by correcting the error analysis result on the premise of the existing deep learning model.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic diagram of a distributed self-optimizing video real-time parsing framework based on active learning 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.
As shown in fig. 1, a distributed self-optimization video real-time analysis framework based on active learning combines the efficient stream computation performance and the deep learning target detection capability of the Storm framework, and realizes real-time video analysis under the condition of parallel input of massive multi-source video data.
The following describes in detail a specific process of a distributed self-optimization video real-time parsing framework based on active learning:
the method comprises the following steps that (1) distributed training of image data is achieved by adopting a Storm framework based on a Hadoop distributed storage system;
performing model fusion on the obtained multiple deep learning models, deploying the models to a distributed parallel video analysis framework, and realizing optimal allocation of computing resources by matching with a multi-terminal parallel service mechanism;
step (3), aiming at multi-source data input, selecting an optimal computing node in a distributed parallel video analysis frame autonomously;
step (4), the result of the video analysis is transmitted to the front end for displaying, and the result and the video image file are cached to a temporary storage server;
step (5), correcting the data in the temporary storage server, particularly the data with analysis errors, and then performing iterative training;
and (6) updating the neural network model in a video analysis frame to realize one-time free iteration.
According to the distributed self-optimization video real-time analysis framework based on active learning, a Storm framework conducts parallel training on distributed stored video data, deep learning models are fused to form the distributed parallel video analysis framework, and optimal distribution of computing resources is achieved under the Storm framework in cooperation with a multi-terminal parallel service mechanism. And performing iterative training on the original model after correction by using the data of the error analysis, and updating the deep learning model in the video analysis frame, so that self-optimization of the frame is realized, and the accuracy and the robustness of the model are improved.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (1)
1. A distributed self-optimization video real-time analysis frame based on active learning is combined with a Storm stream processing frame and a deep learning image analysis model, real-time analysis of videos is achieved under the condition that massive multi-source video data are input in parallel, and self-optimization of the deep learning model is achieved by means of verified analysis results, and the distributed self-optimization video real-time analysis frame comprises the following steps:
the method comprises the following steps that (1) distributed training of image data is achieved by adopting a Storm framework based on a Hadoop distributed storage system;
performing model fusion on the obtained multiple deep learning models, deploying the models to a distributed parallel video analysis framework, and realizing optimal allocation of computing resources by matching with a multi-terminal parallel service mechanism;
step (3), aiming at multi-source data input, selecting an optimal computing node in a distributed parallel video analysis frame autonomously;
step (4), the result of the video analysis is transmitted to the front end for displaying, and the result and the video image file are cached to a temporary storage server;
step (5), correcting the data in the temporary storage server, particularly the data with analysis errors, and then performing iterative training;
and (6) updating the neural network model in a video analysis frame to realize one-time free iteration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910303275.5A CN110708567A (en) | 2019-04-15 | 2019-04-15 | Distributed self-optimization video real-time analysis framework based on active learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910303275.5A CN110708567A (en) | 2019-04-15 | 2019-04-15 | Distributed self-optimization video real-time analysis framework based on active learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110708567A true CN110708567A (en) | 2020-01-17 |
Family
ID=69193105
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910303275.5A Pending CN110708567A (en) | 2019-04-15 | 2019-04-15 | Distributed self-optimization video real-time analysis framework based on active learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110708567A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104992147A (en) * | 2015-06-09 | 2015-10-21 | 中国石油大学(华东) | License plate identification method of deep learning based on fast and slow combination cloud calculation environment |
| CN105336017A (en) * | 2015-09-29 | 2016-02-17 | 爱培科科技开发（深圳）有限公司 | Driving record information processing method and system based on Storm technology |
| CN105654047A (en) * | 2015-12-21 | 2016-06-08 | 中国石油大学(华东) | Online video intelligent processing system based on deep learning in cloud environment |
| CN106874883A (en) * | 2017-02-27 | 2017-06-20 | 中国石油大学(华东) | A kind of real-time face detection method and system based on deep learning |
| CN107067365A (en) * | 2017-04-25 | 2017-08-18 | 中国石油大学(华东) | The embedded real-time video stream processing system of distribution and method based on deep learning |
| CN107871164A (en) * | 2017-11-17 | 2018-04-03 | 济南浪潮高新科技投资发展有限公司 | A kind of mist computing environment personalization deep learning method |
| CN108764456A (en) * | 2018-04-03 | 2018-11-06 | 北京环境特性研究所 | Airborne target identification model construction platform, airborne target recognition methods and equipment |
| CN108921359A (en) * | 2018-07-26 | 2018-11-30 | 安徽大学 | A kind of distribution gas density prediction technique and device |
-
2019
- 2019-04-15 CN CN201910303275.5A patent/CN110708567A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104992147A (en) * | 2015-06-09 | 2015-10-21 | 中国石油大学(华东) | License plate identification method of deep learning based on fast and slow combination cloud calculation environment |
| CN105336017A (en) * | 2015-09-29 | 2016-02-17 | 爱培科科技开发（深圳）有限公司 | Driving record information processing method and system based on Storm technology |
| CN105654047A (en) * | 2015-12-21 | 2016-06-08 | 中国石油大学(华东) | Online video intelligent processing system based on deep learning in cloud environment |
| CN106874883A (en) * | 2017-02-27 | 2017-06-20 | 中国石油大学(华东) | A kind of real-time face detection method and system based on deep learning |
| CN107067365A (en) * | 2017-04-25 | 2017-08-18 | 中国石油大学(华东) | The embedded real-time video stream processing system of distribution and method based on deep learning |
| CN107871164A (en) * | 2017-11-17 | 2018-04-03 | 济南浪潮高新科技投资发展有限公司 | A kind of mist computing environment personalization deep learning method |
| CN108764456A (en) * | 2018-04-03 | 2018-11-06 | 北京环境特性研究所 | Airborne target identification model construction platform, airborne target recognition methods and equipment |
| CN108921359A (en) * | 2018-07-26 | 2018-11-30 | 安徽大学 | A kind of distribution gas density prediction technique and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hu et al. | Learning semantic segmentation of large-scale point clouds with random sampling | |
| Chen et al. | An edge traffic flow detection scheme based on deep learning in an intelligent transportation system | |
| Oh et al. | Fast video object segmentation by reference-guided mask propagation | |
| Wang et al. | Enabling edge-cloud video analytics for robotics applications | |
| WO2021103135A1 (en) | Deep neural network-based traffic classification method and system, and electronic device | |
| Wang et al. | Real-time load reduction in multimedia big data for mobile Internet | |
| Zhang et al. | Casva: Configuration-adaptive streaming for live video analytics | |
| WO2022148248A1 (en) | Image processing model training method, image processing method and apparatus, electronic device, and computer program product | |
| CN105654047A (en) | Online video intelligent processing system based on deep learning in cloud environment | |
| Nigade et al. | Clownfish: Edge and cloud symbiosis for video stream analytics | |
| US20200349346A1 (en) | Facial recognition for multi-stream video using high probability group and facial network of related persons | |
| Chen et al. | Cuttlefish: Neural configuration adaptation for video analysis in live augmented reality | |
| US20200349344A1 (en) | Facial recognition for multi-stream video using high probability group | |
| Jayarajah et al. | Comai: Enabling lightweight, collaborative intelligence by retrofitting vision dnns | |
| Le et al. | Gradient alignment for cross-domain face anti-spoofing | |
| Wang et al. | Accelerating real‐time object detection in high‐resolution video surveillance | |
| Zhang | [Retracted] Sports Action Recognition Based on Particle Swarm Optimization Neural Networks | |
| Zhang et al. | Crossvision: Real-time on-camera video analysis via common roi load balancing | |
| CN110708567A (en) | Distributed self-optimization video real-time analysis framework based on active learning | |
| Zhu et al. | An interpretable multivariate time-series anomaly detection method in cyber-physical systems based on adaptive mask | |
| CN117746172A (en) | Heterogeneous model polymerization method and system based on domain difference perception distillation | |
| Guo et al. | Research on human-vehicle gesture interaction technology based on computer visionbility | |
| Meng et al. | Hierarchical feature aggregation network with semantic attention for counting large‐scale crowd | |
| Liu et al. | A real-time smoke and fire warning detection method based on an improved YOLOv5 model | |
| CN116170638B (en) | Self-attention video stream compression method and system for online action detection task |
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: 20200117 |