CN110348394A - A method of detection video static object - Google Patents
A method of detection video static object Download PDFInfo
- Publication number
- CN110348394A CN110348394A CN201910632936.9A CN201910632936A CN110348394A CN 110348394 A CN110348394 A CN 110348394A CN 201910632936 A CN201910632936 A CN 201910632936A CN 110348394 A CN110348394 A CN 110348394A
- Authority
- CN
- China
- Prior art keywords
- target
- marquee
- deviation
- static object
- single frames
- 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
- 230000003068 static effect Effects 0.000 title claims abstract description 30
- 238000000034 method Methods 0.000 title claims abstract description 17
- 238000001514 detection method Methods 0.000 title claims abstract description 8
- 230000007547 defect Effects 0.000 description 1
- 230000001815 facial effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
Abstract
A method of detection video static object, the specific steps are as follows: step A: camera obtains video streaming image, identifies that video streaming image whether there is by third party's engine and is detected target, if then entering step B;Step B: when detecting target, the two frame single frames picture of front and back where target is obtained;Step C: in two frame single frames picture of front and back, the marquee of target region information is respectively divided out, the marquee in two frame single frames picture of front and back is completely the same;Step D: being completely coincident the target in two frame single frames picture of front and back, and judges the marquee in two frame single frames picture of front and back with the presence or absence of deviation, if then calculating deviation;Step E: setting threshold value compares threshold value and deviation, judges whether target is static object according to comparison result.The present invention proposes a kind of method for detecting video static object, distinguishes dynamic object and static object by detecting the target position deviation of before and after frames.
Description
Technical field
The present invention relates to technical field of video processing more particularly to a kind of methods for detecting video static object.
Background technique
Face recognition technology as one of research topic most popular in computer nowadays visual field, security industry,
The fields such as access control system, attendance checking system and human-computer interaction have a wide range of applications.The research of recognition of face has had decades
History has been acquired a great achievement so far, however existing research work is more still directed to static criteria facial image, and is passed through
The dynamic human face Study of recognition of actual scene is less.Recognition of face is mainly known by video flowing to obtain image information
Not, but this method is often bad to distinguish dynamic object and static object, causes certain puzzlement to identification.
Summary of the invention
It is an object of the invention to propose a kind of method for detecting video static object for the defects of background technique,
Dynamic object and static object are distinguished by detecting the target position deviation of before and after frames.
To achieve this purpose, the present invention adopts the following technical scheme:
A method of detection video static object, specific step is as follows for detection video static object:
Step A: camera obtains video streaming image, identifies video streaming image with the presence or absence of detected by third party's engine
Target, if then entering step B;
Step B: when detecting target, the two frame single frames picture of front and back where target is obtained;
Step C: in two frame single frames picture of front and back, being respectively divided out the marquee of target region information, preceding
Marquee afterwards in two frame single frames pictures is completely the same;
Step D: being completely coincident the target in two frame single frames picture of front and back, and judges two frame single frames picture figure of front and back
Marquee as in whether there is deviation, if then calculating deviation;
Step E: setting threshold value compares threshold value and deviation, judges whether target is static object according to comparison result.
Preferably, when third party's engine detects target, the single frames picture where target is obtained, as the first figure
Picture;
In video streaming image, using the first image as start frame image, a later frame of start frame image is obtained
Single frames picture where target, as the second image.
Preferably, it in the first image, determines target's center and obtains the first minimum of target according to target's center
Wrap up rectangle;
In second image, determines target's center and obtain the second minimum package square of target according to target's center
Shape.
Preferably, target region is obtained with the centre coordinate of the first minimum package rectangle pre-determined distance value that extends outwardly
First marquee of information;
Target region information is obtained with the centre coordinate of the second minimum package rectangle pre-determined distance value that extends outwardly
Second marquee;
The outwardly extending pre-determined distance value of centre coordinate of first minimum package rectangle and the second minimum package rectangle is identical.
Preferably, the target in two frame single frames picture of front and back is completely coincident include:
It is completely coincident the first minimum package rectangle and the center of the second minimum package rectangle;
Four apex coordinates for obtaining the first marquee and the second marquee after being overlapped respectively, successively compare four vertex
Coordinate judges whether grid deviation occur, if there is grid deviation, then further calculates the first marquee and the second marquee
Each apex coordinate deviation, deviation is compared with threshold value, when deviation be greater than threshold value when, then judge that target is
Dynamic object;When deviation is less than threshold value, then judge target for static object.
Preferably, including one coordinates computed deviation of formula is used;
Formula one:
Wherein:
(one.x1, one.y1)、(one.x2, one.y2)、(one.x3, one.y3)、(one.x4, one.y4) respectively indicate
The coordinate on first four vertex of marquee;
(two.x1', two.y1′)、(two.x1', two.y1′)、(two.x1', two.y1′)、(two.x1', two.y1′)
Respectively indicate the coordinate on second four vertex of marquee;
one.x1″、one.x2″、one.x3″、one.x4" respectively indicate the x-axis coordinate of the first marquee and the second marquee
Deviation;
one.y1″、one.y2″、one.y3″、one.y4" respectively indicate the y-axis coordinate of the first marquee and the second marquee
Deviation.
Detailed description of the invention
Fig. 1 is the flow chart of detection video static object of the invention.
Specific embodiment
To further illustrate the technical scheme of the present invention below with reference to the accompanying drawings and specific embodiments.
When carrying out recognition of face, face of the meeting needed for identifying us in captured video, but some faces
Presence, be not the information needed for us, for example face head portrait on billboard etc. is desirable for this type objects
It rejects, the general character of this type objects is to belong to static object, it is therefore desirable to by judging the static object in video flowing come perfect
Recognition of face.
A kind of method of the detection video static object of the present embodiment, as shown in Figure 1, the specific steps are as follows:
Step A: camera obtains video streaming image, identifies video streaming image with the presence or absence of detected by third party's engine
Target, if then entering step B;
In actual production, third party's engine is the prior art, is mainly used for identifying whether there is in video streaming image
Face, if judging the object then there are face to be detected target;
When third party's engine detects target, the single frames picture where target is obtained, as the first image;
In video streaming image, using the first image as start frame image, a later frame of start frame image is obtained
Single frames picture where target, as the second image.
Acquired single frames picture its actually picture rather than video flowing once detecting target intercept video at once
Front and back two field pictures in stream picture, and saved in the form of picture.
Step B: when detecting target, the two frame single frames picture of front and back where target is obtained;
Step C: in two frame single frames picture of front and back, being respectively divided out the marquee of target region information, preceding
Marquee afterwards in two frame single frames pictures is completely the same;
In the first image, determines target's center and obtain the first minimum package square of target according to target's center
Shape;
In second image, determines target's center and obtain the second minimum package square of target according to target's center
Shape.
Target region information is obtained with the centre coordinate of the first minimum package rectangle pre-determined distance value that extends outwardly
First marquee;
Target region information is obtained with the centre coordinate of the second minimum package rectangle pre-determined distance value that extends outwardly
Second marquee;
The outwardly extending pre-determined distance value of centre coordinate of first minimum package rectangle and the second minimum package rectangle is identical.
Since the principle of the present invention is by comparing in front of and after frames image, the alternate position spike of target changes to judge whether target produces
Raw movement, according to judging result and then judges whether target is static object, therefore when judging whether target generates mobile, first
The minimum package rectangle for wrapping target is first obtained out, a choosing is generated with the outwardly extending pre-determined distance value of minimum package rectangle
Frame is taken, frame selects the information in target area, compares the coordinate of two marquees.
Step D: being completely coincident the target in two frame single frames picture of front and back, and judges two frame single frames picture figure of front and back
Marquee as in whether there is deviation, if then calculating deviation;
Step E: setting threshold value compares threshold value and deviation, judges whether target is static object according to comparison result.
The target in two frame single frames picture of front and back is completely coincident include:
It is completely coincident the first minimum package rectangle and the center of the second minimum package rectangle;
Four apex coordinates for obtaining the first marquee and the second marquee after being overlapped respectively, successively compare four vertex
Coordinate judges whether grid deviation occur, if there is grid deviation, then further calculates the first marquee and the second marquee
Each apex coordinate deviation, deviation is compared with threshold value, when deviation be greater than threshold value when, then judge that target is
Dynamic object;When deviation is less than threshold value, then judge target for static object.
Set threshold value is generally 5 units, and when grid deviation is greater than 5 units, the target in the before and after frames is
Attach most importance to complicated target, that is, static object.
Including using one coordinates computed deviation of formula;
Formula one:
Wherein:
(one.x1, one.y1)、(one.x2, one.y2)、(one.x3, one.y3)、(one.x4, one.y4) respectively indicate
The coordinate on first four vertex of marquee;
(two.x1', two.y1′)、(two.x1', two.y1′)、(two.x1', two.y1′)、(two.x1', two.y1′)
Respectively indicate the coordinate on second four vertex of marquee;
one.x1″、one.x2″、one.x3″、one.x4" respectively indicate the x-axis coordinate of the first marquee and the second marquee
Deviation;
one.y1″、one.y′″、one.y3″、one.y4" respectively indicate the y-axis coordinate of the first marquee and the second marquee
Deviation.
The technical principle of the invention is described above in combination with a specific embodiment.These descriptions are intended merely to explain of the invention
Principle, and shall not be construed in any way as a limitation of the scope of protection of the invention.Based on the explanation herein, the technology of this field
Personnel can associate with other specific embodiments of the invention without creative labor, these modes are fallen within
Within protection scope of the present invention.
Claims (6)
1. a kind of method for detecting video static object, it is characterised in that: specific step is as follows for detection video static object:
Step A: camera obtains video streaming image, identifies that video streaming image whether there is by third party's engine and is detected mesh
Mark, if then entering step B;
Step B: when detecting target, the two frame single frames picture of front and back where target is obtained;
Step C: in two frame single frames picture of front and back, the marquee of target region information, front and back two is respectively divided out
Marquee in frame single frames picture is completely the same;
Step D: being completely coincident the target in two frame single frames picture of front and back, and judges in two frame single frames picture of front and back
Marquee whether there is deviation, if then calculating deviation;
Step E: setting threshold value compares threshold value and deviation, judges whether target is static object according to comparison result.
2. a kind of method for detecting video static object according to claim 1, it is characterised in that:
When third party's engine detects target, the single frames picture where target is obtained, as the first image;
In video streaming image, using the first image as start frame image, the target of a later frame of start frame image is obtained
The single frames picture at place, as the second image.
3. a kind of method for detecting video static object according to claim 2, it is characterised in that:
In the first image, determines target's center and obtain the first minimum package rectangle of target according to target's center;
In second image, determines target's center and obtain the second minimum package rectangle of target according to target's center.
4. a kind of method for detecting video static object according to claim 3, it is characterised in that:
The first of target region information is obtained with the centre coordinate of the first minimum package rectangle pre-determined distance value that extends outwardly
Marquee;
The second of target region information is obtained with the centre coordinate of the second minimum package rectangle pre-determined distance value that extends outwardly
Marquee;
The outwardly extending pre-determined distance value of centre coordinate of first minimum package rectangle and the second minimum package rectangle is identical.
5. a kind of method for detecting video static object according to claim 4, it is characterised in that:
The target in two frame single frames picture of front and back is completely coincident include:
It is completely coincident the first minimum package rectangle and the center of the second minimum package rectangle;
Four apex coordinates for obtaining the first marquee and the second marquee after being overlapped respectively successively compare four vertex and sit
Mark, judges whether grid deviation occur, if there is grid deviation, then further calculates the first marquee and the second marquee
Deviation is compared by the deviation of each apex coordinate with threshold value, when deviation is greater than threshold value, then judges that target is
State target;When deviation is less than threshold value, then judge target for static object.
6. a kind of method for detecting video static object according to claim 5, it is characterised in that:
Including using one coordinates computed deviation of formula;
Formula one:
Wherein:
(one.x1, one.y1)、(one.x2, one.y2)、(one.x3, one.y3)、(one.x4, one.y4) respectively indicate first
The coordinate on four vertex of marquee;
(two.x1', two.y1′)、(two.x1', two.y1′)、(two.x1', two.y1′)、(two.x1', two.y1') respectively
Indicate the coordinate on second four vertex of marquee;
one.x1″、one.x2″、one.x3″、one.x4" respectively indicate the x-axis coordinate deviation of the first marquee and the second marquee
Value;
one.y1″、one.y2″、one.y3″、one.y4" respectively indicate the y-axis coordinate deviation of the first marquee and the second marquee
Value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910632936.9A CN110348394A (en) | 2019-07-15 | 2019-07-15 | A method of detection video static object |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910632936.9A CN110348394A (en) | 2019-07-15 | 2019-07-15 | A method of detection video static object |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110348394A true CN110348394A (en) | 2019-10-18 |
Family
ID=68176159
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910632936.9A Pending CN110348394A (en) | 2019-07-15 | 2019-07-15 | A method of detection video static object |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110348394A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112800929A (en) * | 2021-01-25 | 2021-05-14 | 安徽农业大学 | On-line monitoring method for bamboo shoot quantity and high growth rate based on deep learning |
CN113554008A (en) * | 2021-09-18 | 2021-10-26 | 深圳市安软慧视科技有限公司 | Method and device for detecting static object in area, electronic equipment and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103745485A (en) * | 2013-12-31 | 2014-04-23 | 深圳泰山在线科技有限公司 | Method and system for judging object stillness or movement |
CN107563345A (en) * | 2017-09-19 | 2018-01-09 | 桂林安维科技有限公司 | A kind of human body behavior analysis method based on time and space significance region detection |
CN108875480A (en) * | 2017-08-15 | 2018-11-23 | 北京旷视科技有限公司 | A kind of method for tracing of face characteristic information, apparatus and system |
CN109118516A (en) * | 2018-07-13 | 2019-01-01 | 高新兴科技集团股份有限公司 | A kind of target is from moving to static tracking and device |
CN109840565A (en) * | 2019-01-31 | 2019-06-04 | 成都大学 | A kind of blink detection method based on eye contour feature point aspect ratio |
-
2019
- 2019-07-15 CN CN201910632936.9A patent/CN110348394A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103745485A (en) * | 2013-12-31 | 2014-04-23 | 深圳泰山在线科技有限公司 | Method and system for judging object stillness or movement |
CN108875480A (en) * | 2017-08-15 | 2018-11-23 | 北京旷视科技有限公司 | A kind of method for tracing of face characteristic information, apparatus and system |
CN107563345A (en) * | 2017-09-19 | 2018-01-09 | 桂林安维科技有限公司 | A kind of human body behavior analysis method based on time and space significance region detection |
CN109118516A (en) * | 2018-07-13 | 2019-01-01 | 高新兴科技集团股份有限公司 | A kind of target is from moving to static tracking and device |
CN109840565A (en) * | 2019-01-31 | 2019-06-04 | 成都大学 | A kind of blink detection method based on eye contour feature point aspect ratio |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112800929A (en) * | 2021-01-25 | 2021-05-14 | 安徽农业大学 | On-line monitoring method for bamboo shoot quantity and high growth rate based on deep learning |
CN113554008A (en) * | 2021-09-18 | 2021-10-26 | 深圳市安软慧视科技有限公司 | Method and device for detecting static object in area, electronic equipment and storage medium |
CN113554008B (en) * | 2021-09-18 | 2021-12-31 | 深圳市安软慧视科技有限公司 | Method and device for detecting static object in area, electronic equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20210366124A1 (en) | Graphical fiducial marker identification | |
WO2014155979A1 (en) | Tracking processing device and tracking processing system provided with same, and tracking processing method | |
WO2016157327A1 (en) | Video surveillance system and video surveillance method | |
US11948363B2 (en) | Person detection device and person detection method | |
Ribnick et al. | Real-time detection of camera tampering | |
EP2966591B1 (en) | Method and apparatus for identifying salient events by analyzing salient video segments identified by sensor information | |
CN110009650B (en) | Escalator handrail boundary area border crossing detection method and system | |
CN110348394A (en) | A method of detection video static object | |
JP2020149111A (en) | Object tracking device and object tracking method | |
JP7255173B2 (en) | Human detection device and human detection method | |
US20180307896A1 (en) | Facial detection device, facial detection system provided with same, and facial detection method | |
CN113989858B (en) | Work clothes identification method and system | |
JP2008035301A (en) | Mobile body tracing apparatus | |
JP7188067B2 (en) | Human detection device and human detection method | |
US10475191B2 (en) | System and method for identification and suppression of time varying background objects | |
KR101206736B1 (en) | Apparatus and method for detecting rotated face region | |
WO2022091577A1 (en) | Information processing device and information processing method | |
WO2020179638A1 (en) | Person detection device and person detection method | |
Chen et al. | Detection and segmentation of occluded vehicles based on skeleton features | |
US20230103555A1 (en) | Information processing apparatus, information processing method, and program | |
JP3078203B2 (en) | Intruding object recognition device | |
KR101567366B1 (en) | Masking apparatus and method for detecting face | |
JP7524713B2 (en) | Information processing device and information processing method | |
JP2011259044A (en) | Image processing device and image processing method | |
Xu et al. | 2D and 3D video scene text classification |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191018 |