CN104637054B - Image true-false mirror method for distinguishing based on imaging Perspective Principles - Google Patents
Image true-false mirror method for distinguishing based on imaging Perspective Principles Download PDFInfo
- Publication number
- CN104637054B CN104637054B CN201510048232.9A CN201510048232A CN104637054B CN 104637054 B CN104637054 B CN 104637054B CN 201510048232 A CN201510048232 A CN 201510048232A CN 104637054 B CN104637054 B CN 104637054B
- Authority
- CN
- China
- Prior art keywords
- face
- target
- length
- image
- point
- 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.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- 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
- G06T2207/20068—Projection on vertical or horizontal image axis
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The image true-false mirror method for distinguishing based on imaging Perspective Principles of the present invention, including：A) imaging optical axises is definite；B) determines front end face；C) determines object distance；D) determines the end face where target to be identified；E) asks for the distance between both ends of the surface；F) horizontals is definite；G) marks the level point of object to be identified；H) asks for mark pointWithLength in front end face；I) asks for horizontal line sectionThe physical length characterized；J) asks for the length corresponding to the mark line segment of vertical direction；K) discriminating of image true-falses.The image true-false mirror method for distinguishing of the present invention, ask for the actual vertical length corresponding to the mark line segment of object vertical direction to be identified, by the way that the level being obtained, vertical length are compared with entity, it can determine whether out to identify whether target is to be synthesized in background picture, realize that the true and false of digital picture judges.
Description
Technical field
The present invention relates to a kind of image true-false mirror method for distinguishing based on imaging Perspective Principles, more specifically, especially relating to
It is and a kind of by comparing the identification target size size and the actual size size of the target that are reflected in picture or scope
Image true-false mirror method for distinguishing based on imaging Perspective Principles.
Background technology
Video or picture usually contain larger data message, can be as the reasonable evidence in judicial process, to prove
The content to be proved.However, with the progress and development of video and image processing techniques, more and more videos or image exist
The possibility artificially distorted, and can not be reflected and determined whether it is true or false by conventional detection method, it is that image is proposed as judicial evidence
Stern challenge." South China Tiger event " occurred such as 2007, relates to thing picture, in October, 2007 as shown in figure 4, giving
Part, Shaanxi Department of Forestry of Shanxi Province announces to find South China tiger in Shaanxi, and discloses and be stated to be Ankang city Zhenping county town pass town villager
The South China tiger photo that straight imperial 3 days October in 2007 takes, even world's authority's magazine《Science》It also can not be to the true of photo
Reality provides conclusion, it is seen that the technology of distorting of existing digital photos has had reached the degree mixed the spurious with the genuine.
Perspective imaging is also referred to as perspective view, it is that a kind of shape three-dimensional object shows on two dimensional surface
Stereogram, although digital picture shows the object in three dimensions in the plane, but if specific clap can be combined
Scene is taken the photograph, then can calculate the size of the object occurred in frame out, utilizes object and the size constancy of object to be identified
It can determine that the true and false of image.In image when distorting, the personnel that distort simply focus on how replicating target, pasting, being embedded in figure
It as in, and can not judge by existing digital image processing techniques the trace of amending image, often have ignored target in the background
Specific size, the photo of Anhui Ningguo Bureau of Civil Affairs chief et al. the expression of sympathy centenarian such as occurred for 2013, such as Fig. 3 institutes
Show, give the digital picture, just because of size of the target in specific size when picture is distorted is not accounted for, cause often
People can be seen that photo is that PS is crossed, and cause public opinion in commotion.But the picture in South China Tiger event can not with the naked eye find out it
True or false, this just needs specifically to calculate the size that target is reflected in picture, by the way that target is reflected in picture
Size sees whether its size rationally judges the true or false of size compared with the actual size or actual reference dimension.
The content of the invention
The shortcomings that present invention is in order to overcome above-mentioned technical problem provides a kind of image true-false based on imaging Perspective Principles
Reflect method for distinguishing.
The image true-false mirror method for distinguishing based on imaging Perspective Principles of the present invention, feature part is, by following
Step is realized：A) imaging optical axises determine, the object in imaging circumstances according to corresponding to the central point of digital picture,
Determine the optical axis of imaging device during captured imagePosition and direction；B) determines front end face, according to digital picture four
The object in imaging circumstances corresponding to the midpoint on a side, is determined and imaging optical axisPerpendicular end face, this end face are positioned as
Front end face；C) determines object distance, according to focusPosition on optical axis, determines front end faceDistance focal pointAway from
From this distance definition is object distance；D) determines the end face where target to be identified, according to can not be moved around target to be identified
Dynamic object, determines the end face perpendicular with imaging optical axis where target to be identified, this end face is defined as rear end face；
E) asks for the distance between both ends of the surface, according to front end faceAnd rear end facePosition in actual photographed scene determines
Go out the distance between both ends of the surface；F) horizontals is definite, in front end faceMiddle searching position and the immutable object of size
Body, and demarcate 2 object points in its horizontal directionWith, form horizontal line section, and object point is found out in digital pictureWithCorresponding picture pointWith；G) marks the level point of object to be identified, and object to be identified is marked in digital picture
Two mark points in body horizontal directionWith, form horizontal line section；H) asks for mark pointWithLength in front end face
Degree, it is assumed that two mark points in image in same horizontal lineWithIt is by front end faceIn object pointWithIt sends,
According to image-forming principle, pass through formula（1）Horizontal line section existing for hypothesis is obtainedLength：
（1）
I) asks for horizontal line sectionThe physical length characterized, the property possessed according to similar triangles, passes through public affairs
Formula（2）The horizontal line section in target to be identified in digital picture is obtainedCorresponding real standard length：
（2）
J) ask for vertical direction mark line segment corresponding to length, using with step f) to i) identical method, calculate
Go out the actual vertical length corresponding to the mark line segment of vertical direction in target to be identified；K) discriminating of image true-falses, according to step
The real standard length and vertical length corresponding to mark point in rapid target to be identified acquired i) and in step j), will ask
The physical length taken is compared with target entity to be identified or such identification target size range, if difference has been more than by mistake
The permitted scope of difference then shows that identification target is synthesized in background picture, and digital picture is to forge picture；If difference
It is without tampered image in the case of then showing that digital picture identifies target as object is differentiated using this in rational scope.
The image true-false based on imaging Perspective Principles of the present invention reflects method for distinguishing, acquired object distance in step c), step
It is rapid e) measured by distance, in step f) horizontal line section length, horizontal line section length in step h), step i)
In real standard lengthIt is accurate to centimetre；Picture point in step f)WithThe distance between、In step g)
Picture pointWithThe distance betweenIt is calculated using the actual range in number of pixels or image, if acquisition image
On actual range then exact value millimeter.
The beneficial effects of the invention are as follows：The image true-false mirror method for distinguishing based on imaging Perspective Principles of the present invention, for
The present situation of existing digital picture detection means None- identified distorted image means ripe day by day, using object size to be identified not
The principle of denaturation establishes imaging optical axis and focus according to digital picture in actual photographed scene first, then hangs down with optical axis
Straight front end face internal labeling line segment, mark corresponding picture point in the pictureWith, finally it is obtained according to image-forming principle and waits to reflect
Determine horizontal line section in the image of objectPhysical length；Object vertical direction to be identified is asked for according to identical method
Mark line segment corresponding to actual vertical length, by the way that the level being obtained, vertical length are compared with entity, you can judge
Go out to identify whether target is to be synthesized in background picture, and then judge the true and false of digital picture.
Description of the drawings
Fig. 1 is the schematic diagram of perspective imaging；
Fig. 2 is the original for asking for the physical length on image corresponding to horizontal line section in the present invention using the Perspective Principles of imaging
Reason figure；
Fig. 3 is the schematic diagram of the variable synthesising picture of naked eyes；
Fig. 4 is to relate to thing picture in South China Tiger event；
Fig. 5 is the flow chart of the image true-false mirror method for distinguishing based on imaging Perspective Principles of the present invention.
Specific embodiment
The invention will be further described with embodiment below in conjunction with the accompanying drawings.
As shown in Figure 1, giving the perspective imaging principle figure of photographing device, shown horizontal plane is, on horizontal plane
Object is,For optical axis,For projection centre（Focus）,The rear plane vertical with optical axis be perspective plane, objectOn
Object pointWith projection centreLineIt is met at after extension on perspective planePoint,It is exactly object pointPerspective projection.This is just
It is the Perspective Principles described in us.
As shown in figure 5, the flow chart of the image true-false mirror method for distinguishing based on imaging Perspective Principles of the present invention is given,
It is realized by following steps：
A) imaging optical axises determine, the object in imaging circumstances according to corresponding to the central point of digital picture, determine
The optical axis of imaging device when going out captured imagePosition and direction；
B) determines front end face, the object in imaging circumstances according to corresponding to the midpoint on four sides of digital picture, really
It makes and imaging optical axisPerpendicular end face, this end face are positioned as front end face；
As shown in Figure 1,Point andPoint is respectively the central point, laterally following on vertical the right of object in imaging circumstances
Central point can determine that intersection point by it, and then can determine that object distance；In the photo such as provided in Fig. 4, front end face can be
End face in picture where the trunk set greatly on right side.
C) determines object distance, according to focusPosition on optical axis, determines front end faceDistance focal pointDistance,
This distance definition is object distance；
D) determines the end face where target to be identified, according to mobile object can not occur around target to be identified, determines
Go out the end face perpendicular with imaging optical axis where target to be identified, this end face is defined as rear end face；
In Fig. 4, if in picture South China tiger head feature（Such as the size of brave head）As identification target, then can choose
Vertical plane in actual photographed scene where the trunk of the little tree of South China tiger head is rear end face.
E) asks for the distance between both ends of the surface, according to front end faceAnd rear end facePosition in actual photographed scene
It puts, determines the distance between both ends of the surface；
Such as in Fig. 4, distanceBetween vertical plane where as big trunk and the vertical plane where the small trunk of rear end
Actual range.
F) horizontals is definite, in front end faceMiddle searching position and the immutable object of size, and demarcate it
2 object points in horizontal directionWith, form horizontal line section, and object point is found out in digital pictureWithIt is corresponding
Picture pointWith；
As in Fig. 4,2 points of works in the big tree trunk horizontal direction in right side in Fig. 4 can be chosen in actual photographed scene
For object pointWith, and corresponding picture point is marked in digital picture.
G) marks the level point of object to be identified, and two in object horizontal direction to be identified are marked in digital picture
Mark pointWith, form horizontal line section；
As shown in figure 4, two point conducts on South China tiger face width are chosen in digital pictureWith, with south China
Face's width of tiger is as diagnostic characteristics.
H) asks for mark pointWithLength in front end face, it is assumed that two marks in image in same horizontal line
PointWithIt is by front end faceIn object pointWithIt sends, according to image-forming principle, passes through formula（1）Hypothesis is obtained to exist
Horizontal line sectionLength：
（1）
I) asks for horizontal line sectionThe physical length characterized, the property possessed according to similar triangles, passes through public affairs
Formula（2）The horizontal line section in target to be identified in digital picture is obtainedCorresponding real standard length：
（2）
For the South China tiger in picture 4, it is equivalent to be obtained face's width of South China tiger.
J) ask for vertical direction mark line segment corresponding to length, using with step f) to i) identical method, calculate
Go out the actual vertical length corresponding to the mark line segment of vertical direction in target to be identified；
As in Fig. 4, the physical length of South China tiger face can be asked for, face's length by the use of South China tiger is special as differentiating
Sign.
K) discriminating of image true-falses, according to the mark point institute in target to be identified acquired in step i) and step j)
Corresponding real standard length and vertical length, by the physical length asked for and target entity to be identified or such identification target
Size range is compared, if difference has been more than the permitted scope of error, shows to identify that target is to be synthesized to Background
In piece, digital picture is to forge picture；If difference in rational scope, show digital picture using this identify target as
It is without tampered image in the case of differentiating object.
For the South China tiger in Fig. 4, then by judging the South China tiger in calculated digital picture in actual photographed
The width of face, length in scene, to judge the South China tiger face size being obtained whether within the scope of rational, if it exceeds
Rational scope, then it is believed that South China tiger is be embedded in picture in the later stage, if within the scope of rational, surface according to
Identify that this image is not tampered image according to the face feature of South China tiger.
In order to ensure the accuracy of striked data, acquired object distance in step c), distance measured by step e)、
The length of horizontal line section in step f), horizontal line section length in step h), real standard length in step i)It is accurate to centimetre；Picture point in step f)WithThe distance between、Picture point in step g)WithBetween
DistanceIt is calculated using the actual range in number of pixels or image, it is accurate if the actual range on acquisition image
It is worth millimeter.
For the identification target " South China tiger " in Fig. 4, since target not reproduce, the size model of target can only be provided
It encloses so that identification is more difficult.If target is that reproduce object to identifying in picture, such as people, vehicle, article, then with regard to pole
The big feasibility and confidence level for improving the discriminating of the picture true and false.
Claims (2)
- The method for distinguishing 1. a kind of image true-false based on imaging Perspective Principles reflects, which is characterized in that realized by following steps：A) imaging optical axises determine, the object in imaging circumstances according to corresponding to the central point of digital picture, determine to clap The optical axis of imaging device when taking the photograph imagePosition and direction；B) determines front end face, the object in imaging circumstances according to corresponding to the midpoint on four sides of digital picture determines With imaging optical axisPerpendicular end face, this end face are positioned as front end face；C) determines object distance, according to focusPosition on optical axis, determines front end faceDistance focal pointDistance, this away from From being defined as object distance；D) determines the end face where target to be identified, according to mobile object can not occur around target to be identified, determines to treat Identify the end face perpendicular with imaging optical axis where target, this end face is defined as rear end face；E) asks for the distance between both ends of the surface, according to front end faceAnd rear end facePosition in actual photographed scene, really Make the distance between both ends of the surface；F) horizontals is definite, in front end faceMiddle searching position and the immutable object of size, and it is horizontal to demarcate it 2 object points on directionWith, form horizontal line section, and object point is found out in digital pictureWithCorresponding picture PointWith；G) marks the level point of object to be identified, and two marks in object horizontal direction to be identified are marked in digital picture PointWith, form horizontal line section；H) asks for mark pointWithLength in front end face, it is assumed that two mark points in image in same horizontal line WithIt is by front end faceIn object pointWithIt sends, according to image-forming principle, passes through formula（1）Water existing for hypothesis is obtained Horizontal line sectionLength：（1）I) asks for horizontal line sectionThe physical length characterized, the property possessed according to similar triangles, passes through formula（2） The horizontal line section in target to be identified in digital picture is obtainedCorresponding real standard length：（2）J) asks for the length corresponding to the mark line segment of vertical direction, to i) identical method, calculates and treats using with step f) Identify the actual vertical length corresponding to the mark line segment of vertical direction in target；K) discriminating of image true-falses, according to corresponding to the mark point in target to be identified acquired in step i) and step j) Real standard length and vertical length, by the physical length asked for and target entity to be identified or such identification target size Scope is compared, if difference has been more than the permitted scope of error, shows to identify that target is synthesized in background picture , digital picture is to forge picture；If difference in rational scope, shows that digital picture identifies target as differentiation using this It is without tampered image in the case of object.
- The method for distinguishing 2. image true-false according to claim 1 based on imaging Perspective Principles reflects, it is characterised in that：Step C) acquired object distance in, distance measured by step e), in step f) horizontal line section length, water in step h) Horizontal line segment length, real standard length in step i)It is accurate to centimetre；Picture point in step f)WithBetween Distance、Picture point in step g)WithThe distance betweenUsing the actual range in number of pixels or image It calculates, millimeter is accurate to if the actual range on acquisition image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510048232.9A CN104637054B (en) | 2015-01-30 | 2015-01-30 | Image true-false mirror method for distinguishing based on imaging Perspective Principles |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510048232.9A CN104637054B (en) | 2015-01-30 | 2015-01-30 | Image true-false mirror method for distinguishing based on imaging Perspective Principles |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104637054A CN104637054A (en) | 2015-05-20 |
| CN104637054B true CN104637054B (en) | 2018-05-18 |
Family
ID=53215760
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510048232.9A Expired - Fee Related CN104637054B (en) | 2015-01-30 | 2015-01-30 | Image true-false mirror method for distinguishing based on imaging Perspective Principles |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104637054B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102609947A (en) * | 2012-02-10 | 2012-07-25 | 浙江理工大学 | Forgery detection method for spliced and distorted digital photos |
| CN102609948A (en) * | 2012-02-10 | 2012-07-25 | 浙江理工大学 | Manipulation detection method for copy-paste distorted photo digital photos |
-
2015
- 2015-01-30 CN CN201510048232.9A patent/CN104637054B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102609947A (en) * | 2012-02-10 | 2012-07-25 | 浙江理工大学 | Forgery detection method for spliced and distorted digital photos |
| CN102609948A (en) * | 2012-02-10 | 2012-07-25 | 浙江理工大学 | Manipulation detection method for copy-paste distorted photo digital photos |
Non-Patent Citations (5)
| Title |
|---|
| A New Approach to Obtain Height Measurements from Video;Antonio Criminisi etal;《PROCEEDINGS OF SPIE》;19990204;第3576卷;第227-237 * |
| 数字图像被动取证技术综述;赵俊红;《计算机工程与设计》;20111231;第32卷(第8期);第2756-2759页 * |
| 行进视觉中环境对象位置估计;李靖;《中国优秀硕士学位论文全文数据库信息科技辑》;20111231(第12期);第I138-763页 * |
| 视频监控电子人像证据鉴定研究;蔡鑫;《中国优秀硕士学位论文全文数据库社会科学I辑》;20130715(第7期);第G113-76页 * |
| 透视投影变换矩阵及其方程;崔永军;《北京动力经济学院学报》;19940930;第11卷(第3期);第92-95页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104637054A (en) | 2015-05-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110285793B (en) | Intelligent vehicle track measuring method based on binocular stereo vision system | |
| WO2021004548A1 (en) | Vehicle speed intelligent measurement method based on binocular stereo vision system | |
| CN105678742B (en) | A kind of underwater camera scaling method | |
| CN103714535B (en) | Binocular vision measurement system camera parameter online adjustment method | |
| CN107084680B (en) | A kind of target depth measurement method based on machine monocular vision | |
| CN111680588A (en) | Human face gate living body detection method based on visible light and infrared light | |
| CN108288292A (en) | A kind of three-dimensional rebuilding method, device and equipment | |
| CN100428805C (en) | Video camera reference method only using plane reference object image | |
| CN108020175B (en) | multi-grating projection binocular vision tongue surface three-dimensional integral imaging method | |
| CN109859272A (en) | A kind of auto-focusing binocular camera scaling method and device | |
| CN106570899B (en) | Target object detection method and device | |
| WO2011031538A2 (en) | Accurate 3d object reconstruction using a handheld device with a projected light pattern | |
| CN106296811A (en) | A kind of object three-dimensional reconstruction method based on single light-field camera | |
| CN105913013A (en) | Binocular vision face recognition algorithm | |
| CN110110793B (en) | Binocular image rapid target detection method based on double-current convolutional neural network | |
| JP7479729B2 (en) | Three-dimensional representation method and device | |
| WO2018032841A1 (en) | Method, device and system for drawing three-dimensional image | |
| CN104200476B (en) | The method that camera intrinsic parameter is solved using the circular motion in bimirror device | |
| TWI394085B (en) | Method of identifying the dimension of a shot subject | |
| WO2024032233A1 (en) | Stereophotogrammetric method based on binocular vision | |
| CN112465778A (en) | Underwater fish shoal observation device and method | |
| CN107865473B (en) | Characteristics of human body's range unit and its distance measuring method | |
| CN110378995A (en) | A method of three-dimensional space modeling is carried out using projection feature | |
| TW202242803A (en) | Positioning method and apparatus, electronic device and storage medium | |
| CN105354828B (en) | Read and write intelligent identification and the application thereof of reading matter three-dimensional coordinate in scene |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180518 Termination date: 20190130 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |