CN1097959C - Imaging system - Google Patents
Imaging system Download PDFInfo
- Publication number
- CN1097959C CN1097959C CN96180544A CN96180544A CN1097959C CN 1097959 C CN1097959 C CN 1097959C CN 96180544 A CN96180544 A CN 96180544A CN 96180544 A CN96180544 A CN 96180544A CN 1097959 C CN1097959 C CN 1097959C
- Authority
- CN
- China
- Prior art keywords
- image
- reflecting surface
- color
- indirect
- image portion
- 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
Images
Landscapes
- Image Input (AREA)
- Color Television Image Signal Generators (AREA)
Abstract
The present invention provides a practical shooting system which corrects the color of the image of a shot object at a high speed and accurately displays the color of the image. The present invention comprises a shooting element (23) and a lens (21), wherein the shooting element (23) shoots a color image, and the image of an object (0b) is set in the maximum visual field (Vm, Vm) so as to be scattered and reflected through the lens (21) for the incidence onto the reflecting surface (18) on the shooting element (23). The principal coordinates (Xmi, Ymi) of a direct image which is obtained through imaging an object point (P) of the object (0b) on the shooting element (23) are corresponding to the subsidiary coordinates (Xni, Yni) of an indirect image of the object point (P) through the reflecting surface (18) on the shooting element (23). The R, G, B components of pixels in all the principal coordinates are divided by the same components of the subsidiary coordinates so as to realize shooting with color correction.
Description
Technical field
The present invention relates to the object image that a kind of revisal photographs color, and can correctly show the camera chain of this image color.
Background technology
The color of object is subjected to the influence of illuminating position easily, wants often correctly to show that the color of the image of being photographed by video camera is difficult.Even the people under such environment, also can correctly be familiar with the color of object, this ability is called as color invariance.
In existing video camera, also there is not imaging apparatus (for example, imaging apparatus such as CCD element) with this specific character.In order in comprising the camera system of video camera, to realize color invariance, attempt perhaps carrying out complicated revisal by the color of specified point and color are on every side relatively carried out revisal.But these trials or only limit to the revisal of special image perhaps need too much image processing time, and impracticable.
Summary of the invention
The object of the present invention is to provide a kind of color of the object image that can be photographed with the speed revisal of practicality, and can correctly show this image color, the good camera chain of response.
Feature of the present invention is, at the imaging apparatus that comprises the chromatic image of making a video recording with the picture of object is imaged in the camera chain of camera lens of above-mentioned imaging apparatus, in the absolute visual field that constitutes by above-mentioned camera lens and imaging apparatus, be provided with the astigmatism of above-mentioned object is penetrated reflection and incided reflecting surface on the above-mentioned imaging apparatus by above-mentioned camera lens, be provided with each principal coordinate (Xmi of the object-point direct image that imaging obtains on above-mentioned imaging apparatus on the above-mentioned object, Ymi) and each secondary coordinate (Xni of the indirect image of the above-mentioned object-point that on above-mentioned imaging apparatus, obtains by above-mentioned reflecting surface, Yni) corresponding corresponding intrument has the color correcting section that obtains the shooting of color revisal by following formula.
D1(Xmi，Ymi)＝(Rmi/Rni)·S，
D2(Xmi，Ymi)＝(Gmi/Gni)·S，
D3(Xmi，Ymi)＝(Bmi/Bni)·S，
In the formula, D1, D2, D3 are respectively the principal coordinate (Xmi of the shooting after the color revisal, Ymi) R, G, B composition, Rmi, Gmi, Bmi are respectively principal coordinate (Xmi, Ymi) R, G, the B composition of direct pixel (Pm) in, Rni, Gni, Bni are respectively secondary coordinate, and (S is the revisal item for Xni, Yni) R, the G of indirect pixel (Pn), B composition.
According to inventor's described later parsing, distinguished the indirect image representative object point brightness that is arranged on the reflecting surface institute scattered reflection in the absolute visual field.Therefore, remove Rmi, Gmi, Bmi respectively, can offset because the error that the influence of illumination causes with the Rni that represents this brightness, Gni, Bni.This point is by inventor's experimental verification.Again, revisal item S is in order to prevent that to exceed the boundary of amplitude-scaled of device caused saturated when Rmi, Gmi, Bmi are removed later result by Rni, Gni, Bni.
Particularly, make that the amplitude of the direct image portion of the above-mentioned direct image of shooting is wideer than the amplitude of the indirect image portion of taking above-mentioned indirect image in above-mentioned imaging apparatus, can apply flexibly the absolute visual field of imaging apparatus effectively if set above-mentioned reflecting surface.And, as described later,, confirmed can not produce unaccommodated situation to the color revisal even the amplitude of indirect image portion is 25% a degree of absolute visual field.
Wish to be provided with the protection lid that covers above-mentioned absolute visual field light in addition at least in the daylighting portion side of above-mentioned camera lens again.Light beyond the absolute visual field can cause the error of color revisal, utilizes this protection lid can reduce error.
When the design reflectivity face, on the direction side by side of direct image portion of taking above-mentioned direct image and the indirect image portion of taking above-mentioned indirect image, the direct image and the indirect image of above-mentioned object can be similar shapes.In this case, can carry out correct color revisal to the trickle part of flower or small items.
On the other hand, also can the design reflectivity face making in the direct image portion of taking above-mentioned direct image and take on the direction side by side of indirect image portion of above-mentioned indirect image, is certain value with the ratio of above-mentioned direct image portion and above-mentioned indirect image portion corresponding respectively each pixel count.In this case, can simplify the algorithm of color revisal, carry out the color revisal very at high speed.
The shaped design of reflecting surface gets final product by following formula,
Xni＝f(A-tan(2α))/(1+A·tan(2α))，
In the formula, f is the focal length of camera lens, and A is (X/Z), the horizontal direction distance from horizontal datum Ch of X indicated object point P, the vertical range from vertical reference line Cv of Z indicated object point P, α represents reflecting surface 18 and the angle that horizontal line constituted parallel with vertical reference line Cv.
According to inventor's experiment, if being the skin that forms oil membrane from the teeth outwards, above-mentioned reflecting surface constitutes, distinguished to obtain very good color revisal.
When enforcement was of the present invention, as long as the software for realizing above-mentioned corresponding intrument that will be kept in the medium is installed in the general microcomputer, the above-mentioned protection lid that will have above-mentioned reflecting surface simultaneously was installed on the general video camera and gets final product.
Like this,,, can provide a kind of practicality to carry out the color revisal of photographic objects image apace, correctly show the camera chain that this visual color response is good by by reflecting surface contrast indirect image and direct image according to the feature of the invention described above.
Detailed content of the present invention can describe by invention embodiment shown below and embodiment.And the symbol of being charged in the item of claim scope only contrasts with accompanying drawing for convenience, does not limit the invention in the formation shown in the accompanying drawing drawing because of charging to of this symbol.
Description of drawings
Fig. 1 is expressed as the figure that concerns between object Ob, reflecting surface, camera lens and the CCD element of the explanation principle of the invention.
Fig. 2 (a) is for being expressed as the figure of the initial convolution of explanation scattered reflection apperance on reflecting surface, (b) for being expressed as the figure of the secondary convolution of the indirect visual Fn of portion on the CCD element that illustrates that focus departs from.
Fig. 3 is illustrated respectively in (a) Nonlinear Mapping, (b) Linear Mapping, the direct image Im when (c) spotlight throws light on the whole picture plane F and the graph of a relation between the indirect image In.
Fig. 4 is for being expressed as the figure of explanation ratio between each direct pixel set Pm under the situation of (a) Nonlinear Mapping, (b) Linear Mapping and each indirect pixel set Pn respectively.
The curve that concerns between the angle [alpha] of Fig. 5 (a) for the varying level position of indicated object point P and best reflecting surface, (b) curve for concerning between the degree of depth of indicated object point P and the visual field error angle φ.
Fig. 6 (a) is the change curve that the pass between expression illumination intensity and the briliancy ties up to color revisal front and back, (b) ties up to the change curve of color revisal front and back for the pass between expression illumination intensity and the x chromaticity coordinate
Fig. 7 (a) has installed the oblique view of protecting the video camera of lid for expression, and the cross-sectional view of the video camera of protection lid (b) has been installed for expression.
Fig. 8 is the logic diagram of the relevant camera chain of the present invention of expression.
Fig. 9 is equivalent to the figure of Fig. 1 for expression another embodiment of the present invention.
Embodiment
Following with reference to description of drawings embodiment of the present invention.
At first, Fig. 1～5 explanations principle of the present invention.Light path geometry model after example expression shown in Figure 1 is simplified is investigated position relation between the reflection datum mark N of object-point P on the general object Ob and the specified point on the reflecting surface (nasal surface) 18 at this.Object-point P on the object Ob is imaged as direct image Im by the O in the camera lens 21 on the video camera 20 on CCD element 23.On the other hand, the scattered reflection on reflecting surface 18 surfaces of reflector 15 of the picture of the object-point P on the object Ob, by above-mentioned camera lens 21 on CCD element 23 as indirect image In incident.Indirect image In since the scattered reflection of reflecting surface 18 and reflecting surface 18 beyond the focus of camera lens 21, though can imaging, for simplicity, suppose that reflecting surface 18 carries out direct reflection, represent the center of light path with line segment PN and NO for the purpose of convenient.
Among the whole picture plane F of CCD element 23, the line of sight of a pair of maximum (face) scope that Vm, Vm enclosed is a camera lens 21 CCD element 23 can imaging scope, this scope is equivalent to absolute visual field.Certainly, much less, absolute visual field also can launch on the direction of the vertical paper of Fig. 1.In the whole picture plane F corresponding to this scope, the scope indirect image Fn of portion that the border line of sight Vn that the reflecting surface top 18a of absolute visual field line Vm from upper left to the bottom right and, reflecting surface 18 is connected with the O of camera lens 21 is enclosed is the scope of shooting indirect image In.The scope direct image Fm of portion corresponding to remaining part is the scope of shooting direct image Im.
Vertical reference line Cv among Fig. 1 represents the reference axis by the zero point of the horizontal direction at camera lens 21 centers and paper thickness direction, and the vertical reference line Cv of the photographic plane by CCD element 23 represents the reference axis of the datum mark of vertical direction.Image coordinate is by (X, Y Z) are coordinate representation.Symbol X among the figure, Xn, the direct image Im of the object-point P on Xmi and Xni difference indicated object point P, reflection datum mark N, the Fm of direct image portion and the indirect image In of the object-point P of the Fn of indirect image portion and the horizontal range between the horizontal datum Ch.Equally, these each points with respect to the horizontal range symbol Y of horizontal datum Ch on Fig. 1 paper vertical direction, Yn, Ymi and Yni represent.Symbols Z among the figure and Zn be indicated object point P and the horizontal range that reflects datum mark N and vertical reference line Cv respectively.In other words, apart from the degree of depth that Zn represents to reflect datum mark N, the vertical direction distance of object-point P and reflection datum mark N is represented by Z-Zn.
Yet, run into body surface from the light of illumination, reflect according to surperficial optical characteristics.Can describe with following formula in video camera 20 viewed catoptrical intensity I (λ).
The spectral intensity of E (λ) expression illumination distributes in I (λ)=E (λ) ρ (λ) (1) formula, the reflection of ρ (λ) expression object surfaces, and λ represents wavelength.I (λ) be will reflect then and 3 look R, G, B resolved into.With the resolution indirect illumination that dwindles direct image Im in the horizontal direction be reflecting surface 18, mean with this form and measure illumination and obtain indirect image In.
Light by reflecting surface 18 direct reflections is surrounded by the distribution of diffusion light.This will be with the light of different weighting influences as the Fn of the indirect image portion arrival of CCD element 23.For example in Fig. 2 (a), respectively along the light of optical axis S1 and S2 incident for example have with optical axis S1, S2 with direct reflection be the approximate scattered reflection intensity distributions of Gaussian Profile G1, G2 on summit.And will be influenced as the light that the Fn of indirect image portion arrives on the CCD element 23 by the value with intensity DRC1, DRC2 to the light on the specific optical axis Sn of CCD element 23.According to relevant approximate, the light C of the point reflection from the nose can describe with following formula.
C (X, Y)=∫ ∫ Eo (X, Y) ρ o (X, Y) B1 (X, Y) in dXdY (2) formula, the point on the subscript o indicated object thing Ob.Light C represents to arrive from sight the summation of all illuminating raies of the N bar on the nasal surface.At this moment weight coefficient changes according to the angle of light, perhaps the surface roughness of nasal surface and changing.The roughness on the optical characteristics of fuzzy coefficient B1 and reflecting surface 18 and surface is relevant.
If reflecting surface 18 is outside the focus of camera lens 21, these light that reflected by reflecting surface 18 probably project into circle.Its intensity distributions shown in Fig. 2 (b), changes according to Gaussian function is approximate in the diametric(al) of circle of confusion.Therefore, the light income of these pixels of the Fn of indirect image portion in CCD element 23 is the summation with weighting of ox-eye.The size of this ox-eye depends on fuzzy coefficient B2, and this fuzzy coefficient depends on focal length and from the degree of depth of the camera lens 21 reflection datum mark N of video camera 20.
(Xni, Yni) (Xni, Yni) in dXnidYni (3) formula, subscript n i represents the pixel of indirect image In to C to Cni=∫ ∫ B2, secondary coordinate (Xni, Yni) coordinate of expression indirect image on CCD element 23.
Comprise the result of two formulas of two kinds of fuzzy coefficient B1, B2 as combination, carry out the fuzzy operation in two spaces.The focus of a basis on reflecting surface 18 departs from, another according to by reflecting surface 18 outside the focus of camera lens 21 time caused focus on CCD element 23 depart from and carry out.Fuzzy process is carried out in two layers of control respectively.Comprise the brightness that continuous convolution that two formulas of above-mentioned two kinds of fuzzy coefficients carry out can be thought representative object point P by combination.That is, the Fn of indirect image portion that is obtained on the CCD element 23 by being reflected in of reflecting surface 18 can think the brightness of representative object point P or its illumination nearby.
Therefore, carrying out intensity of colour signal D1, D2, D3 that the calculation process of formula (4) expression obtains represents by the color of the object-point P among the object Ob of revisal.This be because, after Rmi, Gmi in each principal coordinate of the color of the brightness of representative object point P or illumination and object-point P self, Bmi are divided by by Rni, Gni, Bni in the secondary coordinate of the brightness of representative object point P or illumination, just can remove the influence of the brightness etc. of object-point P.
D1(Xmi，Ymi)＝(Rmi/Rni)·S
D2(Xmi，Ymi)＝(Gmi/Gni)·S
D3 (Xmi, Ymi)=(Bmi/Bni) in S (3) formula, subscript m is represented direct image Im, n represents that i represents the image on the CCD element 23 by the indirect image In of reflecting surface 18.Again, D1, D2, D3 are respectively the principal coordinate (Xmi of the shooting after the color revisal, Ymi) R, G, B composition, Rmi, Gmi, Bmi are respectively principal coordinate (Xmi, Ymi) R, G, the B composition of direct pixel (Pm) in, Rni, Gn i, Bn i are respectively R, G, the B composition of secondary coordinate (Xni, Yn i) indirect pixel (Pn).(Xmi Ymi) is the coordinate of object-point P direct image that imaging obtained on CCD element 23 to principal coordinate, and (Xni Yni) is the coordinate of the indirect image that obtained by reflecting surface to secondary coordinate on CCD element 23.Coefficient S is in order not allow the value of D1～D3 saturated and regulate absolute value.
Effect as for the reflecting surface 18 of the transducer that detects spatial brightness can obtain to confirm by simple experiment.When the kicker light lamp shone directly on the white wall, relevant camera chain 1 of the present invention can obtain the image seen among Fig. 3 (c).The direct image Im of this spotlight shows as figure approximate in the left side, boundary line and white circle, and its indirect image In projects into the ellipse that horizontal resolution reduces, follows hot spot on every side.The color of the reflection representative illumination of reflecting surface 18.The color that can throw light on changes according to incandescent lamp that uses and color filter.If long and narrow banded optical projection is to white wall the time, spot corresponding among the value conduct of its R, G, B and direct image Im and the indirect image In is measured.The ratio of intensity of colour signal (D1, D2, D3) is even make the change color of illumination also roughly constant.
Below the position of explanation reflecting surface 18 and video camera 20 concerns.
Reflecting surface 18 is α with the angle that horizontal line became parallel with vertical reference line Cv.The angle that the represented reflection ray from reflecting surface 18 of line NO is become with horizontal line is ξ, and the angle that is become with reflecting surface 18 with line-transect NO is β.The angle that vertical line became of line NO and reflecting surface 18 is θ.Because line NO is the light of line PN after reflecting surface 18 direct reflections of expression incident light, so the angle that vertical line became of line PN and reflecting surface 18 also should be identical θ.F represents the focal length of the camera lens 21 of video camera 20.The angle that line PN is become with horizontal line is by x, and the formed object-point horizontal level of line PO and horizontal line angle is by Φ, and line PO and the formed visual field of line PN error angle are represented by ψ.
If the attention object-point then has ψ=Φ-x (5)
If note ∠ PNO, then x+ ξ=2 θ set up.
According to reflection datum mark N to angular dependence, then α=ξ+β sets up.
According to the relation of the vertical line that reflects datum mark N reflecting surface 18 on every side, then β=pi/2-θ.
According to above-mentioned 2 formulas around the reflection datum mark N, then ξ=alpha-beta=α+θ-pi/2 is set up.
Further, θ=ξ-α+pi/2 is set up.
After above-mentioned various arrangement, obtain following formula.
ψ＝Φ-x＝Φ-(2θ-ξ)＝Φ+ξ-2θ＝Φ+ξ-2(ξ-α+π/2)
＝Φ+ξ-2ξ+2α-π＝Φ-ξ+2α-π (6)
The angle [alpha] of reflecting surface 18 can be calculated by following formula.Object-point horizontal level angle Φ can obtain by following formula again.
Φ＝tan
-1((Z-f)/x) (7)
Angle ξ represents the index of the horizontal direction coordinate of reflection datum mark N on the reflecting surface 18 or indirect image In, can obtain by following formula.
ξ＝tan
-1(f/Xni) (8)
At this, the angle [alpha] of the best reflecting surface 18 the when horizontal coordinate of object-point P changes is represented in Fig. 5 (a).Angle [alpha] is that visual field error angle ψ is set at 2 degree, calculates under the situation of a less value.On the other hand, other angle is with its mean value representative.In Fig. 5 (a), Φ is a transverse axis with object-point horizontal level angle, and the value of the angle [alpha] of reflecting surface 18 is the longitudinal axis.When object-point horizontal level angle Φ increases,, wish that the angle [alpha] of reflecting surface 18 suitably reduces in order to keep visual field error angle ψ on a less definite value.
Shown in Fig. 1,3 and 4, each picture lines by the Fm of direct image portion that catches direct image Im and, separate and the Fn of indirect image portion that catches indirect image In is constituted by boundary line DL.The boundary line DL that separates direct image Fm of portion and the Fn of indirect image portion is corresponding with the reflecting surface top 18a of reflecting surface 18.The indirect pixel Pn that mapping of the present invention represents to constitute the Fm of direct image portion accounts for the ratio of the direct pixel set Pm of the Fm of direct image portion.When object Ob when the video camera 20 because angle ψ measures visual field error, the mapping difficulty that becomes.In order to make the visual field deviation between direct image Im and the indirect image In be minimum, ψ is as far as possible little for required angle.If angle ψ is big, object Ob maps directly on the Fm of direct image portion, and the Fn of indirect image portion will can not shine upon, and is perhaps opposite fully.Angle ψ can be with the coordinate top expression of reflecting surface 18.Can derive the relation of following formula from the geometric configuration of Fig. 1.
Tan(x)＝(Z-Zn)/(X-Xn) (9)
Positive contact transformation is asked on the both sides of formula (5), can be obtained following formula.
tanψ＝(tanΦ-tan(x))/(1+tanΦ·tan(x) (10)
Through type (9), (10) can be obtained,
tanψ＝[Xn(Z-f)+Zn·X+X·f-2X·Z]/
[Xn·X+Zn(f-Z)+Z(Z-f)-X
2] (11)
Formula (11) indicated object point (X, Z) and reflecting surface 18 on reflection datum mark N (Xn, the dependence of both sides' angle Zn).In formula (11), get X=0, will obtain the tangent angle on the camera optical axis point shown in the formula (12).
tanψ＝(Xn(Z-f)/(Zn(f-Z)+Z(Z-f)) (12)
If increase the datum mark horizontal range Xn of the horizontal range between reflection datum mark N and the horizontal datum Ch, then error ψ in the visual field increases.Therefore, reflecting surface 18 should be as far as possible and horizontal datum Ch horizontal registration, but visual field error ψ increases along with the increase of the degree of depth Zn of reflection datum mark N.Therefore, compare with the camera lens 21 of video camera 20 should be as far as possible little for the degree of depth Zn of reflection datum mark N.
Object ties up to expression among Fig. 5 (b) apart from the pass of Z and visual field error angle.Increase apart from Z, then visual field error angle ψ reduces.When being quite high value, if, then spend less than 2 when apart from more than 40cm near object Ob time error angle ψ.So long as not high-resolution banded illumination, the problem in the visual field is just inessential.Under common illumination condition, brightness can not change in the short period.Along with the increase from the datum-plane distance X n of reflection datum mark N, error angle ψ increases.Change if the angle [alpha] of reflecting surface 18 is pressed the tendency shown in Fig. 5 (a), its influence will be minimum.
By above-mentioned formula (5) and formula (6), can get x=Φ-ψ=π+ξ-2 α, further ask the tangent of this formula, then following formula is set up.
Tan(x)＝tan(π+ξ-2α)＝tan(ξ-2α) (13)
With following formula substitution formula (7) and formula (8), then can obtain
(Z-Zn)/(X-Xn)＝
[(f-Xni·tan(2α))/(Xni+f·tan(2α)) (14)
(X/Z)=A is set in the formula, after following formula (14) is launched, puts in order, can obtains following formula,
Xni/f＝{[(A-tan(2α)]-[Xn/Z-(Zn/Z)·tan(2α)]}/
[(1+A·tan(2α))-((Xn/Z)·tan(2α)+Zn/Z)] (15)
Further, when Z＞＞Zn, X＞＞during Xn, later half in the molecule of following formula and the denominator is zero, thereby following formula is set up.
Xni＝f[A-tan(2α)]/[1+A·tan(2α)] (16)
This formula is represented the mapping to the horizontal coordinate of the direct image Im of the object Ob that belongs to same scan line SL and indirect image In.Point coordinates Xni on the Fn of indirect image portion corresponding with an object-point on the object Ob does not directly depend on the value of distance, but depends on ratio A=(X/Z).This can illustrate by omitting Zn in formula.If when supposition object Ob was positioned at fully away from the position of video camera 20, it is very little that angle ψ becomes.At this moment, if the object-point P that should gaze at moves along OP, the reflection of the reflecting surface on the line segment PN only changes a little.Described suc as formula (16), mapping has direct relation for the shape of decision reflecting surface 18.
The gimmick of Fig. 3 (a), Fig. 4 (a) expression Nonlinear Mapping, the gimmick of Fig. 3 (b), Fig. 4 (b) expression Linear Mapping.In the relation of the position of Fm of direct image portion and the Fn of indirect image portion, the angle [alpha] of each the small part by how determining Xni, A=(X/Z) in the above-mentioned formula (16) and reflecting surface, mapping can be selected non-linear or linear relation.For the corresponding relation between the indirect pixel Pn of the direct pixel set Pm that illustrates at the Fm of direct image portion on 1 scan line SL and the Fn of indirect image portion, Fig. 4 represents direct pixel set Pm, the pairing relation in the oblique mobile back of pixel Pn indirectly with arrow.The direction of shining upon among the Fm of direct image portion is the direction of arrow Mm, and the mapping direction of the Fn of indirect image portion is the direction of opposite arrow Mn.Usually, the boundary line DL between Fm of direct image portion and the Fn of indirect image portion is vertical with the lower edge of whole picture plane F.
In the Nonlinear Mapping of Fig. 3 (a), Fig. 4 (a), gather Pm divided by each direct pixel that the pixel by different numbers constitutes with each indirect pixel Pn.In this mapping, each one is sized to a/d=b/c among corresponding mutually direct image Im and the indirect image In, that is, on Fm of direct image portion and the Fn of indirect image portion direction side by side, make the direct image Im of object Ob similar with indirect image In.Such mapping is applicable to the situation of the small part of image being carried out correct color revisal, for example the situation of photography flower or wisp.
In the Linear Mapping of Fig. 3 (b), Fig. 4 (b), allow the ratio (Pm/Pn) of each pixel of Fm of direct image portion and the Fn of indirect image portion correspondence be certain value.In this mapping, the size a/d and the b/c of each one are arranged to unequal among corresponding mutually direct image Im and the indirect image In.That is, direct image Im and indirect image In have to be dissmilarity.The part of the Fm of direct image portion is carried out the color revisal with even resolution.This mapping can the high speed processing image rate, can obtain the color revisal image of substantially linear.In addition, the device that carries out the correspondence of above-mentioned distribution can be realized with microcomputer 30 described later.
Reflecting surface 18 integral body are not straight line as shown in Figure 1, and the surperficial angle [alpha] of each small part is all inequality, and integral body is curve form.The reflecting surface 18 of Fig. 1 only is depicted as straight line for convenience of description.
When utilizing line chart design reflectivity face 18, at first, on the CCD element in the Fn of indirect image portion on DL right side, boundary line, but the angle [alpha] of the small part of reflecting surface is determined in veritcal limit boundary line that can the projection video at reflecting surface top 18a.The angle [alpha] of each small part of reflecting surface is determined according to the necessity shown in Fig. 5 (a).In the degree of depth is 1 meter scope, measure by line chart from the small partial-length of the length of direct image projection and corresponding reflecting surface 18.At this moment, the difference of the degree of depth is obtained the numerical value of budgetary estimate from formula (16), can not produce big error.That is, go for mensuration by the line chart of each pixel between Fm of direct image portion and the Fn of indirect image portion from the formula of above-mentioned mapping.
On the other hand, when utilizing numerical expression design reflectivity face 18, at first, begin to calculate from the boundary line of the light that arrives video camera 20 reflecting surface top 18a coordinate (Xo, Yo).When utilizing Linear Mapping,, calculate above-mentioned A=(X/Z) and M=(Xni/f) according to the corresponding relation of Fn of indirect image portion and the Fm of direct image portion.The angle [alpha] of the small part of reflecting surface utilizes formula (16) to determine under this coordinate.Then, utilize formula (17), (18) to calculate that (Xo Yo) departs from the coordinate of the part of slight distance from coordinate.When subscript n and n-1 represent that reflecting surface 18 is divided into small part in two formulas below near n-1 small part of reflecting surface top 18a with away from the relation between n the small part position of the reflecting surface end 18b of portion at this top.Further, (Xn, Yn) angle [alpha] in utilizes formula (16) to calculate to this new coordinate.Order repeats such operation, determines the curved surface of reflecting surface 18.
Z
n＝[Z
n-1-tanα
n-1(M
nf
n-X
n-1)]/[1-M
n·tan(α
n-1)] (17)
X
n＝X
n-1+(Z
n-1-Z
n)/tanα
n-1 (18)
Below, with reference to Fig. 1,3,4,7,8, the formation of relevant camera chain of the present invention is described.
Fig. 7 represents video camera 20 concrete formation nearby, and the front of video camera 20 has partly been installed the protection lid 10 with reflector 15.Protection lid 10 is by the lid body 11 of pyramid shape with for being embedded in the installation portion 12 on the video camera 20 outward.The lid body 11 of this protection lid 10 is in order to prevent the light intrusion video camera 20 beyond a pair of absolute visual field line (face) scope that Vm, Vm surrounded shown in Figure 1.Light beyond a pair of absolute visual field line (face) scope that Vm, Vm surrounded can cause error to the light from reflecting surface 18 for the revisal shooting, thereby wishes to remove.
Fig. 8 represents the logic diagram of camera chain 1, and this camera chain 1 roughly is made of protection lid 10, video camera 20, microcomputer 30, display unit 41 and color printer 42.After regulating light quantities by aperture 22, the image of taking in from the camera lens 21 of video camera 20 is imaged on the CCD element 23.The output signal of CCD element 23 also inputs to the light quantity that photographed image calculates in frame equalization portion 24 except that the image input section 31 that is input to microcomputer 30, the opening amount that the control opening is regulated motor 25 adjusting apertures 22 is saturated with the output signal that prevents CCD element 23.
Color correcting section 34 is calculated D1, D2 and D3 respectively according to formula (4), and its maximum is obtained by maximum selection portion 35 in whole image.The undersaturated adequate value of measuring the revisal item S of coefficient in the formula of can be used as (4) of this maximum, scaler 36 is determined the adequate value of revisal item S in color correcting section 34, D1, D2 and the D3 value of output revisal.For example, the amplitude-scaled that the microcomputer of 8 bits carries out information processing is 256, and the amplitude-scaled of distributing to 1 look in R, G, B three looks is about 85, then allow D1, D2 and D3 value amplitude-scaled maximum 85 with under set revisal item S.16,32 bit microcomputers then can distribute bigger amplitude-scaled, can show trickleer tone.
For the real example foregoing invention, XC-711 (trade mark) colour TV camera that above-mentioned video camera 20 adopts Sony Corporation to make, camera lens 21 adopts the C type (trade mark) of the COSMICARC system of focal length 12.5mm.Color-values adopts MINOLTA system colour difference meter CS-100 (trade mark).The device of the still image of experimental data is obtained in camera chain 1 initial constituting.Skin 17 adopts the grey ticker tape that removes gloss to make for carrying out scattered reflection, has obtained better result when employing is provided with the leather of oil membrane.The maximum of the amplitude of the Fn of indirect image portion and the ratio of whole image F is limited in 25%.Microcomputer adopts the Pentium (trade mark) of Action clock 120MHz, and for the image of 320 * 220 pixels, the processing time is 0.55 second.
The practicality of this camera chain 1 of revisal surface color and polish obtains to confirm to the influence of brightness by studying relevant color quality.The inventor makes up sight and fluorescent lamp is tested several full-color images.Image processing is that the color of the direct image Im color divided by indirect image In is carried out.The color of image has obtained improvement, and dark image brightens, and can confirm its trickle part, too bright visual deepening.The look of making an uproar after the following dark image in 100 luxs is handled by the method for using reflecting surface 18 becomes obvious.
To carry out the measuring of quality after the color revisal in order investigating, to have carried out two independently experiments by this camera chain 1.The red scraps of paper are arranged on the camera plane, under different illumination intensities, compare the color of the scraps of paper.The contrast of the effect of illumination intensity before and after revisal is shown in Fig. 6 (a) in the briliancy of the red scraps of paper.Transverse axis is an illumination intensity, and the longitudinal axis is represented the briliancy of the red scraps of paper.Shown in the curve of [before the revisal], usually, increase the intensity of sight illumination, the brightness of the color scraps of paper increases in the image.Shown in the curve of [after the revisal],, the brightness illumination intensity in the scraps of paper after the revisal also remains unchanged substantially even changing.According to the effect of the illumination intensity in the xy chromaticity coordinate of CIE1931 benchmark shown in Fig. 6 (b).In the curve of Fig. 6 (b) [before the revisal], along with the increase of the illumination intensity shown in the transverse axis, the value of the x chromaticity coordinate of the red scraps of paper shown in the longitudinal axis also increases.This means and when illumination intensity is inequality, produced the distortion of form and aspect with respect to original color.If the x chromaticity coordinate after the observation revisal will be found along with the enhancing of illumination only has only small minimizing.In addition, in Fig. 6 (a) and (b), value when briliancy though be equivalent to throw light on during 100 luxs of the most weak situation and x chromaticity coordinate value usually are different from than this big illumination intensity, but by the condition enactment of change reflecting surface, even also can keep the steadiness of illumination and form and aspect in low intensive illumination.
The color that former image deflects has been removed in the image color revisal that this camera chain 1 of employing reflecting surface 18 is carried out.The varying strength illumination is the histogram all similar of the image of revisal down.From this point, also we can say the influence of not feeling illumination intensity.Shown in Fig. 6 (a) and (b), the color parameter before and after the color revisal when the intensity of illumination changes within the specific limits, we can say that the brightness of color and form and aspect only are limited in the small excursion.
The possibility of another embodiment of the present invention is described at last.
In the above-described embodiment, the full picture plane F of CCD element 23 is a flat shape, and adopting full picture plane F theoretically for example is that some O with camera lens 21 is that the CCD element 23 of the curve form at center is possible.In this case, above-mentioned formula (15) can be replaced as following formula.
tan(2α)＝{A·tanξ+1-[(Zn/Z)+(Xn/Z)·tanξ]}/
{1-A·tanξ+[(Zn/Z)·tanξ-(Xn/Z)]} (19)
Further, when Z＞＞Zn, X＞＞during Xn, later half in the molecule of following formula and the denominator is zero, thereby formula (16) can be replaced as following formula.
tan(2α)＝(A·tanξ+1)/(1-A·tanξ) (20)
The curved design of reflecting surface 18 needs only with formula (20) replacement formula (16).
In the above-described embodiment, though reflecting surface 18 is to be made of the black skin that forms oil membrane, if as long as allow the light of reflecting surface carry out scattered reflection with suitable intensity, the employing surface is that other material formations such as grey of lattice-shaped also can.
In the above-described embodiment, practical example should be shown, in mapping, be divided into the direct pixel set Pm that an indirect pixel Pn and a plurality of direct pixel constitute.But, when equating, the amplitude of Fn of indirect image portion and the Fm of direct image portion also can only be divided into an indirect pixel Pn and a direct pixel Pm, perhaps, the Fn of indirect image portion is possible than indirect pixel set Pn and a direct pixel Pm of being divided into a plurality of indirect pixels formations under the wide situation of the amplitude of the Fm of direct image portion in theory.
Above-mentioned camera chain (device) can utilize in the digital camera for the video camera of photography motion video or the still image of photographing.Again, above-mentioned camera chain also can use in apart from deriving means rightly based on the solid of color.Existing solid is that change point with the colour coding on each scan line detects as characteristic point apart from deriving means.Its characteristic point is by to comparing between the stereo image of the left and right sides, if near colour coding then be consistent.The stability of colour coding is the very big advantage of color constancy of the present invention.This camera is adapted to solid apart from the deriving means, can improves the reliability of three-dimensional mapping.
Claims (9)
1. camera chain, comprise the imaging apparatus (23) of the chromatic image of making a video recording and the picture of object (Ob) is imaged on the camera lens (21) of described imaging apparatus, it is characterized in that (Vm is provided with in Vm) astigmatism of described object (Ob) is penetrated reflection and incided reflecting surface (18) on the described imaging apparatus (23) by described camera lens (21) at the absolute visual field that is made of described camera lens (21) and imaging apparatus (23); And be provided with each principal coordinate (Xmi of object-point (P) direct image that imaging obtains on described imaging apparatus (23) on the described object (Ob), Ymi) and each secondary coordinate (Xni, Yni) the corresponding corresponding intrument (30) of the indirect image of the described object-point (P) that on described imaging apparatus (23), obtains by described reflecting surface (18); And by the color of direct image (Im) is obtained the color correcting section (34) of the shooting of color revisal divided by the color of indirect image (In).
2. camera chain according to claim 1, this imaging apparatus (23) have in the direct image portion (Fm) of the direct image of described object (Ob) photographs object-point (P) and take indirect image portion (Fn) by the indirect image of the resulting object-point of described reflecting surface (18) (P).
3. camera chain according to claim 2 is set described reflecting surface (18) and is made that the amplitude of described direct image portion (Fm) is wideer than the amplitude of described indirect image portion (Fn).
4. camera chain according to claim 2, on the direction side by side of described direct image portion (Fm) and described indirect image portion (Fn), the direct image (Im) and the indirect image (In) of described object (Ob) are similar shapes.
5. camera chain according to claim 2, on the direction side by side of described direct image portion (Fm) and described indirect image portion (Fn), the ratio (Pm/Pn) of each pixel corresponding with described direct image portion (Fm) and described indirect image portion (Fn) difference is a certain value.
6. camera chain according to claim 1 has picture intelligence after scan line (SL) sequential scanning and carries out digitlization and be saved in image input section (31) in the memory.
7. camera chain according to claim 1, color correcting section (34) obtains the color compensating value by following formula,
D1(Xmi，Ymi)＝(Rmi/Rni)·S，
D2(Xmi，Ymi)＝(Gmi/Gni)·S，
D3 (Xmi, Ymi)=(Bmi/Bni) S, in the formula, D1, D2, D3 are respectively the principal coordinate of the shooting after the color revisal, and (Rmi, Gmi, Bmi are respectively principal coordinate (Xmi for Xmi, R Ymi), G, B composition, Ymi) R, G, the B composition of direct pixel (Pm) in, Rni, Gni, Bni are respectively secondary coordinate, and (S is the revisal item for Xni, Yni) R, the G of indirect pixel (Pn), B composition.
8. camera chain according to claim 1 is provided with protection lid (10) in the daylighting portion side of described camera lens (21).
9. camera chain according to claim 1, the shape of described reflecting surface (18) are pressed following formula and are formed,
Xni＝f(A-tan(2α))/(1+A·tan(2α))，
In the formula, f is the focal length of camera lens, A is (X/Z), the horizontal direction distance from horizontal datum (Ch) of X indicated object point (P), the vertical range from vertical reference line (Cv) of Z indicated object point (P), α represent reflecting surface (18) with the parallel angle that horizontal line constituted of vertical reference line (Cv).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN96180544A CN1097959C (en) | 1996-12-17 | 1996-12-17 | Imaging system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN96180544A CN1097959C (en) | 1996-12-17 | 1996-12-17 | Imaging system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1239629A CN1239629A (en) | 1999-12-22 |
| CN1097959C true CN1097959C (en) | 2003-01-01 |
Family
ID=5127976
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN96180544A Expired - Fee Related CN1097959C (en) | 1996-12-17 | 1996-12-17 | Imaging system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1097959C (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH03254727A (en) * | 1990-03-02 | 1991-11-13 | Canon Inc | Image photographing device |
| JPH03286398A (en) * | 1990-04-02 | 1991-12-17 | Sumitomo Electric Ind Ltd | Image processing type traffic flow measuring instrument |
| JPH05122708A (en) * | 1991-10-29 | 1993-05-18 | Sony Corp | Electronic camera |
| CN1076323A (en) * | 1991-10-30 | 1993-09-15 | 汤姆森消费电子有限公司 | Produce the equipment of figure according to canned data |
| JPH0882831A (en) * | 1994-09-09 | 1996-03-26 | Canon Inc | Lens hood |
| JP3254727B2 (en) * | 1992-05-20 | 2002-02-12 | 日新電機株式会社 | Method for forming boron nitride-containing film |
| JP3286398B2 (en) * | 1993-05-28 | 2002-05-27 | 三洋電機株式会社 | Electric vacuum cleaner |
-
1996
- 1996-12-17 CN CN96180544A patent/CN1097959C/en not_active Expired - Fee Related
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH03254727A (en) * | 1990-03-02 | 1991-11-13 | Canon Inc | Image photographing device |
| JPH03286398A (en) * | 1990-04-02 | 1991-12-17 | Sumitomo Electric Ind Ltd | Image processing type traffic flow measuring instrument |
| JPH05122708A (en) * | 1991-10-29 | 1993-05-18 | Sony Corp | Electronic camera |
| CN1076323A (en) * | 1991-10-30 | 1993-09-15 | 汤姆森消费电子有限公司 | Produce the equipment of figure according to canned data |
| JP3254727B2 (en) * | 1992-05-20 | 2002-02-12 | 日新電機株式会社 | Method for forming boron nitride-containing film |
| JP3286398B2 (en) * | 1993-05-28 | 2002-05-27 | 三洋電機株式会社 | Electric vacuum cleaner |
| JPH0882831A (en) * | 1994-09-09 | 1996-03-26 | Canon Inc | Lens hood |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1239629A (en) | 1999-12-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100352274C (en) | Keystone distortion correction of a projector | |
| US7085430B2 (en) | Correcting geometric distortion in a digitally captured image | |
| CN101061367A (en) | Three-dimensional measuring instrument, method, and program | |
| US8823876B2 (en) | Image processing apparatus | |
| CN1732696A (en) | Light source estimating device, light source estimating method, and imaging device and image processing method | |
| CN1910623A (en) | Image conversion method, texture mapping method, image conversion device, server-client system, and image conversion program | |
| CN1744672A (en) | Image capture apparatus and control method therefor | |
| CN1452730A (en) | Panoramic image acquisition device | |
| US20130314541A1 (en) | Methods and arrangements for object pose estimation | |
| CN107194881A (en) | A kind of removal image reflex reflector and method based on photometric stereo | |
| CN107622475B (en) | Gray correction method in image splicing | |
| CN1928523A (en) | Scintillation measuring method of display device and scintillation measuring device | |
| CN110231018A (en) | A kind of structure light measurement method, apparatus and computer readable storage medium | |
| JP2006285763A (en) | Method and device for generating image without shadow for photographic subject, and white board used therefor | |
| CN1467987A (en) | Image processing method, image processing program and memory media storing the program | |
| JPH1069543A (en) | Method and device for reconstituting curved surface of object | |
| JP7114335B2 (en) | IMAGE PROCESSING DEVICE, CONTROL METHOD FOR IMAGE PROCESSING DEVICE, AND PROGRAM | |
| CN1097959C (en) | Imaging system | |
| KR100417629B1 (en) | Imaging System | |
| JP2004133919A (en) | Device and method for generating pseudo three-dimensional image, and program and recording medium therefor | |
| US20040109077A1 (en) | Camera, imaging system and illuminations sensor | |
| US20190130544A1 (en) | Image processing apparatus, image capturing apparatus, image processing method, and storage medium | |
| EP4099218A1 (en) | Product authentication using security marks and optical means | |
| JP2012007953A (en) | Paint unevenness evaluation value calculating method, paint unevenness evaluation value calculating device and paint unevenness evaluating method | |
| KR101491123B1 (en) | An intellectual flash apparatus of digital imaging device and flash methods thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: WD Ref document number: 1024997 Country of ref document: HK |
|
| C19 | Lapse of patent right due to non-payment of the annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |