CN110365942B - Real-time video intelligent analysis method and system - Google Patents
Real-time video intelligent analysis method and system Download PDFInfo
- Publication number
- CN110365942B CN110365942B CN201910650758.2A CN201910650758A CN110365942B CN 110365942 B CN110365942 B CN 110365942B CN 201910650758 A CN201910650758 A CN 201910650758A CN 110365942 B CN110365942 B CN 110365942B
- Authority
- CN
- China
- Prior art keywords
- analysis
- video
- micro
- video data
- service
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a real-time video intelligent analysis method and a system, wherein the method comprises the following steps: acquiring video data acquired by a camera; caching the video data into a micro service instance for analysis to obtain a video analysis result; the video intelligent analysis system comprises at least one analysis node, wherein at least one micro-service instance is deployed on one analysis node, and one micro-service instance is provided with at least one path of access and analysis module of a camera. According to the embodiment of the invention, through the micro-service examples on the analysis nodes in the distributed design, the bandwidth pressure and performance influence caused by multi-path video access is avoided, the video data analysis of more paths of cameras can be supported, and the video analysis efficiency is improved. The video data collected by the same camera forms a plurality of buffer queues for analysis, and the data analysis performance can be improved.
Description
Technical Field
The invention relates to the technical field of video analysis, in particular to a real-time video intelligent analysis method and system.
Background
Video monitoring is an important component of a safety precaution system, and the monitoring system comprises a front-end camera, a transmission cable and a video monitoring platform. Video monitoring is widely applied to many occasions due to intuition, accuracy, timeliness and rich information content. In recent years, with the rapid development of computers, networks and image processing and transmission technologies, video monitoring technologies have been developed greatly, and computer vision and deep learning technologies are applied to assist (replace) manual video monitoring analysis, so that 7 × 24 waiting is performed, abnormal conditions can be found in real time and warning and early warning information can be pushed, and the efficiency of safety monitoring is improved. However, the video data volume is large, deep learning and machine vision analysis need to occupy a large amount of resources, large-scale video analysis meets challenges, most of traditional video analysis products are relatively fixed analysis scenes, single analysis models, poor video data processing performance and low analysis efficiency.
Disclosure of Invention
The invention provides a real-time video intelligent analysis method and a real-time video intelligent analysis system, which solve the problems of poor processing performance and low analysis efficiency of the traditional video analysis algorithm.
The embodiment of the invention provides a real-time video intelligent analysis method, which is applied to a video intelligent analysis system and comprises the following steps:
acquiring video data acquired by a camera;
caching the video data into a micro service instance for analysis to obtain a video analysis result; the video intelligent analysis system comprises at least one analysis node, wherein at least one micro-service instance is deployed on one analysis node, and one micro-service instance is provided with at least one path of access and analysis module of a camera.
The embodiment of the invention also provides a real-time video intelligent analysis system, which comprises: at least one analysis node, at least one micro service instance deployed on one analysis node, and at least one access and analysis module of one camera configured on one micro service instance, wherein the system further comprises:
the acquisition module is used for acquiring video data acquired by the camera;
and the analysis node is used for caching the video data into the micro-service instance for analysis to obtain a video analysis result.
The technical scheme of the invention has the beneficial effects that: through the micro-service examples on the analysis nodes in the distributed design, the bandwidth pressure and performance influence caused by multi-path video access is avoided, the video data analysis of more paths of cameras can be supported, and the video analysis efficiency is improved. The video data collected by the same camera forms a plurality of buffer queues for analysis, and the data analysis performance can be improved.
Drawings
FIG. 1 is a schematic flow chart of a real-time video intelligent analysis method according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating a buffer queue design according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a real-time video intelligent analysis system according to an embodiment of the present invention;
FIG. 4 is a block diagram of an embodiment of a microservice instance;
FIG. 5 is a block diagram of a model detector according to an embodiment of the present invention.
Detailed Description
In order to make the technical problems, technical solutions and advantages of the present invention more apparent, the following detailed description is given with reference to the accompanying drawings and specific embodiments. In the following description, specific details such as specific configurations and components are provided only to help the full understanding of the embodiments of the present invention. Thus, it will be apparent to those skilled in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the invention. In addition, descriptions of well-known functions and constructions are omitted for clarity and conciseness.
It should be appreciated that reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, the appearances of the phrases "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
In various embodiments of the present invention, it should be understood that the sequence numbers of the following processes do not mean the execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
In addition, the terms "system" and "network" are often used interchangeably herein.
In the embodiments provided herein, it should be understood that "B corresponding to a" means that B is associated with a from which B can be determined. It should also be understood that determining B from a does not mean determining B from a alone, but may be determined from a and/or other information.
First embodiment
As shown in fig. 1, an embodiment of the present invention provides a real-time video intelligent analysis method, which specifically includes the following steps:
step 11: and acquiring video data acquired by the camera.
In the present embodiment, only the video data analysis mode of one of the cameras is described, and the video data of the other cameras are analyzed in the same manner.
Step 12: caching the video data into a micro service instance for analysis to obtain a video analysis result; the video intelligent analysis system comprises at least one analysis node, wherein at least one micro-service instance is deployed on one analysis node, and one micro-service instance is provided with at least one path of access and analysis module of a camera.
The video data of one camera is transmitted to one micro service instance (Engine) for analysis through the access and analysis module, a plurality of micro service instances are deployed on one analysis node, and the analysis node can analyze the video data of multiple cameras.
Optionally, before step 12, further comprising: and configuring analysis nodes and micro-service examples for the cameras in different paths. For example, video 1 is distributed to microservice instance 1 on analysis node 1, and video 2 is distributed to microservice instance 2 on analysis node 1.
In order to prevent video image abnormality, N frames of images are cached in each detection period, and specifically, the step of forming at least two buffer queues for video data collected by the same path of camera includes: in video data collected by the same path of camera, extracting N frames of images for caching according to the detection periods of at least two model detectors respectively to form at least two corresponding buffer queues; one model detector corresponds to one buffer queue, the detection periods of different model detectors are different, and N is a positive integer. The video data collected by the cameras in different paths can be respectively extracted from the N frames of images for caching according to the detection period of the same model detector, so that frequent memory allocation and memory release can be avoided by adopting a memory space multiplexing mode.
As shown in fig. 2, a video 1 with a refresh frame rate (or called acquisition frame rate, output frame rate, video frame number acquired within 1 second) of 20 is accessed, the video 1 stores 3 pictures within 1 second according to model 1 buffer cycle, the detection frequency of the model 1 detector is 1 time per second, the buffered 3 pictures are read in the detection period, and the optimal detection is selected, and then the next detection period performs detection analysis according to the same detection mode. The video 1 stores 3 pictures in 2 seconds according to the model 2 buffer in a circulating mode, the detection frequency of the model 2 detector is 1 time every 2 seconds, the 3 buffered pictures are read in the detection period, the optimal pictures are selected for detection, and then the next detection period carries out detection analysis according to the same detection mode. Accessing a video 2 with a refresh frame rate (or called as an acquisition frame rate, an output frame rate, the number of video frames acquired within 1 second) of 25, circularly storing 3 pictures within 2 seconds according to model 2 cache by the video 2, reading the cached 3 pictures in the detection period with the detection frequency of a model 2 detector being 1 time every 2 seconds, selecting the optimal detection, and then carrying out detection analysis in the next detection period according to the same detection mode. The video 2 stores 3 pictures in 10 seconds according to the model 3 buffer memory cycle, the detection frequency of the model 3 detector is 1 time per 10 seconds, the 3 buffered pictures are read in the detection period, the optimal detection is selected, and then the next detection period carries out detection analysis according to the same detection mode.
Further, after step 12, the method further comprises: and when the video analysis result indicates that the at least two paths of cameras have abnormality, pushing alarm information. The video data of one camera can be cached to at least two model detectors for detection and analysis, and when the detection and analysis results of all the model detectors are abnormal, the alarm information is pushed so as to improve the alarm accuracy. Of course, in a scene with a high security requirement, when only one model detector corresponding to one camera is abnormal, the warning information is pushed.
According to the embodiment of the invention, through the micro-service examples on the analysis nodes in the distributed design, the bandwidth pressure and performance influence caused by multi-path video access is avoided, the video data analysis of more paths of cameras can be supported, and the video analysis efficiency is improved. The video data collected by the same camera forms a plurality of buffer queues for analysis, and the data analysis performance can be improved.
Second embodiment
The first embodiment is described with respect to the real-time video intelligent analysis method of the present invention, and the following embodiment will further describe a real-time video intelligent analysis system corresponding thereto with reference to the accompanying drawings.
Specifically, as shown in fig. 3, the real-time video intelligent analysis system according to the embodiment of the present invention includes: the method comprises the following steps that at least one analysis node (for example, node 1, node 2 and node 3 in fig. 3) is deployed, at least one micro-service instance (for example, instance 1, instance 2 and instance 3 in fig. 3) is deployed on one analysis node, one micro-service instance configures an access and analysis module of at least one camera, specifically, the micro-service instance is a core service for implementing video access, processing and analysis, and multiple instances can be deployed on one or more nodes, and generally one GPU card is deployed with one instance. Wherein, the system still includes:
an obtaining module, configured to obtain video data collected by a camera, such as a transmission line between the camera and an analysis node in fig. 3;
and the analysis node is used for caching the video data into the micro-service instance for analysis to obtain a video analysis result.
Wherein, the micro-service example comprises:
the cache thread module is used for forming at least two cache queues for the video data collected by the same path of camera;
and the analysis thread module is used for inputting the at least two buffer queues into the micro service instance for analysis.
Specifically, as shown in fig. 4, a micro-service instance includes an access thread, a cache thread (forming a video cache queue), an analysis thread, and a management interface. The video 1 enters the cache thread through the access thread to form two video cache queues which are respectively input into the model detector 1 and the model detector 2 for detection and analysis. The video 2 enters the cache thread through the access thread to form a video cache queue and is respectively input into the model detector 2 for detection and analysis. The video 3 enters the cache thread through the access thread to form two video cache queues and is respectively input into the model detector 2 and the model detector 4 for detection and analysis. The video 4 enters the cache thread through the access thread to form two video cache queues which are respectively input into the model detector 3 and the model detector 4 for detection and analysis.
Wherein, the thread module of buffering includes:
the access thread and video cache thread module is used for extracting N frames of images for caching in video data collected by the same path of camera according to the detection periods of at least two model detectors respectively to form at least two corresponding buffer queues; wherein, a model detector corresponds to a buffer queue, and the detection period of different model detectors is different.
As shown in fig. 5, the model detector includes: a Unified Interface (Unified Interface) and dynamic libraries supporting C + +, Java, and other language developments. Wherein, the dynamic library comprises: the system comprises a model file, a logic code and an analysis result, wherein the analysis result can output structured data and abnormal alarm data. The detection model library of the model detector is in the form of an extension library so as to be convenient for upgrading and extending and support visual analysis algorithms written by various languages such as C + +, Java, Python and the like.
Wherein, real-time video intelligence analytic system still includes:
and the alarm service module is used for pushing alarm information when the video analysis result indicates that the at least two paths of cameras have abnormality.
Wherein, real-time video intelligence analytic system still includes:
and the management service module is used for configuring analysis nodes and micro-service examples for different cameras.
The system embodiment of the invention is corresponding to the embodiment of the method, all the implementation means in the method embodiment are applicable to the embodiment of the system, and the same technical effect can be achieved. The system avoids bandwidth pressure and performance influence caused by multi-path video access through the micro-service examples on the analysis nodes in a distributed design, can support video data analysis of more paths of cameras, and improves video analysis efficiency. The video data collected by the same camera forms a plurality of buffer queues for analysis, and the data analysis performance can be improved.
Those skilled in the art will appreciate that all or part of the steps for implementing the above embodiments may be performed by hardware, or may be instructed to be performed by associated hardware by a computer program that includes instructions for performing some or all of the steps of the above methods; and the computer program may be stored in a readable storage medium, which may be any form of storage medium.
Furthermore, it is to be noted that in the device and method of the invention, it is obvious that the individual components or steps can be decomposed and/or recombined. These decompositions and/or recombinations are to be regarded as equivalents of the present invention. Also, the steps of performing the series of processes described above may naturally be performed chronologically in the order described, but need not necessarily be performed chronologically, and some steps may be performed in parallel or independently of each other. It will be understood by those skilled in the art that all or any of the steps or elements of the method and apparatus of the present invention may be implemented in any computing device (including processors, storage media, etc.) or network of computing devices, in hardware, firmware, software, or any combination thereof, which can be implemented by those skilled in the art using their basic programming skills after reading the description of the present invention.
Thus, the objects of the invention may also be achieved by running a program or a set of programs on any computing device. The computing device may be a general purpose device as is well known. The object of the invention is thus also achieved solely by providing a program product comprising program code for implementing the method or the apparatus. That is, such a program product also constitutes the present invention, and a storage medium storing such a program product also constitutes the present invention. It is to be understood that the storage medium may be any known storage medium or any storage medium developed in the future. It is further noted that in the apparatus and method of the present invention, it is apparent that each component or step can be decomposed and/or recombined. These decompositions and/or recombinations are to be regarded as equivalents of the present invention. Also, the steps of executing the series of processes described above may naturally be executed chronologically in the order described, but need not necessarily be executed chronologically. Some steps may be performed in parallel or independently of each other.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (6)
1. A real-time video intelligent analysis method is applied to a video intelligent analysis system and is characterized by comprising the following steps:
acquiring video data acquired by a camera;
caching the video data into a micro service instance for analysis to obtain a video analysis result; the video intelligent analysis system comprises at least one analysis node, wherein at least one micro-service instance is deployed on one analysis node, one micro-service instance is configured with at least one path of access and analysis module of a camera, and the micro-service instance is a core service for realizing video access, processing and analysis; the micro-service example comprises the following steps: the method comprises the following steps of accessing a thread, a video cache queue and an analysis thread, wherein the analysis thread comprises at least one model detector;
the step of caching the video data into a micro-service instance for analysis comprises:
video data collected by a camera in the same path enters a cache thread through an access thread to form at least two cache queues, and at least two different parallel cache queues are respectively input to at least two model detectors for detection and analysis; one model detector corresponds to one cache queue, and the detection periods of different model detectors are different; the video data collected by the cameras in different paths can be respectively extracted and cached according to the detection period of the same model detector, wherein N is a positive integer.
2. The method for real-time intelligent analysis of video according to claim 1, wherein said step of obtaining video analysis results is followed by the steps of:
and when the video analysis result indicates that the at least two paths of cameras have abnormality, pushing alarm information.
3. The real-time intelligent video analysis method according to claim 1, wherein before the step of caching the video data in a microservice instance for analysis to obtain a video analysis result, the method further comprises:
and configuring analysis nodes and micro-service examples for different cameras.
4. A real-time video intelligent analysis system, comprising: the system comprises at least one analysis node, at least one micro-service instance is deployed on one analysis node, one micro-service instance is configured with an access and analysis module of at least one camera, and the micro-service instance is a core service for realizing video access, processing and analysis; the micro-service example comprises the following steps: the method comprises the following steps of accessing a thread, a video cache queue and an analysis thread, wherein the analysis thread comprises at least one model detector; wherein the system further comprises:
the acquisition module is used for acquiring video data acquired by the camera;
the analysis node is used for caching the video data into a micro service instance for analysis to obtain a video analysis result;
caching the video data into a micro-service instance for analysis, wherein the caching comprises the following steps:
video data collected by a camera in the same path enters a cache thread through an access thread to form at least two cache queues, and at least two different parallel cache queues are respectively input to at least two model detectors for detection and analysis; one model detector corresponds to one cache queue, and the detection periods of different model detectors are different; the video data collected by the cameras in different paths can be respectively extracted and cached according to the detection period of the same model detector, wherein N is a positive integer.
5. The real-time video intelligent analysis system of claim 4, further comprising:
and the alarm service module is used for pushing alarm information when the video analysis result indicates that the at least two paths of cameras have abnormality.
6. The real-time video intelligent analysis system of claim 5, further comprising:
and the management service module is used for configuring analysis nodes and micro-service examples for different cameras.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910650758.2A CN110365942B (en) | 2019-07-18 | 2019-07-18 | Real-time video intelligent analysis method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910650758.2A CN110365942B (en) | 2019-07-18 | 2019-07-18 | Real-time video intelligent analysis method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110365942A CN110365942A (en) | 2019-10-22 |
| CN110365942B true CN110365942B (en) | 2021-08-20 |
Family
ID=68221063
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910650758.2A Active CN110365942B (en) | 2019-07-18 | 2019-07-18 | Real-time video intelligent analysis method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110365942B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111010534A (en) * | 2019-11-11 | 2020-04-14 | 中国建设银行股份有限公司 | Real-time asynchronous video analysis method and system |
| CN111262838A (en) * | 2020-01-09 | 2020-06-09 | 南方电网科学研究院有限责任公司 | Intelligent analysis method, system and equipment for network security |
| CN115393781A (en) * | 2021-05-08 | 2022-11-25 | 华为技术有限公司 | Video monitoring data processing method and device |
| CN113572997A (en) * | 2021-07-22 | 2021-10-29 | 中科曙光国际信息产业有限公司 | Video stream data analysis method, device, equipment and storage medium |
| CN114466151B (en) * | 2022-04-11 | 2022-07-12 | 武汉中科通达高新技术股份有限公司 | Video storage system, computer equipment and storage medium of national standard camera |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8547878B2 (en) * | 2010-05-18 | 2013-10-01 | Lsi Corporation | Modularized scheduling engine for traffic management in a network processor |
| CN102638456B (en) * | 2012-03-19 | 2015-09-23 | 杭州海康威视数字技术股份有限公司 | Based on magnanimity real-time video code stream intelligent analysis method and the system thereof of cloud computing |
| CN103366568B (en) * | 2013-06-26 | 2015-10-07 | 东南大学 | Traffic section vehicle queue's video detecting method and system |
| CN104283951B (en) * | 2014-09-29 | 2018-03-27 | 华为技术有限公司 | The method, apparatus and system of a kind of instance migration |
| CN106817562A (en) * | 2015-12-01 | 2017-06-09 | 欧阳卓明 | A kind of platform software with video monitoring image intellectual analysis function |
| CN109005433B (en) * | 2018-09-04 | 2019-05-21 | 北京邮电大学 | A kind of video cloud service platform architecture and implementation method |
-
2019
- 2019-07-18 CN CN201910650758.2A patent/CN110365942B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN110365942A (en) | 2019-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110365942B (en) | Real-time video intelligent analysis method and system | |
| CN111177095B (en) | Log analysis method, device, computer equipment and storage medium | |
| CN109165573B (en) | Method and device for extracting video feature vector | |
| EP3848852B1 (en) | Method and apparatus for detecting temporal action of video, electronic device, storage medium, and computer program product | |
| US20170293606A1 (en) | Methods and systems for optimized selection of data features for a neuro-linguistic cognitive artifical intelligence system | |
| CN111935663B (en) | Sensor data stream processing method, device, medium and electronic equipment | |
| US20180143897A1 (en) | Determining idle testing periods | |
| US11875784B2 (en) | Methods and systems for optimized selection of data features for a neuro-linguistic cognitive artificial intelligence system | |
| CN111611479B (en) | Data processing method and related device for network resource recommendation | |
| CN111226226A (en) | Motion-based object detection method, object detection device and electronic equipment | |
| CN111796997A (en) | Log information processing method and device and electronic equipment | |
| CN113762906A (en) | Method, device and equipment for alarming task period delay and storage medium | |
| CN111327946A (en) | Video quality evaluation and feature dictionary training method, device and medium | |
| Lu et al. | Turbo: Opportunistic enhancement for edge video analytics | |
| CN114245232B (en) | Video abstract generation method and device, storage medium and electronic equipment | |
| CN111552613A (en) | Thread timeout processing method and device and electronic equipment | |
| KR101091185B1 (en) | Apparatus and method for analyzing news data | |
| CN114419502A (en) | Data analysis method and device and storage medium | |
| CN112040090A (en) | Video stream processing method and device, electronic equipment and storage medium | |
| CN112766470A (en) | Feature data processing method, instruction sequence generation method, device and equipment | |
| JP2013196700A (en) | Video processor, video processing method and equipment | |
| Gunasekaran et al. | Correlating log messages for system diagnostics | |
| CN115033457B (en) | Multi-source data real-time acquisition method and system capable of monitoring and early warning | |
| CN111046228B (en) | Video processing method based on stream calculation | |
| CN114697761B (en) | Processing method, processing device, terminal equipment and 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 |