CN107948586A - Trans-regional moving target detecting method and device based on video-splicing - Google Patents
Trans-regional moving target detecting method and device based on video-splicing Download PDFInfo
- Publication number
- CN107948586A CN107948586A CN201711120659.0A CN201711120659A CN107948586A CN 107948586 A CN107948586 A CN 107948586A CN 201711120659 A CN201711120659 A CN 201711120659A CN 107948586 A CN107948586 A CN 107948586A
- Authority
- CN
- China
- Prior art keywords
- splicing
- video
- video stream
- trans
- detecting method
- 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
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000001514 detection method Methods 0.000 claims abstract description 67
- 238000004422 calculation algorithm Methods 0.000 claims description 15
- 238000012544 monitoring process Methods 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 claims description 10
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 7
- 230000004927 fusion Effects 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 238000012360 testing method Methods 0.000 claims description 5
- 230000007246 mechanism Effects 0.000 abstract description 4
- 238000004458 analytical method Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 230000000007 visual effect Effects 0.000 description 5
- 238000010801 machine learning Methods 0.000 description 4
- HUTDUHSNJYTCAR-UHFFFAOYSA-N ancymidol Chemical compound C1=CC(OC)=CC=C1C(O)(C=1C=NC=NC=1)C1CC1 HUTDUHSNJYTCAR-UHFFFAOYSA-N 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000007689 inspection Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005111 flow chemistry technique Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/285—Analysis of motion using a sequence of stereo image pairs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
- G06T7/337—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving reference images or patches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
- G06T2207/10021—Stereoscopic video; Stereoscopic image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of trans-regional moving target detecting method and device based on video-splicing, the described method includes：Read input of the multichannel live video stream as video flowing at the same time；The multichannel live video stream is spliced, splicing video flowing is obtained and is stored in splicing video stream data area；The splicing video flowing of the data is read from splicing video stream data area at the same time, carries out moving object detection.The present invention is handled splicing video flowing using multi-thread mechanism, improves moving object detection efficiency.
Description
Technical field
The present invention relates to monitor video process field, more particularly to a kind of trans-regional moving target inspection based on video-splicing
Survey method and apparatus.
Background technology
During growing in smart city and artificial intelligence, based on the intellectual analysis of monitor video in people day
Often played an increasingly important role in life.And moving object detection is the basis of monitor video intellectual analysis, video intelligence
Can analysis in many functions have relied on moving object detection as a result, for example, crowd density statistics, intelligent barrier avoiding, mesh
Mark is retrieved, driven in the wrong direction, swarming into and break, hover, so the moving object detection based on monitor video has stronger researching value.
But the moving object detection of monitoring field is all based on single monitoring area (i.e. the region of single camera monitoring) to realize at present,
So as to which the visual field of the intellectual analysis to video is also limited to single monitoring area, the visual field of moving object detection can not be expanded.
With the continuous development of computer vision technique, current video-splicing technology and its application are very ripe,
Image mosaic is exactly that several single pictures with certain overlapping region are synthesized the picture of a large scene, so as to expand
The scene visual field of single picture and content.And in field of video monitoring, in order to the content in video is identified rapidly and
Analysis, the splicing efficiency for different video flowings have very high requirement.
Meanwhile during the application of computer vision technique is constantly expanded, for the result of moving object detection
There is the requirement of high-accuracy, high time efficiency.And the moving object detection algorithm based on traditional machine learning can not meet
The demand of detection.
Therefore, the motion estimate efficiency across video area how is improved, is that those skilled in the art urgently solve at present
Technical problem certainly.
The content of the invention
To overcome above-mentioned the deficiencies in the prior art, the present invention provides a kind of cross-domain area's moving target based on video-splicing
Detection method, can utilize video-splicing technology to realize the video-splicing based on more monitoring areas so as to expand monitoring visual field, herein
On the basis of using depth learning technology realize moving object detection, reach the high-accuracy of moving object detection, high time efficiency
Requirement, eliminate the limitation of single monitoring area target detection, expand the regional vision of moving object detection, it is achieved thereby that
Trans-regional moving object detection.
To achieve the above object, the present invention adopts the following technical scheme that：
A kind of trans-regional moving target detecting method based on video-splicing, comprises the following steps：
Read input of the multichannel live video stream as video flowing at the same time；
The multichannel live video stream is spliced, splicing video flowing is obtained and is stored in splicing video stream data area；
The splicing video flowing of the data is read from splicing video stream data area at the same time, carries out moving object detection.
Further, the method further includes after carrying out moving object detection：By the moving target detected in video flowing
In mark and show in real time.
Further, multichannel live video stream is carried out splicing includes：
Registration is carried out to the multi-path video stream of reading, registration parameter is calculated and preserves；
For the multi-path video stream subsequently read in, registration parameter is read, is merged based on the registration parameter.
Further, the multi-path video stream of described pair of reading, which carries out registration, includes：
(1) feature extraction is carried out to the multi-path video stream of reading；
(2) feature of extraction is matched；
(3) according to the matching result of feature, camera parameter and homography matrix are calculated；
(4) image coordinate system of different video stream is mapped to by same third-party spherical coordinate according to homography matrix
System；
(5) it is mapped to according to different video stream homography conversion in same coordinate system, between calculating different video stream
Overlapping region；
(6) splicing seams between different video stream are found in overlapping region dynamic using max-flow min-cut algorithm, calculated
Subsequent registration parameter simultaneously preserves.
Further, the multi-path video stream fusion subsequently read in is included：
Registration parameter is read, homography conversion is carried out to the video flowing subsequently read；
The fusion of video flowing is carried out based on the registration parameter.
Further, the moving object detection includes：
Target in monitoring video frame is labeled, as training and test data set；
Based on training and test data set training moving object detection model；
The splicing video flowing for reading splicing video stream data area in real time carries out moving object detection.
Further, it is described to identify the coordinate and classification information for including label target.
Further, the moving object detection model is to be based on DarkNet frame application yolo network models.
Second purpose according to the present invention, present invention also offers a kind of trans-regional moving target inspection based on video-splicing
Survey device, including memory, processor (including CPU and GPU) and storage are on a memory and the meter that can run on a processor
Calculation machine program, the processor realize the trans-regional moving object detection side based on video-splicing when performing described program
Method.
3rd purpose according to the present invention, present invention also offers a kind of computer-readable recording medium, is stored thereon with
Computer program, performs the trans-regional moving object detection side based on video-splicing when which is executed by processor
Method.
Beneficial effects of the present invention
1st, the present invention is handled splicing video flowing using multi-thread mechanism, in video stream splicing and moving object detection
Between establish a splicing video stream data area and be responsible for storage splicing video flowing, for splicing video stream data area, video-splicing
Process is responsible for the video flowing after having spliced being stored in the data field, and the splicing that detection process of moving target is responsible for reading the data regards
Frequency flow so that video stream splicing and moving object detection are carried out at the same time parallel, not be pipeline system but parallel type, improve
The operational efficiency of algorithm entirety.
2nd, the registration parameter that the present invention calculates between different video stream first in video-splicing is preserved into file, then right
In subsequent video stream, read the registration parameter and realize fusion, realize the real-time of video stream splicing.
3rd, the present invention make use of the middle yolo algorithms of depth learning technology to complete moving target in moving object detection part
Detection, relative to the target detection of traditional machine learning algorithm, is all greatly improved in speed and accuracy rate.
Brief description of the drawings
The accompanying drawings which form a part of this application are used for providing further understanding of the present application, and the application's shows
Meaning property embodiment and its explanation are used to explain the application, do not form the improper restriction to the application.
Fig. 1 is a kind of trans-regional moving target detecting method overall flow schematic diagram based on video-splicing of the present invention；
Fig. 2 is the overall flow diagram of video-splicing；
Fig. 3 is video stream splicing flow in a kind of cross-domain area's moving target detecting method based on video-splicing in the present invention
Schematic diagram.
Embodiment
It is noted that described further below is all illustrative, it is intended to provides further instruction to the application.It is unless another
Indicate, all technical and scientific terms used herein has usual with the application person of an ordinary skill in the technical field
The identical meanings of understanding.
It should be noted that term used herein above is merely to describe embodiment, and be not intended to restricted root
According to the illustrative embodiments of the application.As used herein, unless the context clearly indicates otherwise, otherwise singulative
It is also intended to include plural form, additionally, it should be understood that, when in the present specification using term "comprising" and/or " bag
Include " when, it indicates existing characteristics, step, operation, device, component and/or combinations thereof.
In the case where there is no conflict, the feature in the embodiment and embodiment in the application can be mutually combined.
Embodiment one
Present embodiment discloses a kind of trans-regional moving target detecting method based on video-splicing, overall structure and flow
Figure is as shown in Figure 1, it includes two large divisions：Video-splicing, moving object detection.
Video flowing inputs
By video integration platform, real-time video stream data is read from the collection of video access server, while read more
Input of the real-time video flowing in road as video flowing.
Real-time video splices
Image mosaic is several single pictures with certain overlapping region to be synthesized the picture of a large scene, from
And expand the scene visual field and the content of picture.The overall flow chart of video-splicing is as shown in Fig. 2, including feature extraction, feature
Matching, camera parameter estimation, homography conversion, searching splicing seams, image mosaic.
Current splicing is all based on video frame to realize, the whole algorithm of the splicing for plurality of pictures is more complicated
, time complexity is very high.It is to need the high time for the moving object detection in the monitoring field video intelligent analysis of security protection
The requirement of efficiency, so the high time complexity of splicing is can not to meet moving object detection real-time requirement.
A kind of cross-domain area's moving target detecting method based on video-splicing of the present invention, proposes one in video-splicing
The method of kind optimization allows the splicing of video flowing to meet real-time.The splicing of video flowing is divided into two modules by the present invention：
Registration module and concatenation module, its structure chart are as shown in Figure 3.
For the video stream splicing registration module in the present invention, the first frame of different video stream is read first, it is basic herein
The surf features of upper extraction different video frame are described as follow-up feature, and surf features generally have scale invariability and calculating
The characteristics such as speed is fast, compare and are adapted to apply in video-splicing.Extract after the surf features of different video frame, it is necessary to difference
The feature of video frame extraction carries out registration, to judge whether it has common characteristic point, selects arest neighbors and time neighbour's ratio to calculate
Method carrys out processing feature matching, the distance (selecting Euclidean distance in the present invention) of arest neighbors and time neighbour's feature pair is calculated, if most
Neighbour's feature is less than Euclidean distance Euclidean distance divided by secondary neighbour's feature the threshold value of one setting, then judges its arest neighbors spy
Sign is to being identical characteristic point.According to the matching result of characteristic point, combining camera and image coordinate system and world coordinate system
Mapping relations calculate the parameter of camera, including Intrinsic Matrix and outer parameter matrix.
Since different video flowings is what different cameras were shot, so their image coordinate system is not in same seat
In mark system, so needing the image coordinate system of different video stream being mapped in same third party's coordinate system.This third party
Coordinate system is typically chosen spherical coordinate or cylindrical coordinate system, and spherical coordinate is selected in the present invention.Utilize camera parameter
Calculating can obtain homography matrix, different video streaming image coordinate systems can be mapped to by homography matrix same
A spherical coordinate.Overlapping region between different video stream can be calculated by homography conversion, video-splicing is desirable
A splicing seams are found in the overlapping region of different video stream, this splicing seams is found using max-flow min-cut algorithm dynamic.
After the splicing seams of video flowing are found, the Parametric registration process for input video stream is just completed, that is, is extracted
Surf features, carry out characteristic matching, camera parameter estimation, homography conversion, splicing seams searching.Registration process is exactly calculated for subsequent
Splice required parameter (including the size of input video stream, homography matrix, mask of splicing seams etc.), follow-up splicing
Journey is namely based on this registration parameter and carries out image co-registration, and the present invention will subsequently splice required parameter after registration and press
It is written in xml document and is preserved according to customized form so that it is once to calculate that parameter, which calculates,.
After registration module, the whole flow process of video-splicing need not be repeated after the new video flowing of concatenation module reading
, only need to read the parameter that registration module is written in xml document in the present invention, according to homography matrix parameter to access
Video flowing directly carry out homography conversion and be mapped in the same coordinate system, directly utilize more ripples according to the mask parameters of splicing seams
Section (building gaussian sum laplacian pyramid) is merged, and completes the splicing of live video stream.
Splice video flow processing
For the splicing video flowing of splicing output, if directly carry out pipeline system carries out moving target using yolo algorithms
Detection, it may appear that can not meet the high time efficiency of moving object detection.The present invention is using multi-thread mechanism to splicing video flowing
Handled, a splicing video stream data area is established between video stream splicing and moving object detection be responsible for storing splicing and regard
Frequency flows, and for splicing video stream data area, video-splicing is responsible for the video flowing after having spliced being stored in the data field, moving target
Detection is responsible for reading the splicing video flowing of the data so that and video stream splicing and moving object detection are carried out at the same time parallel, without
Be pipeline system but parallel type, the efficiency of time can be improved.
Moving object detection
In video intelligent analysis, the target detection limitation based on traditional machine learning algorithm is very big, and detection is accurate
Rate is low and easily leak detection and error detection occurs, and the time efficiency of detection can not meet real-time detection demand, and traditional machine
It is characterized in needing in device learning algorithm manually designing.A kind of cross-domain area's moving target based on video-splicing of the present invention
Detection method, in moving object detection part, make use of the middle yolo algorithms of depth learning technology to complete moving object detection,
Speed and accuracy rate are all greatly improved.It is initial to need to mark substantial amounts of target (mesh to be detected based on monitoring video frame
Mark coordinate and classification information) data set, moving object detection model is trained based on DarkNet frame application yolo network models,
Trained model is preserved.
After completing detection model training process, target detection of the invention disposably initializes trained yolo detections mould
Type, the follow-up splicing video flowing for reading splicing video stream data area in real time carry out moving object detection.
Trans-regional testing result
That is detected for yolo detection models on splicing video flowing detects the coordinate and class of moving target as a result, preserving
Other information, and mark according to coordinate and classification information position and the classification of moving target in real time in video streaming, and real-time display.
Embodiment two
The purpose of the present embodiment is to provide a kind of computing device.
A kind of trans-regional moving object detection device based on video-splicing, including memory, processor (CPU and GPU)
And storage is on a memory and the computer program that can run on a processor, when the processor performs described program realization with
Lower step, including：
Step 1：Read input of the multichannel live video stream as video flowing at the same time；
Step 2：The multichannel live video stream is spliced, splicing video flowing is obtained and is stored in splicing video fluxion
According to area；
Step 3：The splicing video flowing of the data is read from splicing video stream data area at the same time, carries out moving object detection.
Embodiment three
The purpose of the present embodiment is to provide a kind of computer-readable recording medium.
A kind of computer-readable recording medium, is stored thereon with computer program, which performs when being executed by processor
Following steps：
Step 1：Read input of the multichannel live video stream as video flowing at the same time；
Step 2：The multichannel live video stream is spliced, splicing video flowing is obtained and is stored in splicing video fluxion
According to area；
Step 3：The splicing video flowing of the data is read from splicing video stream data area at the same time, carries out moving object detection.
Each step involved in the device of above example two and three is corresponding with embodiment of the method one, embodiment
Reference can be made to the related description part of embodiment one.Term " computer-readable recording medium " is construed as including one or more
The single medium or multiple media of instruction set；Any medium is should also be understood as including, any medium can be stored, compiled
Code or carrying are used for the instruction set performed by processor and processor is performed the either method in the present invention.
The present invention is handled splicing video flowing using multi-thread mechanism so that video stream splicing and moving object detection
Be carried out at the same time parallel, without be pipeline system but parallel type, improve the operational efficiency of algorithm entirety.In video-splicing
The registration parameter between different video stream is calculated first to preserve into file, then for subsequent video stream, reads the registration ginseng
Number realizes fusion, realizes the real-time of video stream splicing.In moving object detection part, make use of in depth learning technology
Yolo algorithms complete moving object detection, relative to the target detection of traditional machine learning algorithm, in speed and accuracy rate all
It is greatly improved.
It will be understood by those skilled in the art that each module or each step of the invention described above can be filled with general computer
Put to realize, alternatively, they can be realized with the program code that computing device can perform, it is thus possible to which they are stored
Performed in the storage device by computing device, either they are fabricated to respectively each integrated circuit modules or by they
In multiple modules or step be fabricated to single integrated circuit module to realize.The present invention be not restricted to any specific hardware and
The combination of software.
Although above-mentioned be described the embodiment of the present invention with reference to attached drawing, model not is protected to the present invention
The limitation enclosed, those skilled in the art should understand that, on the basis of technical scheme, those skilled in the art are not
Need to make the creative labor the various modifications that can be made or deformation still within protection scope of the present invention.
Claims (10)
1. a kind of trans-regional moving target detecting method based on video-splicing, it is characterised in that comprise the following steps：
Read input of the multichannel live video stream as video flowing at the same time；
The multichannel live video stream is spliced, splicing video flowing is obtained and is stored in splicing video stream data area；At the same time
The splicing video flowing of the data is read from splicing video stream data area, carries out moving object detection.
2. the trans-regional moving target detecting method based on video-splicing as claimed in claim 1, it is characterised in that the side
Method further includes after carrying out moving object detection：The moving target detected is marked and shown in real time in video streaming.
3. the trans-regional moving target detecting method based on video-splicing as claimed in claim 1, it is characterised in that by multichannel
Live video stream, which carries out splicing, to be included：
Registration is carried out to the multi-path video stream of reading, registration parameter is calculated and preserves；
For the multi-path video stream subsequently read in, registration parameter is read, is merged based on the registration parameter.
4. the trans-regional moving target detecting method based on video-splicing as claimed in claim 3, it is characterised in that described right
The multi-path video stream of reading, which carries out registration, to be included：
(1) feature extraction is carried out to the multi-path video stream of reading；
(2) feature of extraction is matched；
(3) according to the matching result of feature, camera parameter and homography matrix are calculated；
(4) image coordinate system of different video stream is mapped to by same third-party spherical coordinate according to homography matrix；
(5) it is mapped to according to different video stream homography conversion in same coordinate system, it is overlapping between calculating different video stream
Region；
(6) splicing seams between different video stream, calculated for subsequent are found in overlapping region dynamic using max-flow min-cut algorithm
Registration parameter simultaneously preserves.
5. the trans-regional moving target detecting method based on video-splicing as claimed in claim 4, it is characterised in that to follow-up
The multi-path video stream fusion of reading includes：
Registration parameter is read, homography conversion is carried out to the video flowing subsequently read；
The fusion of video flowing is carried out based on the registration parameter.
6. the trans-regional moving target detecting method based on video-splicing as claimed in claim 1, it is characterised in that the fortune
Moving-target detection includes：
Target in monitoring video frame is labeled, as training and test data set；
Based on training and test data set training moving object detection model；
The splicing video flowing for reading splicing video stream data area in real time carries out moving object detection.
7. the trans-regional moving target detecting method based on video-splicing as claimed in claim 6, it is characterised in that the mark
Knowing includes the coordinate and classification information of label target.
8. the trans-regional moving target detecting method based on video-splicing as claimed in claim 6, it is characterised in that the fortune
Moving-target detection model is to be based on DarkNet frame application yolo network models.
9. a kind of trans-regional moving object detection device based on video-splicing, including memory, processor and it is stored in storage
On device and the computer program that can run on a processor, it is characterised in that the processor is realized such as when performing described program
The trans-regional moving target detecting method based on video-splicing described in claim 1-8.
10. a kind of computer-readable recording medium, is stored thereon with computer program, it is characterised in that the program is by processor
The trans-regional moving target detecting method based on video-splicing as described in claim 1-8 is performed during execution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711120659.0A CN107948586B (en) | 2017-11-14 | 2017-11-14 | Trans-regional moving target detecting method and device based on video-splicing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711120659.0A CN107948586B (en) | 2017-11-14 | 2017-11-14 | Trans-regional moving target detecting method and device based on video-splicing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107948586A true CN107948586A (en) | 2018-04-20 |
| CN107948586B CN107948586B (en) | 2019-04-16 |
Family
ID=61935063
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711120659.0A Active CN107948586B (en) | 2017-11-14 | 2017-11-14 | Trans-regional moving target detecting method and device based on video-splicing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107948586B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109714623A (en) * | 2019-03-12 | 2019-05-03 | 北京旷视科技有限公司 | Image presentation method, device, electronic equipment and computer readable storage medium |
| CN109934848A (en) * | 2019-03-07 | 2019-06-25 | 贵州大学 | A method of the moving object precise positioning based on deep learning |
| CN113873200A (en) * | 2021-09-26 | 2021-12-31 | 珠海研果科技有限公司 | Image identification method and system |
| CN114222162A (en) * | 2021-12-07 | 2022-03-22 | 浙江大华技术股份有限公司 | Video processing method, video processing device, computer equipment and storage medium |
| CN114842028A (en) * | 2022-05-07 | 2022-08-02 | 深圳先进技术研究院 | Cross-video target tracking method, system, electronic equipment and storage medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101593353A (en) * | 2008-05-28 | 2009-12-02 | 日电（中国）有限公司 | Image processing method and equipment and video system |
| CN102063724A (en) * | 2010-11-25 | 2011-05-18 | 四川省绵阳西南自动化研究所 | Panoramic virtual alert target relay tracking device |
| CN102156863A (en) * | 2011-05-16 | 2011-08-17 | 天津大学 | Cross-camera tracking method for multiple moving targets |
| CN102682435A (en) * | 2012-05-14 | 2012-09-19 | 四川大学 | Multi-focus image edge detection method based on space relative altitude information |
| CN103020941A (en) * | 2012-12-28 | 2013-04-03 | 昆山市工业技术研究院有限责任公司 | Panoramic stitching based rotary camera background establishment method and panoramic stitching based moving object detection method |
| CN103985254A (en) * | 2014-05-29 | 2014-08-13 | 四川川大智胜软件股份有限公司 | Multi-view video fusion and traffic parameter collecting method for large-scale scene traffic monitoring |
| CN104535047A (en) * | 2014-09-19 | 2015-04-22 | 燕山大学 | Multi-agent target tracking global positioning system and method based on video stitching |
| CN104581000A (en) * | 2013-10-12 | 2015-04-29 | 北京航天长峰科技工业集团有限公司 | Method for rapidly retrieving motional trajectory of interested video target |
| CN104639916A (en) * | 2015-03-04 | 2015-05-20 | 合肥巨清信息科技有限公司 | Large-scene multi-target tracking shooting video monitoring system and monitoring method thereof |
| CN105407278A (en) * | 2015-11-10 | 2016-03-16 | 北京天睿空间科技股份有限公司 | Panoramic video traffic situation monitoring system and method |
-
2017
- 2017-11-14 CN CN201711120659.0A patent/CN107948586B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101593353A (en) * | 2008-05-28 | 2009-12-02 | 日电（中国）有限公司 | Image processing method and equipment and video system |
| CN102063724A (en) * | 2010-11-25 | 2011-05-18 | 四川省绵阳西南自动化研究所 | Panoramic virtual alert target relay tracking device |
| CN102156863A (en) * | 2011-05-16 | 2011-08-17 | 天津大学 | Cross-camera tracking method for multiple moving targets |
| CN102682435A (en) * | 2012-05-14 | 2012-09-19 | 四川大学 | Multi-focus image edge detection method based on space relative altitude information |
| CN103020941A (en) * | 2012-12-28 | 2013-04-03 | 昆山市工业技术研究院有限责任公司 | Panoramic stitching based rotary camera background establishment method and panoramic stitching based moving object detection method |
| CN104581000A (en) * | 2013-10-12 | 2015-04-29 | 北京航天长峰科技工业集团有限公司 | Method for rapidly retrieving motional trajectory of interested video target |
| CN103985254A (en) * | 2014-05-29 | 2014-08-13 | 四川川大智胜软件股份有限公司 | Multi-view video fusion and traffic parameter collecting method for large-scale scene traffic monitoring |
| CN104535047A (en) * | 2014-09-19 | 2015-04-22 | 燕山大学 | Multi-agent target tracking global positioning system and method based on video stitching |
| CN104639916A (en) * | 2015-03-04 | 2015-05-20 | 合肥巨清信息科技有限公司 | Large-scene multi-target tracking shooting video monitoring system and monitoring method thereof |
| CN105407278A (en) * | 2015-11-10 | 2016-03-16 | 北京天睿空间科技股份有限公司 | Panoramic video traffic situation monitoring system and method |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109934848A (en) * | 2019-03-07 | 2019-06-25 | 贵州大学 | A method of the moving object precise positioning based on deep learning |
| CN109934848B (en) * | 2019-03-07 | 2023-05-23 | 贵州大学 | Method for accurately positioning moving object based on deep learning |
| CN109714623A (en) * | 2019-03-12 | 2019-05-03 | 北京旷视科技有限公司 | Image presentation method, device, electronic equipment and computer readable storage medium |
| CN109714623B (en) * | 2019-03-12 | 2021-11-16 | 北京旷视科技有限公司 | Image display method and device, electronic equipment and computer readable storage medium |
| CN113873200A (en) * | 2021-09-26 | 2021-12-31 | 珠海研果科技有限公司 | Image identification method and system |
| CN113873200B (en) * | 2021-09-26 | 2024-02-02 | 珠海研果科技有限公司 | Image identification method and system |
| CN114222162A (en) * | 2021-12-07 | 2022-03-22 | 浙江大华技术股份有限公司 | Video processing method, video processing device, computer equipment and storage medium |
| CN114222162B (en) * | 2021-12-07 | 2024-04-12 | 浙江大华技术股份有限公司 | Video processing method, device, computer equipment and storage medium |
| CN114842028A (en) * | 2022-05-07 | 2022-08-02 | 深圳先进技术研究院 | Cross-video target tracking method, system, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107948586B (en) | 2019-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sahu et al. | Artificial intelligence (AI) in augmented reality (AR)-assisted manufacturing applications: a review | |
| CN110717527B (en) | Method for determining target detection model by combining cavity space pyramid structure | |
| CN107948586A (en) | Trans-regional moving target detecting method and device based on video-splicing | |
| CN114202672A (en) | Small target detection method based on attention mechanism | |
| CN103262119B (en) | For the method and system that image is split | |
| CN110147743A (en) | Real-time online pedestrian analysis and number system and method under a kind of complex scene | |
| CN109635882A (en) | Salient object detection method based on multi-scale convolution feature extraction and fusion | |
| CN108830185B (en) | Behavior identification and positioning method based on multi-task joint learning | |
| CN111311475A (en) | Detection model training method and device, storage medium and computer equipment | |
| CN110705566B (en) | Multi-mode fusion significance detection method based on spatial pyramid pool | |
| CN110956646B (en) | Target tracking method, device, equipment and storage medium | |
| CN113609896A (en) | Object-level remote sensing change detection method and system based on dual-correlation attention | |
| CN109410211A (en) | The dividing method and device of target object in a kind of image | |
| Rong et al. | Pest Identification and Counting of Yellow Plate in Field Based on Improved Mask R‐CNN | |
| CN105005760A (en) | Pedestrian re-identification method based on finite mixture model | |
| CN110866936A (en) | Video labeling method, tracking method, device, computer equipment and storage medium | |
| CN112634369A (en) | Space and or graph model generation method and device, electronic equipment and storage medium | |
| CN110298281A (en) | Video structural method, apparatus, electronic equipment and storage medium | |
| WO2023221608A1 (en) | Mask recognition model training method and apparatus, device, and storage medium | |
| CN111738036A (en) | Image processing method, device, equipment and storage medium | |
| CN109919149A (en) | Object mask method and relevant device based on object detection model | |
| CN113850136A (en) | Yolov5 and BCNN-based vehicle orientation identification method and system | |
| CN110544202A (en) | parallax image splicing method and system based on template matching and feature clustering | |
| CN112417947A (en) | Method and device for optimizing key point detection model and detecting face key points | |
| CN115147644A (en) | Method, system, device and storage medium for training and describing image description model |
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 |