WO2019113869A1 - Camera array-based panoramic optical field acquisition device, processing method and computing equipment - Google Patents
Camera array-based panoramic optical field acquisition device, processing method and computing equipment Download PDFInfo
- Publication number
- WO2019113869A1 WO2019113869A1 PCT/CN2017/116057 CN2017116057W WO2019113869A1 WO 2019113869 A1 WO2019113869 A1 WO 2019113869A1 CN 2017116057 W CN2017116057 W CN 2017116057W WO 2019113869 A1 WO2019113869 A1 WO 2019113869A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- panoramic
- panorama
- array
- pictures
- parameters
- Prior art date
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 15
- 230000003287 optical effect Effects 0.000 title abstract description 3
- 239000011159 matrix material Substances 0.000 claims abstract description 12
- 238000000034 method Methods 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 13
- 230000003068 static effect Effects 0.000 claims description 4
- 230000001360 synchronised effect Effects 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 10
- 238000003384 imaging method Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/90—Arrangement of cameras or camera modules, e.g. multiple cameras in TV studios or sports stadiums
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/222—Studio circuitry; Studio devices; Studio equipment
- H04N5/262—Studio circuits, e.g. for mixing, switching-over, change of character of image, other special effects ; Cameras specially adapted for the electronic generation of special effects
- H04N5/265—Mixing
Definitions
- the present invention relates to the field of image processing technologies, and in particular, to a camera array-based panoramic light field collecting device, a processing method, and a computing device.
- the process of light field imaging includes the acquisition of the light field and the corresponding light field data processing. From the structural point of view, the collection of the light field mainly includes two ways of multi-camera combination and single-camera transformation.
- the former represented by the Massachusetts Institute of Technology and Stanford University, used a multi-camera array to build a light field acquisition system, reconstructing the images captured by each camera, and calculating the four-dimensional light field.
- the single-camera light field acquisition method introduces an optical modulation element into a single camera, changing its imaging structure, thereby redistributing the four-dimensional light field inside the camera to a two-dimensional detector plane.
- the light field panorama can not only realize the three-degree-of-freedom rotation when viewing, but also realize the multiple-focusing according to the needs of the viewer, and realize the three-dimensional displacement of the small angle to a certain extent, giving the viewer more An immersive experience.
- the image array is processed first, and a series of images with different focal lengths are obtained, and the images of different series of focal lengths are spliced.
- the panoramic light field image obtained by the method avoids the first method.
- the problem of zoom failure but because different stitching parameters are used for images at different focal lengths, problems such as blurring and misalignment are still unavoidable; some researchers have proposed to improve this algorithm, using full-focus images for stitching, and then Apply the stitching parameters to images at different focal lengths, but there are still defects in dealing with certain factors that are very relevant to the direction of the light (reflections, shadows, etc.);
- a panoramic light field collecting device a processing method, and a computing device based on a camera array are needed.
- An aspect of the present invention provides a camera array-based panoramic light field collecting device, the device comprising a positive N-shaped cylinder, and uniformly distributing K lenses of the same parameter on each side of the cylinder.
- the K cameras having the same attribute form a matrix camera array;
- the variable N of the cylinders takes a minimum value.
- Another invention of the present invention provides a method of all-light field processing, the method comprising:
- the pictures taken by the lens corresponding to the different positions on the different sides of the N-shaped cylinder are spliced into a panoramic picture to generate K panoramic array images;
- the K panoramic array maps generated in a mosaic are viewed as a loop in the horizontal direction, and a refocusing calculation with edge processing is performed.
- the K panoramic array maps are generated as follows:
- the generated first panorama and the K-1 panorama array are generated to generate K panoramic array images.
- the template parameter is a parameter set that can be multiplexed when the relative positions of the respective lenses of each group of pictures are changed.
- the K lenses with the same parameters are simultaneously exposed to obtain K ⁇ N pictures.
- the K lenses with the same parameters are synchronously exposed to acquire K ⁇ N pictures.
- Yet another aspect of the present invention is to provide a computing device suitable for use in an all-light field processing method, the computing device storing a plurality of instructions, loading and executing:
- the pictures taken by the lens corresponding to the different positions on the different sides of the N-shaped cylinder are spliced into a panoramic picture to generate K panoramic array images;
- the K panoramic array maps generated in a mosaic are viewed as a loop in the horizontal direction, and a refocusing calculation with edge processing is performed.
- the K panoramic array maps are generated as follows:
- the generated first panorama and the K-1 panorama array are generated to generate K panoramic array images.
- the template parameter is a parameter set that can be multiplexed when the relative positions of the respective lenses of each group of pictures are changed.
- the camera array-based panoramic light field collecting device, the processing method and the computing device provided by the invention adopt the lens synchronous exposure with the same parameters, splicing the lens group of the corresponding position, and using the template parameters calculated by the first group of panoramas.
- the mosaic of other groups of panoramas eliminates the problems of blurring and misalignment caused by different stitching parameters of images at different focal lengths.
- the panoramic light field collecting device, the processing method and the computing device provided by the invention can effectively cope with the defects caused by the factors which are related to the light direction.
- the panoramic light field collecting device, the processing method and the computing device provided by the invention reduce the calculation amount requirement, and the lens is not required to be calibrated in advance, thereby reducing the implementation difficulty of the panoramic light field restoration.
- FIG. 1 is a schematic structural view showing a panoramic light field collecting device based on a camera array of the present invention
- FIG. 2 is a schematic view showing the relationship between the number of sides of the positive N-shaped cylinder of the present invention and the scene of the lens;
- FIG. 3 is a schematic diagram showing a lens synchronous exposure acquisition picture with the same parameters of the present invention.
- FIG. 4 is a schematic diagram showing the generation of a panoramic array pattern array map of the present invention.
- Figure 5 is a diagram showing the refocusing calculation of the panoramic array diagram of the present invention.
- FIG. 1 is a schematic structural diagram of a panoramic light field collecting device based on a camera array according to the present invention.
- a panoramic light field collecting device based on a camera array includes a positive N-shaped cylinder 100 in a regular polygonal cylinder.
- K lenses 101 having the same parameters
- K lenses having the same properties constitute a matrix camera array.
- the video signal of each lens in this embodiment is represented by V nk .
- the value of the number of sides N of the positive N-shaped cylinder 100 is determined by the coincidence degree of the scene captured by the lens at the corresponding position on the adjacent surface of the cylinder, as shown in FIG. 2, the positive N-shaped cylinder of the present invention.
- an example of a regular hexagonal cylinder is taken as an example.
- the surface adjacent to the a-plane of the cylinder is the b-plane and the c-plane, and the lens 101a at the edge of the a-plane is The position of the lens 101b at the b-edge position corresponds to the position, and the lens 101a at the a-side edge position corresponds to the position of the lens 101c at the c-edge position.
- the angles ⁇ of the shooting scenes of all the lenses are the same, and the coincidence degree of the scenes taken by the lenses corresponding to the positions on the adjacent faces of the cylinders determines the number N of sides of the positive N-shaped cylinders.
- the edge position of the a-planes in the embodiment is
- the photographing scene of the lens 101b at the edge position of the lens 101a and the b-plane edge needs to coincide, and the photographing scene of the lens 101a at the edge position of the a-plane edge and the lens 101c at the c-plane edge position need to coincide.
- the variable N of the cylinders takes a minimum value.
- the photographing scene of the lens 101a at the edge position of the a-plane edge coincides with the photographing scene of the lens 101b at the edge position of the b-plane edge
- the photographing scene of the lens 101a at the edge position of the a-plane edge coincides with the photographing scene of the lens 101c at the edge position of the c-plane
- the positive N-shape The number of sides N of the cylinder is the minimum value.
- Illustrative in this embodiment is a regular hexagonal cylinder.
- the length of the matrix camera array on each side of the positive N-shaped cylinder is x
- the width of the matrix camera array is y
- K x ⁇ y
- the video signal of each lens is V nk .
- the image is collected by the camera-based panoramic light field collecting device provided by the present invention, and the collected image is processed by the all-light field processing method.
- the all-light field processing method includes:
- K lenses with the same parameters are simultaneously exposed to obtain K ⁇ N pictures.
- the K lenses with the same parameters are simultaneously exposed to obtain K ⁇ N pictures.
- a picture taken by a lens corresponding to a position on a different surface of the N-shaped cylinder is spliced into a panoramic picture to generate K panoramic array patterns.
- the K panoramic array maps are generated as follows:
- the lens with the same parameters as the synchronous exposure image is obtained.
- the first lens of the first surface of the positive N-shaped cylinder is taken as a picture of V 11 , ..., a positive N-shaped column.
- the picture taken at the last lens exposure of the first side of the body is V 1k .
- the first shot of the last face of the positive N-sided cylinder is taken as V N1
- the last shot of the last face of the positive N-sided cylinder is taken as V Nk .
- the first lens corresponding to all the positions on the face is obtained by the stitching algorithm to obtain the first panorama.
- the first panorama is calculated as follows:
- P 1 ( ⁇ ) is the first panorama
- Stitch is the splicing function
- ⁇ is the template parameter.
- the template parameter refers to a set of parameters that can be reused when the relative positions of the respective shots of each group of pictures are changed. When the subsequent panorama is generated, the parameter template is directly used.
- the remaining K-1 group pictures are sequentially stitched by the stitching algorithm using the template parameter ⁇ to generate a K-1 panorama.
- the picture of all the lens exposures of the second face of the positive N-shaped cylinder is V 21 , V 22 ... V 2k , ..., the picture of all the lens exposures of the last face of the positive N-sided cylinder Is V N1 , V N2 ... V Nk .
- the K-1 panorama is generated by the splicing algorithm in turn. Specifically, the K-1 panorama is calculated as follows:
- P 2 ( ⁇ ) is the second panorama
- P K ( ⁇ ) is the Kth panorama.
- the selection of the splicing algorithm is not specifically limited in the foregoing embodiment, and the embodiment is merely an exemplary description. In some embodiments, a person skilled in the art may select a better splicing algorithm according to a specific implementation situation.
- the present invention generates a schematic diagram of a panoramic array pattern array diagram, and generates a first panoramic image and the K-1 panoramic image array to generate K panoramic array images.
- the schematic diagram of the refocusing calculation of the panoramic array diagram of the present invention is performed, and the generated K panoramic array images are viewed as a loop in the horizontal direction, and the refocusing calculation with edge processing is performed, and the panoramic image is generated and displayed after calculation.
- Edge processed three dimensional display calculations may be employed in some embodiments.
- the computing device used in the embodiment for the all-light field processing method stores a plurality of instructions, and the computing device loads and executes the stored instructions:
- K lenses with the same parameters are simultaneously exposed to obtain K ⁇ N pictures.
- the K lenses with the same parameters are simultaneously exposed to obtain K ⁇ N pictures.
- a picture taken by a lens corresponding to a position on a different surface of the N-shaped cylinder is spliced into a panoramic picture to generate K panoramic array patterns.
- the K panoramic array maps are generated as follows:
- the first lens of the first face of the positive N-sided cylinder is taken as a picture of V 11
- the first lens of the first face of the positive N-shaped cylinder is taken as a picture of V 1k .
- the first shot of the last face of the positive N-sided cylinder is taken as V N1
- the last shot of the last face of the positive N-sided cylinder is taken as V Nk .
- the first lens corresponding to all the positions on the face is obtained by the stitching algorithm to obtain the first panorama.
- the first panorama is calculated as follows:
- P 1 ( ⁇ ) is the first panorama
- Stitch is the splicing function
- ⁇ is the template parameter.
- the template parameter refers to a set of parameters that can be reused when the relative positions of the respective shots of each group of pictures are changed. When the subsequent panorama is generated, the parameter template is directly used.
- the remaining K-1 group pictures are sequentially stitched by the stitching algorithm using the template parameter ⁇ to generate a K-1 panorama.
- the picture of all the lens exposures of the second face of the positive N-shaped cylinder is V 21 , V 22 ... V 2k , ..., the picture of all the lens exposures of the last face of the positive N-sided cylinder Is V N1 , V N2 ... V Nk .
- the K-1 panorama is generated by the splicing algorithm in turn. Specifically, the K-1 panorama is calculated as follows:
- P 2 ( ⁇ ) is the second panorama
- P K ( ⁇ ) is the Kth panorama.
- the selection of the splicing algorithm is not specifically limited in the foregoing embodiment, and the embodiment is merely an exemplary description. In some embodiments, a person skilled in the art may select a better splicing algorithm according to a specific implementation situation.
- the generated first panorama and the K-1 panorama array are generated to generate K panoramic array images.
- the generated K panoramic array maps are viewed as a loop in the horizontal direction, and a refocusing calculation with edge processing is performed, and a panoramic image is generated for display by calculation.
- the camera array-based panoramic light field collecting device, the processing method and the computing device provided by the invention adopt the lens synchronous exposure with the same parameters, splicing the lens group of the corresponding position, and using the template parameters calculated by the first group of panoramas.
- the mosaic of other groups of panoramas eliminates the problems of blurring and misalignment caused by different stitching parameters of images at different focal lengths.
- the panoramic light field collecting device, the processing method and the computing device provided by the invention can effectively cope with the defects caused by the factors which are related to the light direction.
- the panoramic light field collecting device, the processing method and the computing device provided by the invention reduce the calculation amount requirement, and the lens is not required to be calibrated in advance, thereby reducing the implementation difficulty of the panoramic light field restoration.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Studio Devices (AREA)
- Stereoscopic And Panoramic Photography (AREA)
Abstract
Provided by the present invention are a camera array-based panoramic optical field acquisition device, processing method and computing equipment, the device comprising a regular N-sided column, K lenses having the same parameters being evenly distributed on each side surface of the column, and the K lens having the same attributes forming a matrix camera array, wherein, the value of N is determined by a coincidence degree of a scene captured by the lenses of a corresponding position on adjacent surfaces of the column, the length of the matrix camera array is x, the width of the matrix camera array is y, and K=x×y. The present invention employs synchronous exposure of lenses which have identical parameters, thus eliminating the problems of virtualization and dislocation caused by different splicing parameters used in images under different focal lengths.
Description
本发明涉及图像处理技术领域,特别涉及一种基于相机阵列的全景光场采集装置、处理方法和计算设备。The present invention relates to the field of image processing technologies, and in particular, to a camera array-based panoramic light field collecting device, a processing method, and a computing device.
光场成像的过程包括光场的采集以及相应的光场数据处理。从结构上来分,光场的采集主要包括多相机组合和单相机改造两种方式。前者以麻省理工学院和斯坦福大学为代表,先后采用多相机阵列的方式搭建了光场采集系统,从每个相机拍摄到的图片进行重构,计算出四维光场。单相机光场采集方式在单个相机中引入光学调制元件,改变其成像结构,从而将相机内部的四维光场重新分布到一个二维的探测器平面上。The process of light field imaging includes the acquisition of the light field and the corresponding light field data processing. From the structural point of view, the collection of the light field mainly includes two ways of multi-camera combination and single-camera transformation. The former, represented by the Massachusetts Institute of Technology and Stanford University, used a multi-camera array to build a light field acquisition system, reconstructing the images captured by each camera, and calculating the four-dimensional light field. The single-camera light field acquisition method introduces an optical modulation element into a single camera, changing its imaging structure, thereby redistributing the four-dimensional light field inside the camera to a two-dimensional detector plane.
自从能够商用的光场相机问世之后,因为光场具有传统摄像手段所不具有的优势,研究者就考虑将在传统摄像中已经得到广泛应用的全景图像拼接应用于光场图像中。与传统全景图像相比,光场全景在观看时不仅能够实现三自由度的转动,还能够实现根据观看者的需要多次聚焦,并且在一定程度上实现小角度的三维位移,给观看者更具沉浸感的体验。Since the introduction of a commercially available light field camera, because the light field has advantages that conventional imaging methods do not have, researchers have considered the application of panoramic image stitching that has been widely used in conventional imaging to light field images. Compared with the traditional panoramic image, the light field panorama can not only realize the three-degree-of-freedom rotation when viewing, but also realize the multiple-focusing according to the needs of the viewer, and realize the three-dimensional displacement of the small angle to a certain extent, giving the viewer more An immersive experience.
现有的实现光场图像拼接的方法主要有两种,并且都是基于现有商用光场相机的:There are two main methods for realizing light field image stitching, and they are based on existing commercial light field cameras:
一、先对拍摄的图像阵列进行处理,获得一系列不同焦距下的图像,对这一系列不同焦距下的图像进行拼接,此方法处理后得到的全景光场图像避免了第一种方法会产生的变焦失效问题,但因为还是对不同焦距下的图像采用了不同的拼接参数,虚化、错位等问题仍无法避免;有研究者提出对这一算法进行改进,使用全焦图像进行拼接,再将拼接参数应用于不同焦距下的图像,但在应对某些与光线方向非常相关的因素时(反射、阴影等)仍然会有缺陷出现;First, the image array is processed first, and a series of images with different focal lengths are obtained, and the images of different series of focal lengths are spliced. The panoramic light field image obtained by the method avoids the first method. The problem of zoom failure, but because different stitching parameters are used for images at different focal lengths, problems such as blurring and misalignment are still unavoidable; some researchers have proposed to improve this algorithm, using full-focus images for stitching, and then Apply the stitching parameters to images at different focal lengths, but there are still defects in dealing with certain factors that are very relevant to the direction of the light (reflections, shadows, etc.);
二、直接对光线进行处理,将事先拍摄好的固定角度的光场图像用 来进行拼接,由于光场图像拍摄角度固定,相邻两张图像中的每条光线的关系就能通过计算得到,但需要非常大的计算能力以及事先的精确校准,实施难度较大。Second, directly processing the light, and using the fixed-angle light field image photographed in advance for splicing, since the light field image capturing angle is fixed, the relationship of each light in the adjacent two images can be calculated. But it requires a lot of computing power and accurate calibration in advance, which is difficult to implement.
因此,为了解决上述问题,需要一种基于相机阵列的全景光场采集装置、处理方法和计算设备。Therefore, in order to solve the above problems, a panoramic light field collecting device, a processing method, and a computing device based on a camera array are needed.
发明内容Summary of the invention
本发明的一各方面在于提供一种基于相机阵列的全景光场采集装置,所述装置包括一个正N边形柱体,在所述柱体的每一个侧面均匀分布K个参数相同的镜头,所述K个属性相同的镜头组成矩阵相机阵列;An aspect of the present invention provides a camera array-based panoramic light field collecting device, the device comprising a positive N-shaped cylinder, and uniformly distributing K lenses of the same parameter on each side of the cylinder. The K cameras having the same attribute form a matrix camera array;
其中,N的取值由所述柱体相邻面上对应位置的镜头所拍摄场景的重合度确定,所述矩阵相机阵列的长度为x,所述矩阵相机阵列的宽度为y,K=x×y。Wherein, the value of N is determined by the coincidence degree of the scene captured by the lens at the corresponding position on the adjacent surface of the cylinder, the length of the matrix camera array is x, and the width of the matrix camera array is y, K=x ×y.
优选地,所述柱体相邻面上对应位置的镜头所拍摄的场景刚好重合时,所述柱体的变数N取最小值。Preferably, when the scenes taken by the lenses corresponding to the positions on the adjacent faces of the cylinders coincide, the variable N of the cylinders takes a minimum value.
本发明的另一个发明在于提供一种全光场处理方法,所述方法包括:Another invention of the present invention provides a method of all-light field processing, the method comprising:
所述K个参数相同的镜头同步曝光,获取K×N个图片;The K pixels with the same parameters are synchronously exposed to obtain K×N pictures;
将N边形柱体不同面上对应位置的镜头拍摄的图片进行全景图片拼接,生成K个全景阵列图;The pictures taken by the lens corresponding to the different positions on the different sides of the N-shaped cylinder are spliced into a panoramic picture to generate K panoramic array images;
拼成生成的所述K个全景阵列图在水平方向看成一个循环,进行具有边缘处理的重聚焦计算。The K panoramic array maps generated in a mosaic are viewed as a loop in the horizontal direction, and a refocusing calculation with edge processing is performed.
优选地,所述K个全景阵列图按照如下方法生成:Preferably, the K panoramic array maps are generated as follows:
根据拼接算法对第一组图片进行全景拼接生成第一张全景图,并计算生成所述第一张全景图的模板参数;Performing panoramic stitching on the first group of pictures according to the stitching algorithm to generate a first panorama, and calculating template parameters for generating the first panorama;
将余下的K-1组图片使用所述模板参数,依次通过拼接算法拼接生成K-1张全景图;Using the template parameters of the remaining K-1 group pictures, and sequentially stitching to generate a K-1 panorama by stitching algorithms;
将生成的所述第一张全景图与所述K-1张全景图阵列,生成K个全景阵列图。The generated first panorama and the K-1 panorama array are generated to generate K panoramic array images.
优选地,所述模板参数为拍摄每一组图片的各个镜头相对位置不变时可以复用的参数集合。Preferably, the template parameter is a parameter set that can be multiplexed when the relative positions of the respective lenses of each group of pictures are changed.
优选地,当拍摄静态物体时,所述K个参数相同的镜头同步曝光,获取K×N个图片。Preferably, when a static object is photographed, the K lenses with the same parameters are simultaneously exposed to obtain K×N pictures.
优选地,当拍动态物体时,所述K个参数相同的镜头同步曝光,获取K×N个图片。Preferably, when the dynamic object is photographed, the K lenses with the same parameters are synchronously exposed to acquire K×N pictures.
本发明的再一个方面在于提供一种适用于全光场处理方法的计算设备,所述计算设备存储多条指令,加载并执行:Yet another aspect of the present invention is to provide a computing device suitable for use in an all-light field processing method, the computing device storing a plurality of instructions, loading and executing:
控制全景光场采集装置的K个参数相同的镜头同步曝光,获取K×N个图片;Controlling the K-parameters with the same K parameters of the panoramic light field collecting device to obtain K×N pictures;
将N边形柱体不同面上对应位置的镜头拍摄的图片进行全景图片拼接,生成K个全景阵列图;The pictures taken by the lens corresponding to the different positions on the different sides of the N-shaped cylinder are spliced into a panoramic picture to generate K panoramic array images;
拼成生成的所述K个全景阵列图在水平方向看成一个循环,进行具有边缘处理的重聚焦计算。The K panoramic array maps generated in a mosaic are viewed as a loop in the horizontal direction, and a refocusing calculation with edge processing is performed.
优选地,所述K个全景阵列图按照如下方法生成:Preferably, the K panoramic array maps are generated as follows:
根据拼接算法对第一组图片进行全景拼接生成第一张全景图,并计算生成所述第一张全景图的模板参数;Performing panoramic stitching on the first group of pictures according to the stitching algorithm to generate a first panorama, and calculating template parameters for generating the first panorama;
将余下的K-1组图片使用所述模板参数,依次通过拼接算法拼接生成K-1张全景图;Using the template parameters of the remaining K-1 group pictures, and sequentially stitching to generate a K-1 panorama by stitching algorithms;
将生成的所述第一张全景图与所述K-1张全景图阵列,生成K个全景阵列图。The generated first panorama and the K-1 panorama array are generated to generate K panoramic array images.
优选地,所述模板参数为拍摄每一组图片的各个镜头相对位置不变时可以复用的参数集合。Preferably, the template parameter is a parameter set that can be multiplexed when the relative positions of the respective lenses of each group of pictures are changed.
本发明提供的基于相机阵列的全景光场采集装置、处理方法和计算设备采用参数完全相同的镜头同步曝光,对相应位置的镜头分组进行拼接,并且将第一组全景图计算的模板参数用于其他组全景图拼接,消除了不同焦距下的图像采用了不同的拼接参数造成的虚化、错位等问题。The camera array-based panoramic light field collecting device, the processing method and the computing device provided by the invention adopt the lens synchronous exposure with the same parameters, splicing the lens group of the corresponding position, and using the template parameters calculated by the first group of panoramas. The mosaic of other groups of panoramas eliminates the problems of blurring and misalignment caused by different stitching parameters of images at different focal lengths.
本发明提供的全景光场采集装置、处理方法和计算设备,能够有效应对与光线方向非常相关的因素出现时造成的缺陷。The panoramic light field collecting device, the processing method and the computing device provided by the invention can effectively cope with the defects caused by the factors which are related to the light direction.
本发明提供的全景光场采集装置、处理方法和计算设备降低了计算量要求降低,无需事先对镜头进行校准,降低了全景光场还原的实施难度。The panoramic light field collecting device, the processing method and the computing device provided by the invention reduce the calculation amount requirement, and the lens is not required to be calibrated in advance, thereby reducing the implementation difficulty of the panoramic light field restoration.
应当理解,前述大体的描述和后续详尽的描述均为示例性说明和解释,并不应当用作对本发明所要求保护内容的限制。It is to be understood that the foregoing general descriptions
参考随附的附图,本发明更多的目的、功能和优点将通过本发明实施方式的如下描述得以阐明,其中:Further objects, features, and advantages of the present invention will be made apparent by the following description of the embodiments of the invention.
图1示意性示出了本发明基于相机阵列的全景光场采集装置的结构示意图;1 is a schematic structural view showing a panoramic light field collecting device based on a camera array of the present invention;
图2示出了本发明正N边形柱体的边数与镜头拍摄场景的关系示意图;2 is a schematic view showing the relationship between the number of sides of the positive N-shaped cylinder of the present invention and the scene of the lens;
图3示出了本发明参数相同的镜头同步曝光获取图片的示意图;FIG. 3 is a schematic diagram showing a lens synchronous exposure acquisition picture with the same parameters of the present invention; FIG.
图4示出了本发明生成全景阵列图阵列图的示意图;4 is a schematic diagram showing the generation of a panoramic array pattern array map of the present invention;
图5示出了本发明全景阵列图重聚焦计算的示意图。Figure 5 is a diagram showing the refocusing calculation of the panoramic array diagram of the present invention.
通过参考示范性实施例,本发明的目的和功能以及用于实现这些目的和功能的方法将得以阐明。然而,本发明并不受限于以下所公开的示范性实施例;可以通过不同形式来对其加以实现。说明书的实质仅仅是帮助相关领域技术人员综合理解本发明的具体细节。Objects and functions of the present invention, and methods for achieving the objects and functions will be clarified by referring to the exemplary embodiments. However, the invention is not limited to the exemplary embodiments disclosed below; it can be implemented in various forms. The essence of the description is merely to assist those skilled in the relevant art to understand the specific details of the invention.
在下文中,将参考附图描述本发明的实施例,相关技术术语应当是本领域技术人员所熟知的。在附图中,相同的附图标记代表相同或类似的部件,或者相同或类似的步骤,除非另有说明。下面通过具体实施例对本发明的内容进行说明。如图1所示本发明基于相机阵列的全景光场采集装置的结构示意图,实施例中一种基于相机阵列的全景光场采集装置,包括一个正N边形柱体100,在正多边形柱体柱体100的每一个侧面均匀分布K个参数相同的镜头101,K个属性相同的镜头组成矩阵相机阵列,具体地本实施例中每个镜头的视屏信号用V
nk表示。
Hereinafter, embodiments of the present invention will be described with reference to the drawings, and related technical terms should be well known to those skilled in the art. In the figures, the same reference numerals are used to refer to the same or similar parts, or the same or similar steps, unless otherwise stated. The contents of the present invention will be described below by way of specific examples. FIG. 1 is a schematic structural diagram of a panoramic light field collecting device based on a camera array according to the present invention. In the embodiment, a panoramic light field collecting device based on a camera array includes a positive N-shaped cylinder 100 in a regular polygonal cylinder. Each of the sides of the cylinder 100 is uniformly distributed with K lenses 101 having the same parameters, and K lenses having the same properties constitute a matrix camera array. Specifically, the video signal of each lens in this embodiment is represented by V nk .
根据本发明,正N边形柱体100的边数N的取值由柱体相邻面上对应位置的镜头所拍摄场景的重合度确定,如图2所示本发明正N边形柱体的边数与镜头拍摄场景的关系示意图,本实施例中示例性的以正六边 形柱体为例,柱体a面相邻的面为b面和c面,a面边缘位置的镜头101a与b面边缘位置的镜头101b位置对应,同时a面边缘位置的镜头101a与c面边缘位置的镜头101c位置对应。所有镜头的拍摄场景的角度β相同,柱体相邻面上对应位置的镜头所拍摄场景的重合度确定正N边形柱体的边数N,根据本发明,实施例中a面边缘位置的镜头101a与b面边缘位置的镜头101b的拍摄场景需要重合,a面边缘位置的镜头101a与c面边缘位置的镜头101c的拍摄场景需要重合。According to the present invention, the value of the number of sides N of the positive N-shaped cylinder 100 is determined by the coincidence degree of the scene captured by the lens at the corresponding position on the adjacent surface of the cylinder, as shown in FIG. 2, the positive N-shaped cylinder of the present invention. A schematic diagram of the relationship between the number of sides and the scene of the lens shooting. In the embodiment, an example of a regular hexagonal cylinder is taken as an example. The surface adjacent to the a-plane of the cylinder is the b-plane and the c-plane, and the lens 101a at the edge of the a-plane is The position of the lens 101b at the b-edge position corresponds to the position, and the lens 101a at the a-side edge position corresponds to the position of the lens 101c at the c-edge position. The angles β of the shooting scenes of all the lenses are the same, and the coincidence degree of the scenes taken by the lenses corresponding to the positions on the adjacent faces of the cylinders determines the number N of sides of the positive N-shaped cylinders. According to the present invention, the edge position of the a-planes in the embodiment is The photographing scene of the lens 101b at the edge position of the lens 101a and the b-plane edge needs to coincide, and the photographing scene of the lens 101a at the edge position of the a-plane edge and the lens 101c at the c-plane edge position need to coincide.
本发明所述柱体相邻面上对应位置的镜头所拍摄的场景刚好重合时,所述柱体的变数N取最小值。实施例中a面边缘位置的镜头101a与b面边缘位置的镜头101b的拍摄场景刚好重合,a面边缘位置的镜头101a与c面边缘位置的镜头101c的拍摄场景刚好重合,则正N边形柱体的边数N为最小值。本实施中示例性的为正六边形柱体。When the scenes taken by the lenses corresponding to the positions on the adjacent faces of the cylinders of the present invention are exactly coincident, the variable N of the cylinders takes a minimum value. In the embodiment, the photographing scene of the lens 101a at the edge position of the a-plane edge coincides with the photographing scene of the lens 101b at the edge position of the b-plane edge, and the photographing scene of the lens 101a at the edge position of the a-plane edge coincides with the photographing scene of the lens 101c at the edge position of the c-plane, and the positive N-shape The number of sides N of the cylinder is the minimum value. Illustrative in this embodiment is a regular hexagonal cylinder.
应当理解上述只是示例性的说明,本领域技术人员应当理解并不想与上述的实施例,对于相邻面上的位置对应的所有镜头的拍摄场景都应当相互重合。It should be understood that the above is merely an illustrative description, and those skilled in the art should understand that the shooting scenes of all the lenses corresponding to the positions on the adjacent faces should be coincident with each other.
本发明中正N边形柱体的每一个侧面上的矩阵相机阵列的长度为x,所述矩阵相机阵列的宽度为y,K=x×y,每个镜头的视屏信号为V
nk。
In the present invention, the length of the matrix camera array on each side of the positive N-shaped cylinder is x, the width of the matrix camera array is y, K = x × y, and the video signal of each lens is V nk .
利用本发明所提供的基于相机阵列的全景光场采集装置采集图片,对采集的图片通过全光场处理方法进行处理,具体实施例中全光场处理方法包括:The image is collected by the camera-based panoramic light field collecting device provided by the present invention, and the collected image is processed by the all-light field processing method. In the specific embodiment, the all-light field processing method includes:
K个参数相同的镜头同步曝光,获取K×N个图片。当拍摄静态物体时,所述K个参数相同的镜头同步曝光,获取K×N个图片。K lenses with the same parameters are simultaneously exposed to obtain K×N pictures. When a static object is photographed, the K lenses with the same parameters are simultaneously exposed to obtain K×N pictures.
当拍摄动态物体时,所述K个参数相同的镜头同步曝光,获取K×N个图片。When a dynamic object is photographed, the K lenses with the same parameters are simultaneously exposed to obtain K×N pictures.
将N边形柱体不同面上对应位置的镜头拍摄的图片进行全景图片拼接,生成K个全景阵列图。A picture taken by a lens corresponding to a position on a different surface of the N-shaped cylinder is spliced into a panoramic picture to generate K panoramic array patterns.
实施例中,K个全景阵列图按照如下方法生成:In an embodiment, the K panoramic array maps are generated as follows:
根据拼接算法对第一组图片进行全景拼接生成第一张全景图,并计算生成所述第一张全景图的模板参数。如图3所示参数相同的镜头同步曝光获取图片的示意图,实施例中正N边形柱体的第一个面的第一个镜 头曝光拍摄的图片为V
11,……,正N边形柱体的第一个面的最后镜头曝光拍摄的图片为V
1k。
Performing panoramic stitching on the first group of pictures according to the stitching algorithm to generate a first panorama, and calculating template parameters for generating the first panorama. As shown in FIG. 3, the lens with the same parameters as the synchronous exposure image is obtained. In the embodiment, the first lens of the first surface of the positive N-shaped cylinder is taken as a picture of V 11 , ..., a positive N-shaped column. The picture taken at the last lens exposure of the first side of the body is V 1k .
依次类推,正正N边形柱体的最后一个面的第一个镜头曝光拍摄的图片为V
N1,……,正N边形柱体的最后一个面的最后镜头曝光拍摄的图片为V
Nk。
And so on, the first shot of the last face of the positive N-sided cylinder is taken as V N1 , ..., the last shot of the last face of the positive N-sided cylinder is taken as V Nk .
所有面上位置对应的第一个镜头通过拼接算法得到第一张全景图,具体地,第一张全景图按照如下方法计算得到:The first lens corresponding to all the positions on the face is obtained by the stitching algorithm to obtain the first panorama. Specifically, the first panorama is calculated as follows:
P
1(δ)=Stitch
1(V
11,V
21,...,V
N1)
P 1 (δ)=Stitch 1 (V 11 , V 21 ,...,V N1 )
其中,P
1(δ)为第一张全景图,Stitch为拼接函数,δ为模板参数。模板参数是指拍摄每一组图片的各个镜头相对位置不变时可以复用的参数集合,在后续的全景图生成时,直接使用该参数模板。
Where P 1 (δ) is the first panorama, Stitch is the splicing function, and δ is the template parameter. The template parameter refers to a set of parameters that can be reused when the relative positions of the respective shots of each group of pictures are changed. When the subsequent panorama is generated, the parameter template is directly used.
将余下的K-1组图片使用模板参数δ,依次通过拼接算法拼接生成K-1张全景图。The remaining K-1 group pictures are sequentially stitched by the stitching algorithm using the template parameter δ to generate a K-1 panorama.
实施例中正N边形柱体的第二个面的所有镜头曝光拍摄的图片为V
21,V
22…V
2k,……,正N边形柱体的最后一个面的所有镜头曝光拍摄的图片为V
N1,V
N2…V
Nk。依次通过拼接算法生成K-1张全景图,具体地,K-1张全景图按如下方法计算得到:
In the embodiment, the picture of all the lens exposures of the second face of the positive N-shaped cylinder is V 21 , V 22 ... V 2k , ..., the picture of all the lens exposures of the last face of the positive N-sided cylinder Is V N1 , V N2 ... V Nk . The K-1 panorama is generated by the splicing algorithm in turn. Specifically, the K-1 panorama is calculated as follows:
P
2(δ)=Stitch
2(V
12,V
22,...,V
N2),……,P
K(δ)=Stitch
K(V
1K,V
2K,...,V
NK)。
P 2 (δ)=Stitch 2 (V 12 , V 22 , . . . , V N2 ), . . . , P K (δ)=Stitch K (V 1K , V 2K , . . . , V NK ).
P
2(δ)为第二张全景图,……,P
K(δ)为第K张全景图。
P 2 (δ) is the second panorama, ..., P K (δ) is the Kth panorama.
在上述实施例中对于拼接算法的选择不做具体限定,实施例只是示例性说明,在一些实施方式中本领域技术人员可以根据具体的实施情形选择更优的拼接算法。The selection of the splicing algorithm is not specifically limited in the foregoing embodiment, and the embodiment is merely an exemplary description. In some embodiments, a person skilled in the art may select a better splicing algorithm according to a specific implementation situation.
如图4所示本发明生成全景阵列图阵列图的示意图,将生成的第一张全景图与所述K-1张全景图阵列,生成K个全景阵列图。实施例中总共生成由K=x×y个全景图组成的全景图阵列,全景图阵列的长度为x,宽度为y。As shown in FIG. 4, the present invention generates a schematic diagram of a panoramic array pattern array diagram, and generates a first panoramic image and the K-1 panoramic image array to generate K panoramic array images. In the embodiment, a total of a panorama array composed of K=x×y panoramas is generated, and the panorama array has a length of x and a width of y.
如图5所示本发明全景阵列图重聚焦计算的示意图,拼成生成的K个全景阵列图在水平方向看成一个循环,进行具有边缘处理的重聚焦计算,经过计算将生成全景图像进行显示。在一些实施例中可以采用边缘 处理的三维显示计算。As shown in FIG. 5, the schematic diagram of the refocusing calculation of the panoramic array diagram of the present invention is performed, and the generated K panoramic array images are viewed as a loop in the horizontal direction, and the refocusing calculation with edge processing is performed, and the panoramic image is generated and displayed after calculation. . Edge processed three dimensional display calculations may be employed in some embodiments.
本实施例中对于全光场处理方法使用的计算设备存储多条指令,利用计算设备加载并执行存储的指令:The computing device used in the embodiment for the all-light field processing method stores a plurality of instructions, and the computing device loads and executes the stored instructions:
K个参数相同的镜头同步曝光,获取K×N个图片。当拍摄静态物体时,所述K个参数相同的镜头同步曝光,获取K×N个图片。K lenses with the same parameters are simultaneously exposed to obtain K×N pictures. When a static object is photographed, the K lenses with the same parameters are simultaneously exposed to obtain K×N pictures.
当拍摄动态物体时,所述K个参数相同的镜头同步曝光,获取K×N个图片。When a dynamic object is photographed, the K lenses with the same parameters are simultaneously exposed to obtain K×N pictures.
将N边形柱体不同面上对应位置的镜头拍摄的图片进行全景图片拼接,生成K个全景阵列图。K个全景阵列图按照如下方法生成:A picture taken by a lens corresponding to a position on a different surface of the N-shaped cylinder is spliced into a panoramic picture to generate K panoramic array patterns. The K panoramic array maps are generated as follows:
根据拼接算法对第一组图片进行全景拼接生成第一张全景图,并计算生成所述第一张全景图的模板参数。实施例中正N边形柱体的第一个面的第一个镜头曝光拍摄的图片为V
11,……,正N边形柱体的第一个面的最后镜头曝光拍摄的图片为V
1k。
Performing panoramic stitching on the first group of pictures according to the stitching algorithm to generate a first panorama, and calculating template parameters for generating the first panorama. In the embodiment, the first lens of the first face of the positive N-sided cylinder is taken as a picture of V 11 , ..., the first lens of the first face of the positive N-shaped cylinder is taken as a picture of V 1k .
依次类推,正正N边形柱体的最后一个面的第一个镜头曝光拍摄的图片为V
N1,……,正N边形柱体的最后一个面的最后镜头曝光拍摄的图片为V
Nk。
And so on, the first shot of the last face of the positive N-sided cylinder is taken as V N1 , ..., the last shot of the last face of the positive N-sided cylinder is taken as V Nk .
所有面上位置对应的第一个镜头通过拼接算法得到第一张全景图,具体地,第一张全景图按照如下方法计算得到:The first lens corresponding to all the positions on the face is obtained by the stitching algorithm to obtain the first panorama. Specifically, the first panorama is calculated as follows:
P
1(δ)=Stitch
1(V
11,V
21,...,V
N1)
P 1 (δ)=Stitch 1 (V 11 , V 21 ,...,V N1 )
其中,P
1(δ)为第一张全景图,Stitch为拼接函数,δ为模板参数。模板参数是指拍摄每一组图片的各个镜头相对位置不变时可以复用的参数集合,在后续的全景图生成时,直接使用该参数模板。
Where P 1 (δ) is the first panorama, Stitch is the splicing function, and δ is the template parameter. The template parameter refers to a set of parameters that can be reused when the relative positions of the respective shots of each group of pictures are changed. When the subsequent panorama is generated, the parameter template is directly used.
将余下的K-1组图片使用模板参数δ,依次通过拼接算法拼接生成K-1张全景图。The remaining K-1 group pictures are sequentially stitched by the stitching algorithm using the template parameter δ to generate a K-1 panorama.
实施例中正N边形柱体的第二个面的所有镜头曝光拍摄的图片为V
21,V
22…V
2k,……,正N边形柱体的最后一个面的所有镜头曝光拍摄的图片为V
N1,V
N2…V
Nk。依次通过拼接算法生成K-1张全景图,具体地,K-1张全景图按如下方法计算得到:
In the embodiment, the picture of all the lens exposures of the second face of the positive N-shaped cylinder is V 21 , V 22 ... V 2k , ..., the picture of all the lens exposures of the last face of the positive N-sided cylinder Is V N1 , V N2 ... V Nk . The K-1 panorama is generated by the splicing algorithm in turn. Specifically, the K-1 panorama is calculated as follows:
P
2(δ)=Stitch
2(V
12,V
22,…,V
N2),……,P
K(δ)=Stitch
K(V
1K,V
2K,...,V
NK)。
P 2 (δ)=Stitch 2 (V 12 , V 22 , . . . , V N2 ), . . . , P K (δ)=Stitch K (V 1K , V 2K , . . . , V NK ).
P
2(δ)为第二张全景图,……,P
K(δ)为第K张全景图。
P 2 (δ) is the second panorama, ..., P K (δ) is the Kth panorama.
在上述实施例中对于拼接算法的选择不做具体限定,实施例只是示例性说明,在一些实施方式中本领域技术人员可以根据具体的实施情形选择更优的拼接算法。The selection of the splicing algorithm is not specifically limited in the foregoing embodiment, and the embodiment is merely an exemplary description. In some embodiments, a person skilled in the art may select a better splicing algorithm according to a specific implementation situation.
将生成的第一张全景图与所述K-1张全景图阵列,生成K个全景阵列图。实施例中总共生成由K=x×y个全景图组成的全景图阵列,全景图阵列的长度为x,宽度为y。拼成生成的K个全景阵列图在水平方向看成一个循环,进行具有边缘处理的重聚焦计算,经过计算将生成全景图像进行显示。The generated first panorama and the K-1 panorama array are generated to generate K panoramic array images. In the embodiment, a total of a panorama array composed of K=x×y panoramas is generated, and the panorama array has a length of x and a width of y. The generated K panoramic array maps are viewed as a loop in the horizontal direction, and a refocusing calculation with edge processing is performed, and a panoramic image is generated for display by calculation.
本发明提供的基于相机阵列的全景光场采集装置、处理方法和计算设备采用参数完全相同的镜头同步曝光,对相应位置的镜头分组进行拼接,并且将第一组全景图计算的模板参数用于其他组全景图拼接,消除了不同焦距下的图像采用了不同的拼接参数造成的虚化、错位等问题。The camera array-based panoramic light field collecting device, the processing method and the computing device provided by the invention adopt the lens synchronous exposure with the same parameters, splicing the lens group of the corresponding position, and using the template parameters calculated by the first group of panoramas. The mosaic of other groups of panoramas eliminates the problems of blurring and misalignment caused by different stitching parameters of images at different focal lengths.
本发明提供的全景光场采集装置、处理方法和计算设备,能够有效应对与光线方向非常相关的因素出现时造成的缺陷。The panoramic light field collecting device, the processing method and the computing device provided by the invention can effectively cope with the defects caused by the factors which are related to the light direction.
本发明提供的全景光场采集装置、处理方法和计算设备降低了计算量要求降低,无需事先对镜头进行校准,降低了全景光场还原的实施难度。The panoramic light field collecting device, the processing method and the computing device provided by the invention reduce the calculation amount requirement, and the lens is not required to be calibrated in advance, thereby reducing the implementation difficulty of the panoramic light field restoration.
结合这里披露的本发明的说明和实践,本发明的其他实施例对于本领域技术人员都是易于想到和理解的。说明和实施例仅被认为是示例性的,本发明的真正范围和主旨均由权利要求所限定。Other embodiments of the invention will be apparent to those skilled in the <RTIgt; The description and the examples are to be considered as illustrative only, and the true scope and spirit of the invention are defined by the claims.
Claims (10)
- 一种基于相机阵列的全景光场采集装置,其特征在于,所述装置包括一个正N边形柱体,在所述柱体的每一个侧面均匀分布K个参数相同的镜头,所述K个属性相同的镜头组成矩阵相机阵列;A panoramic light field collecting device based on a camera array, characterized in that the device comprises a positive N-shaped cylinder, and K lenses of the same parameter are evenly distributed on each side of the cylinder, the K A lens with the same properties constitutes a matrix camera array;其中,N的取值由所述柱体相邻面上对应位置的镜头所拍摄场景的重合度确定,所述矩阵相机阵列的长度为x,所述矩阵相机阵列的宽度为y,K=x×y。Wherein, the value of N is determined by the coincidence degree of the scene captured by the lens at the corresponding position on the adjacent surface of the cylinder, the length of the matrix camera array is x, and the width of the matrix camera array is y, K=x ×y.
- 根据权利要求1所述的装置,其特征在于,所述柱体相邻面上对应位置的镜头所拍摄的场景刚好重合时,所述柱体的变数N取最小值。The device according to claim 1, wherein the variable N of the cylinder takes a minimum value when the scenes taken by the lenses at corresponding positions on the adjacent faces of the cylinders coincide.
- 一种利用权利要求1或2所述装置采集的全光场处理方法,其特征在于,所述方法包括:An all-light field processing method for collecting the apparatus according to claim 1 or 2, wherein the method comprises:所述K个参数相同的镜头同步曝光,获取K×N个图片;The K pixels with the same parameters are synchronously exposed to obtain K×N pictures;将N边形柱体不同面上对应位置的镜头拍摄的图片进行全景图片拼接,生成K个全景阵列图;The pictures taken by the lens corresponding to the different positions on the different sides of the N-shaped cylinder are spliced into a panoramic picture to generate K panoramic array images;拼成生成的所述K个全景阵列图在水平方向看成一个循环,进行具有边缘处理的重聚焦计算。The K panoramic array maps generated in a mosaic are viewed as a loop in the horizontal direction, and a refocusing calculation with edge processing is performed.
- 根据权利要求3所述的方法,其特征在于,所述K个全景阵列图按照如下方法生成:The method according to claim 3, wherein said K panoramic array patterns are generated as follows:根据拼接算法对第一组图片进行全景拼接生成第一张全景图,并计算生成所述第一张全景图的模板参数;Performing panoramic stitching on the first group of pictures according to the stitching algorithm to generate a first panorama, and calculating template parameters for generating the first panorama;将余下的K-1组图片使用所述模板参数,依次通过拼接算法拼接生成K-1张全景图;Using the template parameters of the remaining K-1 group pictures, and sequentially stitching to generate a K-1 panorama by stitching algorithms;将生成的所述第一张全景图与所述K-1张全景图阵列,生成K个全景阵列图。The generated first panorama and the K-1 panorama array are generated to generate K panoramic array images.
- 根据权利要求4所述的方法,其特征在于,所述模板参数为拍摄每一组图片的各个镜头相对位置不变时可以复用的参数集合。The method according to claim 4, wherein the template parameter is a parameter set that can be multiplexed when the relative positions of the respective lenses of each group of pictures are changed.
- 根据权利要求3所述的方法,其特征在于,当拍摄静态物体时,所述K个参数相同的镜头同步曝光,获取K×N个图片。The method according to claim 3, wherein when the static object is photographed, the K lenses having the same parameters are simultaneously exposed to obtain K×N pictures.
- 根据权利要求3所述的方法,其特征在于,当拍摄动态物体时,所述K个参数相同的镜头同步曝光,获取K×N个图片。The method according to claim 3, wherein when the dynamic object is photographed, the K lenses having the same parameters are synchronously exposed to acquire K×N pictures.
- 一种适用于权利要求3至7至任一权利要求所述方法的计算设备,其特征在于,所述计算设备存储多条指令,加载并执行:A computing device suitable for use in the method of any one of claims 3 to 7 to claim 1, wherein the computing device stores a plurality of instructions, loads and executes:控制全景光场采集装置的K个参数相同的镜头同步曝光,获取K×N个图片;Controlling the K-parameters with the same K parameters of the panoramic light field collecting device to obtain K×N pictures;将N边形柱体不同面上对应位置的镜头拍摄的图片进行全景图片拼接,生成K个全景阵列图;The pictures taken by the lens corresponding to the different positions on the different sides of the N-shaped cylinder are spliced into a panoramic picture to generate K panoramic array images;拼成生成的所述K个全景阵列图在水平方向看成一个循环,进行具有边缘处理的重聚焦计算。The K panoramic array maps generated in a mosaic are viewed as a loop in the horizontal direction, and a refocusing calculation with edge processing is performed.
- 根据权利要求8所述的计算设备,其特征在于,所述K个全景阵列图按照如下方法生成:The computing device of claim 8 wherein said K panoramic array maps are generated as follows:根据拼接算法对第一组图片进行全景拼接生成第一张全景图,并计算生成所述第一张全景图的模板参数;Performing panoramic stitching on the first group of pictures according to the stitching algorithm to generate a first panorama, and calculating template parameters for generating the first panorama;将余下的K-1组图片使用所述模板参数,依次通过拼接算法拼接生成K-1张全景图;Using the template parameters of the remaining K-1 group pictures, and sequentially stitching to generate a K-1 panorama by stitching algorithms;将生成的所述第一张全景图与所述K-1张全景图阵列,生成K个全景阵列图。The generated first panorama and the K-1 panorama array are generated to generate K panoramic array images.
- 根据权利要求8所述的计算设备,其特征在于,所述模板参数为拍摄每一组图片的各个镜头相对位置不变时可以复用的参数集合。The computing device according to claim 8, wherein the template parameter is a parameter set that can be multiplexed when the relative positions of the respective lenses of each group of pictures are changed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2017/116057 WO2019113869A1 (en) | 2017-12-14 | 2017-12-14 | Camera array-based panoramic optical field acquisition device, processing method and computing equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2017/116057 WO2019113869A1 (en) | 2017-12-14 | 2017-12-14 | Camera array-based panoramic optical field acquisition device, processing method and computing equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2019113869A1 true WO2019113869A1 (en) | 2019-06-20 |
Family
ID=66818878
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2017/116057 WO2019113869A1 (en) | 2017-12-14 | 2017-12-14 | Camera array-based panoramic optical field acquisition device, processing method and computing equipment |
Country Status (1)
Country | Link |
---|---|
WO (1) | WO2019113869A1 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110536055A (en) * | 2019-08-28 | 2019-12-03 | 北京拙河科技有限公司 | A kind of video camera |
CN115988338A (en) * | 2022-07-29 | 2023-04-18 | 南京理工大学 | Far field signal inversion reconstruction method based on compound eye camera array |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1479525A (en) * | 2002-06-27 | 2004-03-03 | Integrated design of omnidirectional camera and microphone array | |
CN103338343A (en) * | 2013-05-29 | 2013-10-02 | 山西绿色光电产业科学技术研究院(有限公司) | Multi-image seamless splicing method and apparatus taking panoramic image as reference |
CN104506761A (en) * | 2014-12-20 | 2015-04-08 | 中国地质大学(武汉) | 360-degree panoramic stereoscopic camera |
CN106371281A (en) * | 2016-11-02 | 2017-02-01 | 辽宁中蓝电子科技有限公司 | Multi-module 360-degree space scanning and positioning 3D camera based on structured light |
CN108156395A (en) * | 2017-12-13 | 2018-06-12 | 中国电子科技集团公司电子科学研究院 | Panorama optical field acquisition device, processing method and computing device based on camera array |
-
2017
- 2017-12-14 WO PCT/CN2017/116057 patent/WO2019113869A1/en active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1479525A (en) * | 2002-06-27 | 2004-03-03 | Integrated design of omnidirectional camera and microphone array | |
CN103338343A (en) * | 2013-05-29 | 2013-10-02 | 山西绿色光电产业科学技术研究院(有限公司) | Multi-image seamless splicing method and apparatus taking panoramic image as reference |
CN104506761A (en) * | 2014-12-20 | 2015-04-08 | 中国地质大学(武汉) | 360-degree panoramic stereoscopic camera |
CN106371281A (en) * | 2016-11-02 | 2017-02-01 | 辽宁中蓝电子科技有限公司 | Multi-module 360-degree space scanning and positioning 3D camera based on structured light |
CN108156395A (en) * | 2017-12-13 | 2018-06-12 | 中国电子科技集团公司电子科学研究院 | Panorama optical field acquisition device, processing method and computing device based on camera array |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110536055A (en) * | 2019-08-28 | 2019-12-03 | 北京拙河科技有限公司 | A kind of video camera |
CN115988338A (en) * | 2022-07-29 | 2023-04-18 | 南京理工大学 | Far field signal inversion reconstruction method based on compound eye camera array |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP3163535B1 (en) | Wide-area image acquisition method and device | |
CN110782394A (en) | Panoramic video rapid splicing method and system | |
CN106462944B (en) | High-resolution panorama VR generator and method | |
US20180213218A1 (en) | Equipment and method for promptly performing calibration and verification of intrinsic and extrinsic parameters of a plurality of image capturing elements installed on electronic device | |
US6023588A (en) | Method and apparatus for capturing panoramic images with range data | |
US9946955B2 (en) | Image registration method | |
CN107358577B (en) | Rapid splicing method of cubic panoramic image | |
JP2017108387A (en) | Image calibrating, stitching and depth rebuilding method of panoramic fish-eye camera and system thereof | |
WO2010061956A1 (en) | Stereoscopic image processing device, method, recording medium and stereoscopic imaging apparatus | |
JPH07294215A (en) | Method and apparatus for processing image | |
CN109166076B (en) | Multi-camera splicing brightness adjusting method and device and portable terminal | |
KR20160051473A (en) | Method of setting algorithm for image registration | |
CN111866523B (en) | Panoramic video synthesis method and device, electronic equipment and computer storage medium | |
CN112648935A (en) | Image processing method and device and three-dimensional scanning system | |
CN108156383B (en) | High-dynamic billion pixel video acquisition method and device based on camera array | |
WO2019113869A1 (en) | Camera array-based panoramic optical field acquisition device, processing method and computing equipment | |
CN110276714B (en) | Method and device for synthesizing rapid scanning panoramic image | |
CN108156395A (en) | Panorama optical field acquisition device, processing method and computing device based on camera array | |
KR101801100B1 (en) | Video providing apparatus and method for supporting production of immersive contents | |
CN114022562A (en) | Panoramic video stitching method and device capable of keeping integrity of pedestrians | |
CN110796690B (en) | Image matching method and image matching device | |
CN107959797A (en) | Panorama optical field acquisition device, processing method and computing device based on camera array | |
CN112738425A (en) | Real-time video splicing system with multiple cameras for acquisition | |
JPH1198532A (en) | Stereoscopic image pickup device and stereoscopic display device | |
CN110086994A (en) | A kind of integrated system of the panorama light field based on camera array |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 17934805 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 17934805 Country of ref document: EP Kind code of ref document: A1 |