Embodiment
Consult Fig. 3 A, the embodiment of the invention provides a kind of photocentre video camera altogether, and this common photocentre video camera comprises: prism 10 and a plurality of camera heads 20,
Prism 10; Have a plurality of reflective surfaces and each reflective surface all posts polarizing coating; The reflective surface of said prism is used for the incident ray that sees through polarizing coating is reflected; The reflection ray of output polarizationization, on said each reflective surface on the polarization direction of polarizing coating and the adjacent reflective surface polarization direction of polarizing coating different;
A plurality of camera heads 20; Corresponding one by one with each reflective surface of said prism 10; And polaroid is installed all before the camera lens of each camera head; Said a plurality of camera head is used to obtain and utilizes each polaroid that the reflection ray of the polarization of each reflective surface of said prism reflection is carried out the light after the polarization manipulation, wherein, the polarization direction of the polaroid before the camera lens of each camera head with its one to one on the reflective surface of said prism the polarization direction of polarizing coating identical.
Wherein, the polarization direction of the polarizing coating on the polarization direction of the polarizing coating on each reflective surface of prism and the adjacent reflective surface can be vertical, also can be other angles, do not influence realization of the present invention.
For the technical scheme that makes the embodiment of the invention is clearer, consult Fig. 3 B, be that example is described the principle of work of photocentre video camera altogether with the reflective surface 100 of prism:
Post polarizing coating 101 on the reflective surface 100 of prism, the reflective surface 100 of said prism is used for the incident ray that sees through polarizing coating 101 is reflected the reflection ray of output polarizationization;
Camera head 20; Corresponding with the reflective surface 100 of said prism; And before the camera lens 200 of camera head 20 polaroid 201 is installed, the reflection ray that said camera head 20 is used to obtain the polarization of reflective surface 100 reflections that utilize 201 pairs of said prisms of polaroid carries out polarization manipulation light afterwards.
For more effective prevention video camera receives and its light that adjacent reflective surface reflected of reflective surface one to one; The photocentre video camera can also comprise altogether: dividing plate; The reflection ray of polarization that is used for stopping a reflective surface output of said prism gets into another the reflective surface corresponding camera head adjacent with this reflective surface, the boundary line coplane of said dividing plate and said adjacent two reflective surfaces.
Dividing plate can be fixedly connected with the shell of said altogether photocentre video camera, the structure that perhaps is one of the forming with the shell of said common photocentre video camera.
On the reflective surface of the common photocentre video camera that the embodiment of the invention provides polarizing coating is arranged; Before the camera lens of camera head polaroid is arranged; Because the polarization direction of the preceding polaroid of the polarization direction of polarizing coating and the pairing camera head lens of this reflective surface is identical on each reflective surface; Different with the polarization direction of polaroid before the pairing camera head lens of adjacent reflective surface; Cause camera head can only receive, can not receive the reflection ray of adjacent reflective surface (this adjacent reflective surface is and the adjacent reflective surface of the pairing reflective surface of said camera head), like this from the reflection ray on the reflective surface corresponding with it; The image that is formed by the reflection ray of two adjacent reflective surfaces of prism just can not be aliasing in together, just can not form the dip-dye band of emergence.
In order to make the embodiment of the invention clearer, following embodiment will be described in detail photocentre video camera altogether provided by the invention.
Fig. 4 shows the light path synoptic diagram based on a reflective surface up, supposes that incident ray 301 is natural light, and its polarization direction is circular polarization, and the polarization direction of polarizing coating 101 and polaroid 201 all is a horizontal polarization.
100 pairs of incident raies 301 that see through polarizing coating 101 of the reflective surface of prism 10 reflect; The reflection ray 302 of output polarizationization; Because the polarization direction of polarizing coating 101 is a horizontal polarization, so incident ray 301 sees through the linearly polarized light that becomes horizontal polarization behind the polarizing coating 101, the linearly polarized light of 100 pairs of these horizontal polarizations of reflective surface reflects; Output reflection light 302, this reflection ray 302 also are the linearly polarized lights of horizontal polarization.
201 pairs of reflection rays of polaroid 302 before the camera lens 200 of camera head 20 carry out polarization to be handled; Because reflection ray 302 is linearly polarized lights of horizontal polarization; The polarization direction of polaroid 201 is a horizontal polarization, is injected in the camera lens 200 of camera head 20 so this reflection ray 302 can see through polaroid 201.
The reflection ray 302 that camera head 20 receives the polarization that utilizes 100 reflections of 201 pairs of said prism 10 reflective surfaces of polaroid carries out polarization manipulation light afterwards.
On the reflective surface of the common photocentre video camera that the embodiment of the invention provides polarizing coating is arranged; Before the camera lens of camera head polaroid is arranged; Because the polarization direction of the preceding polaroid of the polarization direction of polarizing coating and the pairing camera head lens of this reflective surface is identical on each reflective surface; Different with the polarization direction of polaroid before the pairing camera head lens of adjacent reflective surface; Cause camera head can only receive, can not receive the reflected light of adjacent reflective surface (this adjacent reflective surface is and the adjacent reflective surface of the pairing reflective surface of said camera head), like this from the reflected light on the reflective surface corresponding with it; The image that is formed by the reflected light of two adjacent reflective surfaces of prism just can not be aliasing in together, just can not form the dip-dye band of emergence.
Fig. 5 shows the light path synoptic diagram based on a reflective surface down, supposes that incident ray 301 is natural light, and its polarization direction is circular polarization, and the polarization direction of polarizing coating 101 and polaroid 201 all is a horizontal polarization.Wherein, reflective surface, polaroid and camera head, repeat no more at this with embodiment illustrated in fig. 4 similar to the processing of light among this embodiment.
Optional, as shown in Figure 6, between reflective surface 100 and polarizing coating 101, transparent protective film 102 can also be arranged, be used to prevent that the coating on the reflective surface from coming off.Wherein, Incident ray 301 orders see through polarizing coating 101 and incide reflective surface 100 with transparent protective film 102; 100 pairs of incident raies 301 of reflective surface reflect, and reflection ray 302 orders that obtain see through transparent protective film 102, polarizing coating 101 directive camera heads 20.
Owing to adopted polaroid and polarizing coating; Light can reduce light quantity during through polaroid and polarizing coating; So arriving the catoptrical light quantity of camera head lens can be littler than the light quantity of incident light; Usually be merely 50% of incident light light quantity, so can adopt the aperture of regulating camera head to remedy the light loss that polaroid and polarizing coating bring.Aperture is one and is used for controlling light through camera lens that get into the device of the light quantity of light-sensitive surface in the fuselage, aperture size can be represented with the f value: the diameter of the focal length/camera lens aperture of f value=camera lens.The f value of aperture is usually between 2.8 to 16.The f value of aperture is littler, and the light quantity of light-sensitive surface just the more in the entering fuselage in the same unit interval.For example aperture is adjusted to f5.6 from f8, one-level that aperture is big, and simultaneously, the light quantity that gets into light-sensitive surface in the fuselage is also many one times.
The common photocentre video camera of the invention described above embodiment introduction stops camera head reception and its light that adjacent reflective surface reflected of reflective surface one to one through adopting optical principle; For this camera head of more effective prevention receives and its light that adjacent reflective surface reflected of reflective surface one to one; The photocentre video camera can also comprise altogether: dividing plate; The structure that is fixedly connected with the shell of said altogether photocentre video camera or is one of the forming with the shell of said common photocentre video camera; The reflection ray of polarization that is used for stopping a reflective surface output of said prism gets into the corresponding camera head of another reflective surface, and wherein, a said reflective surface and said another reflective surface are two adjacent reflective surfaces; Wherein, the boundary line coplane of said dividing plate and said adjacent two reflective surfaces.Concrete, the reflective surface of prism can up also can be down, and the reflective surface of prism is up the time, dividing plate be positioned at two adjacent reflective surface boundary lines directly over; The reflective surface of prism is down the time, dividing plate be positioned at two adjacent reflective surface boundary lines under.
As shown in Figure 7; P1 is a prism, and it has M1, M2 and three reflective surfaces of M3, and these three reflective surfaces up; C1, C2 and C3 are respectively the camera head of corresponding three reflective surface M1, M2 and M3; For the reflected light that makes M1 and M3 does not get into the camera lens of C2, on the boundary line S12 of M1 and these two adjacent reflective surfaces of M2, place way-board B12, placement way-board B23 on the boundary line S23 of M3 and these two adjacent reflective surfaces of M2.Wherein, way-board B12 and S12 coplane, the bottom surface D12 of way-board B12 overlaps in the projection of prism P1 bottom surface with S12 in the projection of prism P1 bottom surface; Way-board B23 and S23 coplane, the bottom surface D23 of way-board B23 overlaps in the projection of prism P1 bottom surface with S23 in the projection of prism P1 bottom surface.B12 and B23 have been divided into three parts being total to photocentre video camera volume inside like this; The corresponding reflective surface of each part and a camera head; Way-board can block the reflected light of adjacent reflective surface, makes camera head in this segment space can not receive the reflected light of adjacent reflective surface.
Consult Fig. 8, a kind of embodiment of image processing method is provided, this method embodiment comprises:
801, obtain the scene image that camera head is taken.
802, obtain the brightness value of each pixel in the marginal portion image of said scene image; Wherein, the formed image of light of the scene of the marginal portion image of the said scene image regional reflex that to be the said camera head reflective surface corresponding with said camera head of catching have a common boundary with the reflective surface that is adjacent.
This method is applicable to the common photocentre video camera that the foregoing description provides, and also is applicable to the common photocentre video camera that prior art provides.Photocentre video camera altogether as shown in Figure 7; Because reflective surface M2 and M1 and M3 are adjacent respectively; Marginal portion image in the scene image that the video camera that M2 is corresponding is taken comprises: the formed image of light of the scene of the regional reflex that the formed image of light of the scene of the regional reflex that reflective surface M2 and M1 have a common boundary and reflective surface M2 and M3 have a common boundary.Marginal portion image in the scene image that the video camera that M1 is corresponding is taken comprises: the formed image of light of the scene of the regional reflex that reflective surface M2 and M1 have a common boundary.
803, the regional reflex light that has a common boundary with the reflective surface that is adjacent according to the brightness value of each pixel in the marginal portion image of said scene image and the reflective surface corresponding with said camera head that preset forms the luminance compensation coefficient of each pixel in the image, and the brightness of each pixel in the marginal portion image of said scene image is compensated.
Before this step; In order to obtain the reflective surface corresponding with said camera head forms each pixel in the image with the regional reflex light of the reflective surface boundary that is adjacent luminance compensation coefficient; This method further comprises: obtain the marginal portion image and the non-marginal portion image of the monochrome image that said camera head takes, the marginal portion image of said monochrome image is the light formed image of the said camera head reflective surface corresponding with said camera head of catching with the monochromatic object of the regional reflex of the reflective surface boundary that is adjacent; The non-marginal portion image of said monochrome image is the part except that the image of said marginal portion in the said monochrome image; Obtain the average brightness value of pixel in the non-marginal portion image of said monochrome image, with the brightness value of each pixel in the marginal portion image of said monochrome image with respect to the ratio of said average brightness value brightness decay coefficient as each pixel in the marginal portion image of said monochrome image; Confirm that regional reflex light that the inverse of the brightness decay coefficient of each pixel in the marginal portion image of said monochrome image has a common boundary with the reflective surface that is adjacent as the said reflective surface corresponding with camera head forms the luminance compensation coefficient of respective pixel in the image.Wherein, monochromatic object can be monochromatic template, and this template is white template or gray scale template; Accordingly, the R of each pixel, G, B value equate or approximately equal in the monochrome image.
Wherein, The process that the brightness of each pixel in the marginal portion image of scene image is compensated can be to handle by frame; For speed up processing; Can adopt graphic process unit (Graphic ProcessingUnit, GPU), (Digital Signal Processing DSP) waits device to handle to digital signal processor.
The regional reflex light that the reflective surface corresponding with camera head that the embodiment of the invention presets through utilization has a common boundary with the reflective surface that is adjacent forms the brightness decay coefficient of each pixel in the image; Brightness to each pixel in the captured edge of image parts of images compensates, and has eliminated the blanking bar that juncture area caused by two reflective surfaces.
The regional reflex light that the acquisition reflective surface corresponding with camera head that the following specific descriptions embodiment of the invention provides has a common boundary with the reflective surface that is adjacent forms the process of the luminance compensation coefficient of each pixel in the image:
First kind of mode consulted Fig. 9, and it specifically comprises:
901, obtain the marginal portion image and the non-marginal portion image of the monochrome image of said camera head shooting.
902, obtain the average brightness value of pixel in the non-marginal portion image of said monochrome image.
Concrete, ask the brightness value sum of each pixel in the image of non-marginal portion, use said then and, obtain average brightness value divided by the number of pixel in the image of non-marginal portion.
903, with the brightness value of each pixel in certain delegation of marginal portion image in the monochrome image respectively divided by said average brightness value, obtain the brightness decay coefficient of each pixel in this row.
Concrete, shown in figure 10, the width of the marginal portion image in the phantom order color image is w, in the image coordinate of a certain pixel be (x, y), when asking y to be determined value in this step, the brightness decay coefficient of the pixel of different x correspondences, wherein, x ∈ [0, w] supposes B
xBe the brightness value of this pixel, the span of this brightness value is 0~255, and then this pixel intensity attenuation coefficient is:
Wherein,
Be the average brightness value of pixel in the non-marginal portion image of monochrome image, its span is 0~255, λ
xSpan be λ
x∈ (0,1], i.e. 0<λ
x≤1.
Following definite mode of describing the marginal portion picture traverse w in the monochrome image:
Concrete, judge whether the brightness value of each pixel in the monochrome image delegation is lower than threshold value, find the nearest pixel that is lower than threshold value in range image center, said pixel is w to the distance on monochrome image border.
904, confirm the brightness decay coefficient of each pixel in other row of marginal portion image of monochrome image, wherein, the brightness decay coefficient of pixel that said other row and said certain delegation have identical horizontal ordinate is identical.
905, confirm that regional reflex light that the inverse of the brightness decay coefficient of each pixel in the marginal portion image of monochrome image has a common boundary with the reflective surface that is adjacent as the said reflective surface corresponding with camera head forms the luminance compensation coefficient of respective pixel in the image.
The second way: consult Figure 11, it specifically comprises:
1101, obtain the marginal portion image and the non-marginal portion image of the monochrome image of said camera head shooting.
1102, obtain the average brightness value of pixel in the non-marginal portion image of said monochrome image.
1103, with the brightness value of each pixel of marginal portion image in the monochrome image divided by said average brightness value, obtain the brightness decay coefficient of each pixel.
1104, confirm that regional reflex light that the inverse of the brightness decay coefficient of each pixel in the marginal portion image of monochrome image has a common boundary with the reflective surface that is adjacent as the said reflective surface corresponding with camera head forms the luminance compensation coefficient of respective pixel in the image.
Need to prove; For each camera head,, then can only ask the brightness decay coefficient of pixel in the marginal portion image of a monochrome image if when self imaging conditions of camera head is constant; When if self imaging conditions of camera head changes; Such as changing f-number, the juncture area of then corresponding two reflective surfaces also can change, and need ask the brightness decay coefficient of the marginal portion image of monochrome image again.General, when f-number increased, the juncture area of two reflective surfaces also can become greatly, and when f-number reduced, the juncture area of two reflective surfaces also can reduce.
Optional; The brightness of each pixel compensates in the marginal portion image of the scene image that each camera head is taken, and obtains after the scene image after each self compensation, in order to obtain panoramic picture; This method also comprises: the scene image after a plurality of compensation is spliced, obtain panoramic picture.
Consult Figure 12 A, the embodiment of the invention provides a kind of image processing apparatus, and it mainly comprises:
Image acquisition unit 1201 to be compensated is used to obtain the scene image that camera head is taken;
Brightness value acquiring unit 1202 is used for obtaining the brightness value of marginal portion each pixel of image of said scene image; Wherein, the formed image of light of the scene of the marginal portion image of the said scene image regional reflex that to be the said camera head reflective surface corresponding with said camera head of catching have a common boundary with the reflective surface that is adjacent;
Compensating unit 1203; Be used for forming the luminance compensation coefficient of each pixel in the image with the regional reflex light of the reflective surface boundary that is adjacent, the brightness of each pixel in the marginal portion image of said scene image is compensated according to the brightness value of marginal portion each pixel of image of said scene image and the reflective surface corresponding that presets with said camera head.
Wherein, The luminance compensation coefficient that the regional reflex light that the reflective surface corresponding with said camera head has a common boundary with the reflective surface that is adjacent forms each pixel in the image can be just to be configured in advance on the said camera head, also can be that camera head utilizes that the brightness value of each pixel calculates in the monochrome image.
Camera head forms the luminance compensation coefficient of each pixel in the image for the regional reflex light that calculates the reflective surface corresponding with said camera head and have a common boundary with the reflective surface that is adjacent, consults Figure 12 B, and this camera head also comprises:
Mono-color image acquiring unit 1204; Be used to obtain the marginal portion image and the non-marginal portion image of the monochrome image that said camera head takes, the marginal portion image of said monochrome image is the light formed image of the said camera head reflective surface corresponding with said camera head of catching with the monochromatic object of the regional reflex of the reflective surface boundary that is adjacent; The non-marginal portion image of said monochrome image is the part except that the image of said marginal portion in the said monochrome image; The brightness decay coefficient is confirmed unit 1205; Be used for obtaining the average brightness value of the non-marginal portion image pixel of said monochrome image; With the brightness value of each pixel in the marginal portion image of said monochrome image with respect to the ratio of said average brightness value brightness decay coefficient as each pixel in the marginal portion image of said monochrome image; Can specifically be used for brightness value with certain each pixel of delegation of the marginal portion image of monochrome image divided by said average brightness value; Obtain the brightness decay coefficient of each pixel in said certain delegation; Confirm the brightness decay coefficient of each pixel in other row of marginal portion image of said monochrome image, wherein, the brightness decay coefficient of pixel that said other row and said certain delegation have identical horizontal ordinate is identical.Concrete attenuation coefficient confirms that method sees also the description of step 903-904 and step 1103 among the method embodiment, repeats no more at this.The luminance compensation coefficient is confirmed unit 1206, is used for confirming that the inverse of brightness decay coefficient of marginal portion each pixel of image of said monochrome image forms the luminance compensation coefficient of each pixel in the image as the regional reflex light of the said same reflective surface boundary that is adjacent of reflective surface corresponding with camera head.
The regional reflex light that two reflective surfaces that the embodiment of the invention presets through utilization have a common boundary forms the brightness decay coefficient of each pixel in the image; Brightness to each pixel in the captured edge of image parts of images compensates, and has eliminated the blanking bar that the zone caused that is had a common boundary by two reflective surfaces.
One of ordinary skill in the art will appreciate that all or part of step that realizes in the foregoing description method is to instruct relevant hardware to accomplish through program; Described program can be stored in a kind of computer-readable recording medium; ROM (read-only memory) for example, disk or CD etc.
More than common photocentre video camera, image processing method and device that the embodiment of the invention provided have been carried out detailed introduction; Used concrete example among this paper principle of the present invention and embodiment are set forth, the explanation of above embodiment just is used for helping to understand method of the present invention and core concept thereof; Simultaneously, for one of ordinary skill in the art, according to thought of the present invention, the part that on embodiment and range of application, all can change, in sum, this description should not be construed as limitation of the present invention.