CN110379130A - A kind of Medical nursing shatter-resistant adjustable voltage system based on multi-path high-definition SDI video - Google Patents
A kind of Medical nursing shatter-resistant adjustable voltage system based on multi-path high-definition SDI video Download PDFInfo
- Publication number
- CN110379130A CN110379130A CN201910576131.7A CN201910576131A CN110379130A CN 110379130 A CN110379130 A CN 110379130A CN 201910576131 A CN201910576131 A CN 201910576131A CN 110379130 A CN110379130 A CN 110379130A
- Authority
- CN
- China
- Prior art keywords
- video
- resistant
- medical nursing
- motion vector
- shatter
- 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
- 230000000474 nursing effect Effects 0.000 title claims abstract description 24
- 230000033001 locomotion Effects 0.000 claims abstract description 51
- 238000012706 support-vector machine Methods 0.000 claims abstract description 9
- 238000005538 encapsulation Methods 0.000 claims abstract description 6
- 239000000284 extract Substances 0.000 claims abstract description 6
- 230000000877 morphologic effect Effects 0.000 claims abstract description 6
- 238000001914 filtration Methods 0.000 claims abstract description 5
- 238000012545 processing Methods 0.000 claims description 20
- 238000000034 method Methods 0.000 claims description 13
- 238000001514 detection method Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 6
- 230000006835 compression Effects 0.000 claims description 3
- 238000007906 compression Methods 0.000 claims description 3
- 150000001875 compounds Chemical class 0.000 claims description 2
- 230000009471 action Effects 0.000 abstract description 8
- 238000012549 training Methods 0.000 abstract description 3
- 230000006399 behavior Effects 0.000 description 22
- 238000004458 analytical method Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 230000001360 synchronised effect Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 206010000117 Abnormal behaviour Diseases 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
- G08B21/0407—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L65/00—Network arrangements, protocols or services for supporting real-time applications in data packet communication
- H04L65/60—Network streaming of media packets
- H04L65/75—Media network packet handling
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
- H04N19/136—Incoming video signal characteristics or properties
- H04N19/137—Motion inside a coding unit, e.g. average field, frame or block difference
- H04N19/139—Analysis of motion vectors, e.g. their magnitude, direction, variance or reliability
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
- H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
- H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/85—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using pre-processing or post-processing specially adapted for video compression
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/222—Studio circuitry; Studio devices; Studio equipment
- H04N5/262—Studio circuits, e.g. for mixing, switching-over, change of character of image, other special effects ; Cameras specially adapted for the electronic generation of special effects
- H04N5/268—Signal distribution or switching
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- Gerontology & Geriatric Medicine (AREA)
- General Health & Medical Sciences (AREA)
- Emergency Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Social Psychology (AREA)
- Psychology (AREA)
- Psychiatry (AREA)
- Computer Networks & Wireless Communication (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of Medical nursing shatter-resistant adjustable voltage systems based on multi-path high-definition SDI video.Acquisition original video simultaneously initializes;Coded treatment extracts two intermediate parameters of motion vector and macro block type information, and then obtains motion vector figure, filters out local bright spot, and median filtering is successively carried out to motion vector figure and opens the Morphological scale-space of operation, obtains motion target area;Extract the motion history figure and its gradient orientation histogram feature of each road video;Using the training of gradient orientation histogram feature, the Classification and Identification that the original video classified to unknown Medical nursing falling-resistant carries out behavior act is completed using trained support vector machines;The encapsulation of multi-channel video splicing completes the local of video and audio and plays, plug-flow and storage is broadcast live.The present invention can meet the requirement of real-time of behavior action recognition under multi-channel video scene.For 8 road 1080P videos, the frame per second of action recognition is greater than 30fps.Plug-flow and storage record can be broadcast live to video simultaneously.
Description
Technical field
The invention belongs to video image processing technology application field, the medical treatment based on multi-path high-definition SDI video is referred in particular to
Nurse falling-resistant system research.
Background technique
Activity recognition based on video is to carry out Classification and Identification to personage's behavior act in video scene.With digital video
The development of technology and computer technology, the real-time discriminance analysis of behavior act based on video become a new research hotspot.
Since it is owned by huge market application prospect in various industries such as education, medical treatment, security protection, traffic.Artificial intelligence it is quick
Development imparts the ability of the equipment such as video camera thinking judgement.But video camera is widely available in current industry, passes through rear end
One common video camera can be transformed into AI video camera by the intervention of equipment.In behavior action recognition field, it is big to occur one
The algorithm criticized, such as it is based on CNN, RCNN, Fast RCNN, Faster RCNN, with the increase of network structure scale, in order to full
The demand of sufficient real-time proposes very high requirement for the calculation power of computer.
Currently, the behavior act identification based on video is mainly off-line operation, for the video data kept of recording into
The processing of row discriminance analysis.In terms of real-time identifying processing, the resolution ratio and frame per second of processing are all lower.View based on hardware such as FPGA
Frequency is handled, although can be improved the real-time of system, the development difficulty of FPGA is high and the period is long, is not suitable for multi-channel video field
Behavior act identification under scape.Traditional PC machine, although operational capability is stronger, does not have standard after being equipped with GPU acceleration
SDI video interface, can not the SDI interface high-definition camera machine equipment for multichannel handled in real time.Current row real-time on the market
It is all more expensive for the video camera price of action recognition.Therefore embedded host system can be connect by customizing video input
Mouthful, exploitation is rapidly completed, completes the real-time discriminance analysis of behavior act of multichannel SDI video, keeps common camera intelligent.Therefore
One kind is needed to be able to achieve the intelligentized host equipment of multiple cameras.
Summary of the invention
In order to solve the problems, such as that background technique, the present invention are directed to the Medical nursing based on multi-path high-definition SDI video
Shatter-resistant adjustable voltage system proposes a kind of Medical nursing shatter-resistant adjustable voltage system based on multi-path high-definition SDI video, realizes with falling-resistant
The early warning of function.
The technical solution adopted by the present invention comprises the steps of:
Step S1: high definition SDI video camera acquires original video, detects parameter information (such as resolution ratio, frame per second of each road video
Deng), complete system initialization;
Step S2: coded treatment is carried out using FFmpeg codec, realizes the coding of each road input video, is extracted
Two intermediate parameters of the motion vector of each single-frame images and macro block type information in FFmpeg cataloged procedure;
Step S3: according to the motion vector and macro block type information of each road video of acquisition, by the speed in motion vector
Information is shown in the form of brightness obtains motion vector figure, filters out its corresponding small range part bright spot using macro block type information,
Median filtering is successively carried out to motion vector figure and opens the Morphological scale-space of operation, obtains motion target area, completes quickly fortune
Moving-target detection;
Step S4: after obtaining motion target area, each road video is extracted according to the motion target area of each frame image
Motion history figure (MHI) then extracts the gradient orientation histogram feature of motion history figure;
Step S5: the original video of known Medical nursing falling-resistant classification is obtained into gradient direction by above-mentioned steps processing
Histogram feature inputs the original video of known Medical nursing falling-resistant classification and its corresponding gradient orientation histogram feature
It is trained to support vector machines, the support vector machines after being trained;
The original video classified to unknown Medical nursing falling-resistant is completed using trained support vector machines and carries out behavior
The Classification and Identification of movement;
If recognition detection is to the abnormal behaviour fallen down, then warn;Otherwise without warning.
Step S6: carrying out splicing to the multi-channel video of input, carries out video combinatorial compound and video and compresses volume all the way
Code, can be substantially reduced the bandwidth pressure of network in this way, improve the stability of transmission process.Video and audio is finally packaged place
Reason completes the local of video and audio and plays, plug-flow and storage is broadcast live.
The classification of Medical nursing falling-resistant includes two classes fallen down and do not fallen down.
In the step S3, by video coding process motion vector and macro block type information be used for moving target inspection
It surveys, the moving target locating speed based on H.264 encoded video can be significantly improved.
In the step S4, the quick foreground moving object detection algorithm based on FFmpeg encoder can be improved and obtain
Take the speed of motion history figure.
In the step S4, motion history figure and gradient orientation histogram feature are combined, motion history figure can be more
It mends gradient orientation histogram and ignores the deficiency of information on action sequence, can not only represent the position that movement occurs, it can be with
The information for acting the sequencing in timing is represented, the precision of behavior act identification can be significantly improved.
The processing of the step 3-5 of method of the invention can realize fast-moving target detection in embedded platform, improve inspection
Survey processing speed.
In the step S6 and S7, video compression coding, video and audio encapsulation, live streaming plug-flow, storage and etc. modularization
Processing is synchronized by different processes, improves the operational efficiency of system and the robustness of program.
In the present invention for video encoding, encapsulation, live streaming plug-flow and storage and etc. carried out modularized processing.It improves
The efficiency of system operation, and improve the stability of system.
The present invention having the beneficial effect that compared with the existing technology
The present invention can guarantee real-time video behavior act (the refering in particular to fall down) identification of multichannel SDI HD video video camera.
It realizes and AI video camera is substituted using common SDI video camera, reduce the cost of investment, and realize the real-time processing of video
And early warning.In Medical nursing field, the investment of manpower is reduced, is had for the personnel such as home for destitute, sanatorium falling-resistant early warning
Significance.
The present invention can satisfy the requirement of real-time of behavior action recognition under multi-channel video scene.8 road 1080P are regarded
Frequently, the frame per second of action recognition is greater than 30fps.Plug-flow and storage record can be broadcast live to video simultaneously.
Detailed description of the invention
The present invention will be described in detail with reference to the accompanying drawings and examples, and advantages and features of the invention is enable to be easier to
It is readily appreciated by one skilled in the art, so as to make a clearer definition of the protection scope of the present invention.
Fig. 1 is the Medical nursing shatter-resistant adjustable voltage system audio-video process flow diagram of multi-path high-definition SDI video;
Fig. 2 is the Medical nursing shatter-resistant adjustable voltage system complete machine frame diagram of multi-path high-definition SDI video;
Fig. 3 is the top-down module design figure of Medical nursing shatter-resistant adjustable voltage system of multi-path high-definition SDI video.
Specific embodiment
The invention proposes a kind of complete machine solution party of Medical nursing shatter-resistant adjustable voltage system based on multi-path high-definition SDI video
Case.The main flow that it is handled is as shown in Figure 1.
The system include to the real-time coding of multichannel SDI HD video, Activity recognition early warning, image mosaic, audio-video encapsulation,
The committed steps such as plug-flow, audio-video storage are broadcast live.Original video is inputted by SDI high-definition camera first, MIC inputs original sound
Frequency evidence, then obtain input video resolution ratio and frame per second information complete system (SYS) initialization and video input (VI) and
The configuration of video image processing (VPSS).Then H.264 coded treatment is synchronized to the multi-channel video of input, obtained encoded
Then motion vector and macro block type information obtained in journey restore motion vector figure, cooperate image preprocessing (image grayscale
Change, image denoising, morphological operation) quick moving object detection can be completed, the effect detected according to foreground moving object
Motion history figure is extracted, the gradient orientation histogram feature of image is then extracted, which is used into trained branch
It holds vector machine and carries out Classification and Identification judgement, can effectively judge that specified behavior act carries out early warning.By the video of multichannel
It is combined splicing as needed, compressed encoding is then carried out, audio-video is finally packaged into specified format and is broadcast live
Plug-flow and storage.
Step S1: 8 tunnel high definition SDI video cameras of acquisition are originally inputted, and are detected two pieces of strings and are turned and data chip GV7704
Working condition obtains the resolution ratio and frame per second information of SDI interface input video, then completes a series of initialization and view of system
Frequency input parameter configuration.
8 road high-definition cameras are connected especially by SDI interface, the system two pieces of GV7704 chips of collocation complete SDI video
Data turn the parallel port BT1120 data, and the video data of parallel port is input to Hi3531A primary processor.Hi3531A chip by pair
The reading of GV7704 chip status gets the resolution ratio and frame per second of 8 road input videos.Master chip completes a series of adding for drivings
It carries, and completes the initialization of system, video input and video image processing unit.
Step S2: the original view of every road video input is obtained from the image processing unit corresponding channel being arranged in step S1
Then frequency evidence carries out coded treatment to original video data using the encoder in FFmpeg open source library, using H.264 pressing
Code is reduced the staff, the data volume of video is effectively reduced.The data volume of original video can be reduced into 50-100 after H.264 compressed encoding
Times, it greatly reduces video and stores occupied space and the bandwidth pressure that plug-flow is born is broadcast live.
During FFmpeg coding, the corresponding motion vector information of single frame video image and macro block (mb) type letter are got
Breath, this is all the additional intermediate parameters of video in cataloged procedure.
Step S3: the motion vector figure acquired by step S2 contains more isolated noise, and moving region
Discontinuously, therefore at this stage, according to the motion vector and macro block type information got in step S2, motion vector is restored
Figure and macro block (mb) type figure, by two images combination image preprocessing (specifically including image gray processing, filtering and morphological operation)
The quick positioning of achievable motion target area.
Specifically, according to the motion vector and macro block type information of each road video of acquisition, by the speed in motion vector
Information is shown in the form of brightness obtains motion vector figure, filters out its corresponding small range part bright spot using macro block type information,
Median filtering is successively carried out to motion vector figure and opens the Morphological scale-space of operation, obtains motion target area, completes quickly fortune
Moving-target detection;
Special additional intermediate parameters motion vector and macro block (mb) type letter using in image encoding process in step of the present invention
Breath, by Experimental comparison, the processing of such moving object detection is transported compared to prospects such as traditional GMM, Vibe and CodeBook
Moving-target detection algorithm has in time efficiency to be obviously improved.It is also more applicable in real-time system.
Step S4: required feature is identified in order to extract behavior act in the process.
In order to carry out effective behavior act identification, the feature of selective extraction video frame is identified.It is mentioned according to step S3
The motion target area got, can more rapidly and efficiently acquisition video image motion history figure, include movement
Location information, further include the timing information of behavior act, for behavior act representativeness preferably.
After to image normalization, the gradient orientation histogram information of motion history figure, the HOG that will finally extract are extracted
Identification feature of the feature as final video frame.
In the present system, another characteristic is known as final classification using based on MHI-HOG union feature.By comparative experiments
It was found that the union feature relative to individual HOG feature under the premise of not increasing time efficiency, for behavior act identification
Accuracy rate is obviously improved.Therefore this system final choice MHI-HOG union feature carries out behavior act Classification and Identification.
Step S5:
In order to effectively carry out behavior act identification classification to the union feature of extraction, in the present system mainly for video
Personage's tumble behavior under scene carries out identification early warning.
Enough movement to be identified building sample sets are acquired, and establish label.The identification feature of sample data is extracted, if
Surely trained the number of iterations and training precision, then instruct the behavior act recognition classifier of the support vector machines of building
Practice.
The model parameter of the support vector machines is imported after the completion of training, for each video image frame to be identified
Feature carries out the identification of behavior act using the action recognition classifier of the support vector machines, and the movement judged is carried out
Post-processing carries out early warning processing to dangerous play, can improve accuracy of identification in this way and improve time efficiency.
Step S6: when carrying out processing analysis to multi-channel video, for unitized management and centralized control.
8 road videos of input are subjected to picture composition stacking splicing operation, the video counts that will finally splice according to demand
According to coding compression is carried out, the requirement of memory space can significantly be reduced by being stored and be broadcast live plug-flow for the video that coding has compressed
With the pressure of network bandwidth.
In present invention specific implementation, the splicing for having carried out picture for the multi-path video data of input is operated.Pass through
Multi-picture splicing operation is carried out, to the bandwidth pressure that can be substantially reduced network after synthesis picture progress compressed encoding, ensure that number
According to stablize transmission.
Step S7: in the system by FFmpeg encoder by original video according to H.264 agreement by video encapsulation at
The form of h264, and from the equipment that audio frequency apparatus obtains be packaged into data from original PCM form more general and universal
AAC form.In order to increase the universality of video, video and audio synchronizes according to timestamp information to be packaged into versatility higher
The format of FLV.
According to the timestamp information of the video and audio of acquisition, completion video information is synchronous with audio-frequency information, will be packaged
Video push can carry out the remote access of WEB terminal to server, and user is facilitated to carry out viewing and the video file of real-time video
Playback, facilitate user to be managed video.
Fig. 3 indicates top-down the designing a model of system design.Including human-computer interaction, software systems and data-driven three
A part.Human-computer interaction module can provide the visualization interface of operation for user;Mainly for acquisition in terms of software systems
Video data carries out target detection, feature extraction and Activity recognition;Mainly include in terms of data-driven management to database and
Access, stores necessary system parameter and behavior act information.
In terms of the present invention is applied to Medical nursing falling-resistant, there is great application prospect, the prospect in terms of Medical nursing
It is very big, nurse cost can be significantly reduced, and improve the safety of nurse.
Claims (3)
1. a kind of Medical nursing shatter-resistant adjustable voltage system based on multi-path high-definition SDI video, it is characterised in that: method includes following step
It is rapid:
Step S1: high definition SDI video camera acquires original video, detects the parameter information of each road video, completes system initialization;
Step S2: coded treatment is carried out using FFmpeg codec, extracts each single-frame images in FFmpeg cataloged procedure
Motion vector and macro block type information two intermediate parameters;
Step S3: according to the motion vector and macro block type information of each road video of acquisition, by the velocity information in motion vector
It is shown in the form of brightness and obtains motion vector figure, its corresponding local bright spot is filtered out using macro block type information, to motion vector
Figure successively carries out median filtering and opens the Morphological scale-space of operation, obtains motion target area, completes fast-moving target detection;
Step S4: after obtaining motion target area, the movement of each road video is extracted according to the motion target area of each frame image
History figure (MHI) then extracts the gradient orientation histogram feature of motion history figure;
Step S5: the original video of known Medical nursing falling-resistant classification is obtained into gradient direction histogram by above-mentioned steps processing
The original video of known Medical nursing falling-resistant classification and its corresponding gradient orientation histogram feature are input to branch by figure feature
It holds vector machine to be trained, the support vector machines after being trained;It is completed using trained support vector machines to unknown medical treatment
The original video for nursing falling-resistant classification carries out the Classification and Identification of behavior act;
Step S6: carrying out splicing to the multi-channel video of input, carries out video combinatorial compound video and compressed encoding all the way, most
Video and audio is packaged processing afterwards, the local of video and audio is completed and plays, plug-flow and storage is broadcast live.
2. the Medical nursing shatter-resistant adjustable voltage system according to claim 1 based on multi-path high-definition SDI video, it is characterised in that:
The classification of Medical nursing falling-resistant includes two classes fallen down and do not fallen down.
3. the Medical nursing shatter-resistant adjustable voltage system according to claim 1 based on multi-path high-definition SDI video, it is characterised in that:
In the step S6 and S7, video compression coding, video and audio encapsulation, live streaming plug-flow, storage and etc. modularization pass through difference
Process synchronize processing.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910576131.7A CN110379130B (en) | 2019-06-28 | 2019-06-28 | Medical nursing anti-falling system based on multi-path high-definition SDI video |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910576131.7A CN110379130B (en) | 2019-06-28 | 2019-06-28 | Medical nursing anti-falling system based on multi-path high-definition SDI video |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110379130A true CN110379130A (en) | 2019-10-25 |
CN110379130B CN110379130B (en) | 2021-04-16 |
Family
ID=68251140
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910576131.7A Expired - Fee Related CN110379130B (en) | 2019-06-28 | 2019-06-28 | Medical nursing anti-falling system based on multi-path high-definition SDI video |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110379130B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113840166A (en) * | 2021-08-31 | 2021-12-24 | 南京巨鲨显示科技有限公司 | Method and system for synchronizing audio and video mixing of multi-path streaming media |
CN115620489A (en) * | 2022-10-24 | 2023-01-17 | 深圳市自然资源和不动产评估发展研究中心(深圳市地质环境监测中心) | Slope monitoring and early warning method, system, equipment and medium based on live broadcast plug flow |
CN117640998A (en) * | 2023-12-13 | 2024-03-01 | 北京拓目科技有限公司 | Method and system for collecting multi-channel video data by MVPS (mechanical vapor compression system) series video processing system |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000059796A (en) * | 1998-06-03 | 2000-02-25 | Matsushita Electric Ind Co Ltd | Motion detecting device, motion detecting method and recording medium with motion detection program recorded therein |
CN101478675A (en) * | 2008-01-03 | 2009-07-08 | 中国科学院计算技术研究所 | Semantic events detection method and system in video |
CN104866841A (en) * | 2015-06-05 | 2015-08-26 | 中国人民解放军国防科学技术大学 | Human body object running behavior detection method |
CN107103733A (en) * | 2017-07-06 | 2017-08-29 | 司马大大(北京)智能系统有限公司 | One kind falls down alarm method, device and equipment |
-
2019
- 2019-06-28 CN CN201910576131.7A patent/CN110379130B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000059796A (en) * | 1998-06-03 | 2000-02-25 | Matsushita Electric Ind Co Ltd | Motion detecting device, motion detecting method and recording medium with motion detection program recorded therein |
CN101478675A (en) * | 2008-01-03 | 2009-07-08 | 中国科学院计算技术研究所 | Semantic events detection method and system in video |
CN104866841A (en) * | 2015-06-05 | 2015-08-26 | 中国人民解放军国防科学技术大学 | Human body object running behavior detection method |
CN107103733A (en) * | 2017-07-06 | 2017-08-29 | 司马大大(北京)智能系统有限公司 | One kind falls down alarm method, device and equipment |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113840166A (en) * | 2021-08-31 | 2021-12-24 | 南京巨鲨显示科技有限公司 | Method and system for synchronizing audio and video mixing of multi-path streaming media |
CN115620489A (en) * | 2022-10-24 | 2023-01-17 | 深圳市自然资源和不动产评估发展研究中心(深圳市地质环境监测中心) | Slope monitoring and early warning method, system, equipment and medium based on live broadcast plug flow |
CN117640998A (en) * | 2023-12-13 | 2024-03-01 | 北京拓目科技有限公司 | Method and system for collecting multi-channel video data by MVPS (mechanical vapor compression system) series video processing system |
Also Published As
Publication number | Publication date |
---|---|
CN110379130B (en) | 2021-04-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107004271B (en) | Display method, display apparatus, electronic device, computer program product, and storage medium | |
CN110008797B (en) | Multi-camera multi-face video continuous acquisition method | |
CN110379130A (en) | A kind of Medical nursing shatter-resistant adjustable voltage system based on multi-path high-definition SDI video | |
CN102799935B (en) | Human flow counting method based on video analysis technology | |
CN101631237B (en) | Video monitoring data storing and managing system | |
US9846820B2 (en) | Method and system for coding or recognizing of surveillance videos | |
KR102035592B1 (en) | A supporting system and method that assist partial inspections of suspicious objects in cctv video streams by using multi-level object recognition technology to reduce workload of human-eye based inspectors | |
CN103347167A (en) | Surveillance video content description method based on fragments | |
CN104349074A (en) | Method, device and system for producing a merged digital video sequence | |
CN108509912A (en) | Multipath network video stream licence plate recognition method and system | |
CN103379266A (en) | High-definition web camera with video semantic analysis function | |
CN102231820A (en) | Monitoring image processing method, device and system | |
CN102222104A (en) | Method for intelligently extracting video abstract based on time-space fusion | |
KR102063135B1 (en) | Cctv searching method and apparatus using deep learning | |
CN111585684B (en) | Multi-path camera time alignment method and system for networked monitoring video analysis | |
WO2018030658A1 (en) | Method for detecting, through reconstruction image processing, moving object from stored cctv image | |
CN112866654B (en) | Intelligent video monitoring system | |
CN105357523A (en) | High-order singular value decomposition (HOSVD) algorithm based video compression system and method | |
CN110674787A (en) | Video decompression method and system based on Hog feature and lgb classifier | |
CN114051124A (en) | Video monitoring method, device, equipment and storage medium supporting multi-region monitoring | |
WO2019039661A1 (en) | Method for syntax-based extraction of moving object region of compressed video | |
Ouyang et al. | The comparison and analysis of extracting video key frame | |
CN112883836A (en) | Video detection method for deformation of underground coal mine roadway | |
WO2019124634A1 (en) | Syntax-based method for object tracking in compressed video | |
CN109977891A (en) | A kind of object detection and recognition method neural network based |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210416 |
|
CF01 | Termination of patent right due to non-payment of annual fee |