CN105975910A - Technology processing method used for carrying out video identification on moving object and system thereof - Google Patents
Technology processing method used for carrying out video identification on moving object and system thereof Download PDFInfo
- Publication number
- CN105975910A CN105975910A CN201610271771.3A CN201610271771A CN105975910A CN 105975910 A CN105975910 A CN 105975910A CN 201610271771 A CN201610271771 A CN 201610271771A CN 105975910 A CN105975910 A CN 105975910A
- Authority
- CN
- China
- Prior art keywords
- image
- image information
- moving object
- module
- identification
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06M—COUNTING MECHANISMS; COUNTING OF OBJECTS NOT OTHERWISE PROVIDED FOR
- G06M7/00—Counting of objects carried by a conveyor
- G06M7/08—Counting of objects carried by a conveyor wherein the direction of movement of the objects is changed at the station where they are sensed
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention discloses a technology processing method used for carrying out video identification on a moving object. The method is characterized by storing video image information in an image queue; then extracting each frame of image information successively from the image queue, carrying out target outline identification and model processing and acquiring a center point of each image block; and then using a displacement of the center points of previous and following frames of images to confirm the image block which needs to be calculated; and when the center points leave a detection area end terminal, counting the image blocks. By using the technology processing method, the moving object in a video image can be rapidly and accurately identified. By using the technology processing method used for carrying out video identification on the moving object, the moving object in the video image can be rapidly and accurately identified. The invention also discloses a system of the technology processing method.
Description
Technical field
The present invention relates to the technological treatment of a kind of video identification, more specifically, particularly relate to a kind of for moving object
Carry out the technological treatment of video identification；The invention still further relates to the system of this kind of technological treatment.
Background technology
Traditional belt feeder enumerator is to use infrared sensor to be fixed on the both sides of conveyer belt, and infrared ray, another is launched in one end
One end receives this infrared signal, when target item passes through to block infrared ray from centre, can count by flip-flop number.
According to the operation principle of above-mentioned enumerator, the detected the biggest part of target item identified, and pass in conveyer belt
It is single channel when sending, only target item is passed sequentially through singly infrared sensor and can identify counting one by one.If
Detected target item volume is the thinnest and quantity is the most, if when these target items, batch transmits on a moving belt, infrared
Sensor can not carry out batch identification to it, and traditional belt feeder enumerator can not be identified and calculate.Nowadays, with meter
Calculate the high speed development of machine technology, it would be highly desirable to invent a kind of small objects for motion and carry out the technological treatment of video identification, with
Meet the production needs of modern processing manufacturing industry.
Summary of the invention
It is an object of the invention to provide a kind of technological treatment for moving object being carried out video identification, utilizing this technology
Processing method can identify the moving object in video image fast and accurately.Present invention also offers this kind of technological treatment
System, utilize this system can automatically identify the moving object in video image and carry out quantity statistics.
The previous technical scheme of the present invention is as follows:
A kind of technological treatment for moving object being carried out video identification, including processing step as follows:
1.1 captured in real-time transmit the situation of object and are converted into video image information；
1.2 real-time video image informations described in obtaining step 1.1, and video image information is stored to image queue, with
Time the total quantity of statistics is preset as designated value；
1.3 extract each frame image information from image queue successively and it are carried out objective contour identifying processing；
1.4 by step 1.3 image information of objective contour identifying processing carry out models treated；
1.5 from through step 1.4 image information acquisition of models treated go out the central point of each image block；
Before and after 1.6 utilizations, between two two field pictures, the displacement of central point confirms calculative image block, terminates end in detection region, sends out
The total quantity of statistics, through out-of-date, is increased by 1 by existing calculative image block.
Further, in step 1.3, described objective contour identifying processing include by image information gray processing, binaryzation,
Fill and path detection processes.
Further, in step 1.4, the image information that have passed through the image information that outline identification processes is extracted all Guan Bis
Path also stores to block array.
Further, in step 1.4, each the closed path number extracted is made minimum widith, minimum constructive height and figure
Judge as block area filters, meet the closed path required and store to block array.
Further, in step 1.6, determine the position of central point between before and after two two field picture according to shooting speed and transfer rate
Move tolerance value, the displacement of central point between two two field pictures before and after in image queue is compared with tolerance value, if front and back two
In the range of the displacement of central point is in tolerance value between two field picture, then this central point is confirmed as calculative image block.
The latter technique scheme of the present invention is as follows:
Described a kind of system for moving object carries out video identification, includes successively with lower module:
Image collection module: transmit the situation of object in captured in real-time conveyer belt and be converted into video image information；
Image stored count module: for being stored to image queue by the video image information obtained in real time, simultaneously by statistics
Total quantity is set to designated value；
Outline identification module: for extracting each frame image information from image queue successively and it being carried out at objective contour identification
Reason；
Model building module: the image information of the objective contour identifying processing for outline identification resume module being obtained is carried out at model
Reason；
Central point judge module: go out the central point of each image block for model building module is processed the image information acquisition obtained；
Statistical module: between two two field pictures, the displacement of central point confirms calculative image block before and after utilizing, in detection region
Terminate end, be found to have calculative image block through out-of-date, the total quantity of statistics is increased by 1.
Further, described outline identification module includes:
Gray processing unit, for carrying out gray processing process by image information；
Binarization unit, for carrying out binary conversion treatment by the image information after gray processing；
Fill unit, for by the image completion edge through binary conversion treatment；
Path detection unit, for detecting the contour path of object.
Further, described model building module will carry out each closed path of image zooming-out of outline identification process and stored
To block array.
Further, described model building module extracts each closed path in having carried out the image that outline identification processes,
The filtration that will make minimum widith, minimum constructive height and image block area judges, stores the closed path of satisfied requirement to block number
Group.
Further, in described statistical module, determine center between before and after two two field picture according to shooting speed and transfer rate
The displacement tolerance value of point, compares the displacement of central point between two two field pictures before and after in image queue with tolerance value, if
In the range of front and back the displacement of central point is in tolerance value between two two field pictures, then this central point is confirmed as calculative figure
As block.
Compared with prior art, the device have the advantages that into:
1. a kind of technological treatment for moving object carries out video identification of the present invention, first stores video image information
To image queue, from image queue, then extract each frame image information successively and it is carried out objective contour identification, model
Processing and obtain out the central point of each image block, before and after recycling, between two two field pictures, the displacement of central point confirms calculative
Image block, terminates end when central point leaves detection region, is counted by this image block.Utilize this technological treatment can be quickly accurate
The true moving object identified in video image.
2. the system for moving object carries out video identification of the present invention, includes image collection module, image storage meter successively
Digital-to-analogue block, outline identification module, model building module, central point judge module and statistical module, by video image information successively warp
After crossing these resume module, it is possible to identify the moving object in video image quantity performed statistics.
Accompanying drawing explanation
Fig. 1 is that the present invention for carrying out the control flow chart of the technological treatment of video identification to moving object；
Fig. 2 is that the present invention for carrying out the block diagram of the system of video identification to moving object.
Detailed description of the invention
Below in conjunction with detailed description of the invention, technical scheme is described in further detail, but does not constitute the present invention
Any restriction.
With reference to shown in Fig. 1, a kind of technological treatment for moving object carries out video identification of the present invention, including as follows
Process step:
1.1 utilize and transmit object be converted into video image information in video capture device captured in real-time conveyer belt.
1.2 from video capture device real-time video image information described in obtaining step 1.1, and by video image information store
To image queue, the total quantity of statistics is preset as designated value simultaneously.
1.3 extract each frame image information from image queue successively and it are carried out objective contour identifying processing, described target
Outline identification processes and includes processing image information gray processing, binaryzation, filling and path detection.
1.4 by step 1.3 image information of objective contour identifying processing carry out models treated, will have passed through profile know
The image information of other places reason extracts all closed paths, and each the closed path number extracted is made minimum widith, minimum height
Degree and image block area filter and judge, store the closed path of satisfied requirement to block array.
1.5 from through step 1.4 in the image information of models treated, and the closed path obtaining out each image block is obtained
Central point.
Before and after 1.6 utilizations, between two two field pictures, the displacement of adjacent two central points is confirmed whether as calculative image block, however, it is determined that
Terminate end in detection region, be found to have calculative image block through out-of-date, the total quantity of statistics is increased by 1.Wherein, it is judged that
The method of calculative image block is: first determine the position of central point between before and after two two field picture according to shooting speed and transfer rate
Move tolerance value, the displacement of adjacent two central points between two two field pictures before and after in image queue compared with tolerance value,
If in the range of front and back between two two field pictures, the displacement of central point is in tolerance value, then this central point being confirmed as calculative
Image block.After completing whole video identification process, the total quantity of statistics is and automatically identifies the object sum counted.Profit
The moving object in video image can be identified fast and accurately with this technological treatment.
Shown in reference Fig. 2, a kind of system for realizing technological treatment as claimed in claim 1 of the present invention, successively
Including with lower module:
Image collection module 1, utilizes and transmits the situation of object in image collection module 1 captured in real-time conveyer belt and be converted into video
Image information.
Image stored count module 2, stores the video image information obtained in real time to image queue, simultaneously total by add up
Quantity is set to designated value.
Outline identification module 3: extract each frame image information from image queue successively and it is carried out objective contour identifying processing.
Wherein, described outline identification module 3 includes:
Gray processing unit 31, carries out gray processing process by the image information of acquisition, image is become black-white-gray trichroism.
Binarization unit 32, carries out binary conversion treatment by the image information carrying out gray processing process, makes image information become black
White mode.
Fill unit 33, the Image Acquisition through binary conversion treatment is gone out each contour path, and filling adds broad edge automatically.
Path detection unit 34, detects the contour path of object, if what the contour path of object did not close, according to thing
Body profile characteristic increases access path accordingly so that it is become closed contour path.
Model building module 4, carries out mould by the image information having carried out the objective contour identifying processing that outline identification resume module obtains
Type processes.The image information all closed paths of extraction that outline identification processes, and each Guan Bi that will extract will have been carried out
Number of path is made minimum widith, minimum constructive height and image block area and is filtered judgement, stores the closed path of satisfied requirement to block number
Group.
Central point judge module 5, processes the Guan Bi road obtaining out each image block the image information obtained from model building module
Central point is obtained in footpath.
Statistical module 6, before and after utilization, between two two field pictures, the displacement of central point confirms calculative image block, at detection region knot
Shu Duan, is found to have calculative image block through out-of-date, and the total quantity of statistics is increased by 1.Wherein, it is judged that calculative figure
As the method for block is: first determine the displacement tolerance value of central point between before and after two two field picture according to shooting speed and transfer rate,
The displacement of adjacent two central points between two two field pictures before and after in image queue is compared with tolerance value, if two two field picture front and back
Between in the range of the displacement of central point is in tolerance value, then this central point is confirmed as calculative image block.Complete
After whole video identification process, the total quantity of statistics is and automatically identifies the object sum counted.Utilize this technical finesse side
Method can identify the moving object in video image fast and accurately.
The system for moving object carries out video identification of the present invention, includes image collection module 1, image storage meter successively
Digital-to-analogue block 2, outline identification module 3, model building module 4, central point judge module 5 and statistical module 6, believe video image
After breath sequentially passes through these resume module, it is possible to identify the moving object in video image quantity performed statistics.
The foregoing is only presently preferred embodiments of the present invention, all made in the range of the spirit and principles in the present invention any amendment,
Equivalent and improvement etc., should be included within the scope of the present invention.
Claims (10)
1. one kind for carrying out the technological treatment of video identification to moving object, it is characterised in that includes processing as follows step:
1.1 captured in real-time transmit the situation of object and are converted into video image information；
1.2 real-time video image informations described in obtaining step 1.1, and video image information is stored to image queue, with
Time the total quantity of statistics is preset as designated value；
1.3 extract each frame image information from image queue successively and it are carried out objective contour identifying processing；
1.4 by step 1.3 image information of objective contour identifying processing carry out models treated；
1.5 from through step 1.4 image information acquisition of models treated go out the central point of each image block；
Before and after 1.6 utilizations, between two two field pictures, the displacement of central point confirms calculative image block, terminates end in detection region, sends out
The total quantity of statistics, through out-of-date, is increased by 1 by existing calculative image block.
A kind of technological treatment for moving object carries out video identification the most according to claim 1, its feature exists
In, in step 1.3, described objective contour identifying processing includes image information gray processing, binaryzation, filling and path
Detection processes.
A kind of technological treatment for moving object carries out video identification the most according to claim 1, its feature exists
In, in step 1.4, the image information that have passed through the image information that outline identification processes is extracted all closed paths and stored
To block array.
A kind of technological treatment for moving object carries out video identification the most according to claim 3, its feature exists
In, in step 1.4, each the closed path number extracted is made minimum widith, minimum constructive height and image block area mistake
Filter judges, meets the closed path required and stores to block array.
A kind of technological treatment for moving object carries out video identification the most according to claim 1, its feature exists
In, in step 1.6, determine the displacement tolerance of central point between before and after two two field picture according to shooting speed and transfer rate
Value, compares the displacement of central point between two two field pictures before and after in image queue with tolerance value, if in front and back between two two field pictures
In the range of the displacement of heart point is in tolerance value, then this central point is confirmed as calculative image block.
6. the system being used for realizing technological treatment as claimed in claim 1, it is characterised in that include following successively
Module:
Image collection module (1): transmit the situation of object in captured in real-time conveyer belt and be converted into video image information；
Image stored count module (2): for being stored to image queue by the video image information obtained in real time, simultaneously will system
The total quantity of meter is set to designated value；
Outline identification module (3): for extracting each frame image information from image queue successively and it being carried out objective contour knowledge
Other places are managed；
Model building module (4): the image information of the objective contour identifying processing for outline identification resume module being obtained carries out mould
Type processes；
Central point judge module (5): go out in each image block for model building module is processed the image information acquisition obtained
Heart point；
Statistical module (6): between two two field pictures, the displacement of central point confirms calculative image block before and after utilizing, in detection
End is terminated in region, is found to have calculative image block through out-of-date, and the total quantity of statistics is increased by 1.
System for moving object being carried out video identification the most according to claim 6, it is characterised in that described wheel
Wide identification module (3) including:
Gray processing unit (31), for carrying out gray processing process by image information；
Binarization unit (32), for carrying out binary conversion treatment by the image information after gray processing；
Fill unit (33), for by the image completion edge through binary conversion treatment；
Path detection unit (34), for detecting the contour path of object.
System for moving object being carried out video identification the most according to claim 6, it is characterised in that described mould
Type module (4) will carry out each closed path of image zooming-out of outline identification process and stored to block array.
System for moving object being carried out video identification the most according to claim 8, it is characterised in that described mould
Type module (4) extracts each closed path in having carried out the image that outline identification processes, and will make minimum widith,
The filtration of low height and image block area judges, stores the closed path of satisfied requirement to block array.
System for moving object being carried out video identification the most according to claim 6, it is characterised in that described
Statistical module (6) in, determine the displacement tolerance of central point between before and after two two field picture according to shooting speed and transfer rate
Value, compares the displacement of central point between two two field pictures before and after in image queue with tolerance value, if in front and back between two two field pictures
In the range of the displacement of heart point is in tolerance value, then this central point is confirmed as calculative image block.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610271771.3A CN105975910A (en) | 2016-04-27 | 2016-04-27 | Technology processing method used for carrying out video identification on moving object and system thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610271771.3A CN105975910A (en) | 2016-04-27 | 2016-04-27 | Technology processing method used for carrying out video identification on moving object and system thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105975910A true CN105975910A (en) | 2016-09-28 |
Family
ID=56994560
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610271771.3A Pending CN105975910A (en) | 2016-04-27 | 2016-04-27 | Technology processing method used for carrying out video identification on moving object and system thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105975910A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107403444A (en) * | 2017-09-12 | 2017-11-28 | 深圳市中联讯科技有限公司 | Identifying system |
| CN108830340A (en) * | 2018-05-24 | 2018-11-16 | 北京农业信息技术研究中心 | Cereal is traced to the source particle delivery device |
| CN112235598A (en) * | 2020-09-27 | 2021-01-15 | 深圳云天励飞技术股份有限公司 | Video structured processing method and device and terminal equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103150548A (en) * | 2013-01-31 | 2013-06-12 | 南京吉目希自动化科技有限公司 | Method for improving machine vision system identification accuracy |
| JP2015082323A (en) * | 2013-10-24 | 2015-04-27 | 株式会社リコー | Device and method for detecting object |
| CN105352437A (en) * | 2015-10-21 | 2016-02-24 | 广州视源电子科技股份有限公司 | Board card position detection method and device |
| CN105427626A (en) * | 2015-12-19 | 2016-03-23 | 长安大学 | Vehicle flow statistics method based on video analysis |
-
2016
- 2016-04-27 CN CN201610271771.3A patent/CN105975910A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103150548A (en) * | 2013-01-31 | 2013-06-12 | 南京吉目希自动化科技有限公司 | Method for improving machine vision system identification accuracy |
| JP2015082323A (en) * | 2013-10-24 | 2015-04-27 | 株式会社リコー | Device and method for detecting object |
| CN105352437A (en) * | 2015-10-21 | 2016-02-24 | 广州视源电子科技股份有限公司 | Board card position detection method and device |
| CN105427626A (en) * | 2015-12-19 | 2016-03-23 | 长安大学 | Vehicle flow statistics method based on video analysis |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107403444A (en) * | 2017-09-12 | 2017-11-28 | 深圳市中联讯科技有限公司 | Identifying system |
| CN108830340A (en) * | 2018-05-24 | 2018-11-16 | 北京农业信息技术研究中心 | Cereal is traced to the source particle delivery device |
| CN108830340B (en) * | 2018-05-24 | 2021-06-29 | 北京农业信息技术研究中心 | Grain tracing particle feeding device |
| CN112235598A (en) * | 2020-09-27 | 2021-01-15 | 深圳云天励飞技术股份有限公司 | Video structured processing method and device and terminal equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108986064B (en) | People flow statistical method, equipment and system | |
| CN109308460B (en) | Article detection method, system and computer readable storage medium | |
| EP3648448A1 (en) | Target feature extraction method and device, and application system | |
| CN201255897Y (en) | Human flow monitoring device for bus | |
| CN109446942A (en) | Method for tracking target, device and system | |
| CN105701476A (en) | Machine vision-based automatic identification system and method for production line products | |
| CN105976396A (en) | Method used for identifying moving object on conveyor belt in video image | |
| TWI687159B (en) | Fry counting system and fry counting method | |
| WO2014092552A2 (en) | Method for non-static foreground feature extraction and classification | |
| CN109033985B (en) | Commodity identification processing method, device, equipment, system and storage medium | |
| CN106067031B (en) | Based on artificial mechanism for correcting errors and deep learning network cooperation machine vision recognition system | |
| CN107657244A (en) | A kind of human body tumble behavioral value system and its detection method based on multiple-camera | |
| CN110533654A (en) | The method for detecting abnormality and device of components | |
| US11200681B2 (en) | Motion detection method and motion detection system with low computational complexity and high detection accuracy | |
| CN105975910A (en) | Technology processing method used for carrying out video identification on moving object and system thereof | |
| CN111415339B (en) | Image defect detection method for complex texture industrial product | |
| CN113723325A (en) | Tool defect detection system for prefabricated parts | |
| US11804059B2 (en) | Pedestrian re-identification method and apparatus | |
| CN105956656A (en) | Equipment for counting moving objects | |
| CN111340569A (en) | Store people stream analysis method, device, system, terminal and medium based on cross-border tracking | |
| CN110889347A (en) | Density traffic flow counting method and system based on space-time counting characteristics | |
| CN106780599A (en) | A kind of circular recognition methods and system based on Hough changes | |
| CN112800806A (en) | Object pose detection tracking method and device, electronic equipment and storage medium | |
| CN109064457A (en) | A kind of method and system that the quantity of view-based access control model is accumulative | |
| CN110572618B (en) | Illegal photographing behavior monitoring method, device and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160928 |