CN107169924A - The method for building up and system of three-dimensional panoramic image - Google Patents

Abstract

The invention discloses a method and a system for establishing a three-dimensional panoramic image, and relates to the technical field of virtual reality. A plurality of photographing devices arranged symmetrically and evenly on the hemispherical surface in a spherical position relationship take pictures at the same time in the direction of the center of the hemisphere. The images captured by the device are spliced into a panorama and sent to the virtual reality device to map the panorama to the appearance of the corresponding virtual hemispherical model for displaying the three-dimensional panorama. Since each photographing device simultaneously shoots the center of the hemisphere from different angles from the outside to the inside, it is possible to display the three-dimensional panoramic view corresponding to the user's observation angle in real time when the user circles around the center of the hemisphere, so that the user can change the observation at will. The location greatly increases the scope of the user's observation activity area and improves the user experience. It can solve the problem that the observer's movable area is very limited and the experience effect is poor in the prior art.

Description

Method and system for building three-dimensional panoramic image

technical field

The invention relates to the technical field of virtual reality, in particular to a method and system for establishing a three-dimensional panoramic image.

Background technique

The 3D panoramic image captures the image information of the entire environmental scene through the camera, uses software to stitch and integrate the images, and processes the plane image to obtain a 3D panoramic image. The three-dimensional panoramic image can realize the simulation of the two-dimensional plane image into a real three-dimensional space, and achieve the effect of simulating and reproducing the real environment scene.

With the development of computer software and hardware technology and the increasing popularity of smart wearable devices, head-mounted virtual reality devices display virtual environment images in front of the observer through the image display screen, creating a feeling of being in the virtual environment for the observer. experience.

However, the viewing angles of the three-dimensional panoramic images currently used in head-mounted virtual reality devices are that the position of the observer is fixed, and the position of the observer is the center, and the 360° rotation in situ is observed outwards, and the head can look up and down at the same time. Look down and observe. The scene presented under this viewing angle of view makes the movable area of the observer very limited, and the experience effect is poor.

Contents of the invention

In order to solve the above problems, the present invention provides a method and system for establishing a three-dimensional panoramic image, which can expand the viewing angle and increase the user's sense of immersion.

The invention provides a method for establishing a three-dimensional panoramic image, including:

A plurality of photographing devices are arranged symmetrically and evenly on the hemispherical surface in a spherical position relationship with each other, and take pictures toward the center of the hemisphere at the same time, wherein the images taken by the spatially adjacent photographing devices partially overlap each other;

The image splicing device is respectively connected with a plurality of photographing devices, receives images sent by a plurality of photographing devices, and stitches the images sent by a plurality of photographing devices into a panorama, and the panorama is a hemispherical central panorama;

The virtual reality device is connected with the image splicing device, receives the panorama after splicing and processing by the image splicing device, maps the panorama to the appearance of the corresponding virtual hemispherical model to display the three-dimensional panorama, and the virtual hemispherical model and the hemisphere Face to face.

Optionally, the hemispherical surface is divided into multiple parts by spherical dimension lines and spherical longitude lines, and each part is intersected by adjacent longitude lines and dimension lines to obtain at least 3 intersection points, and the plurality of photographing devices include Wide-angle camera at the intersection.

Optionally, the system also includes:

The image compression device compresses the panorama after the image splicing device splicing process, and the image compression device is arranged in the image splicing device, or is externally placed in the image splicing device and connected with the image splicing device;

The image decompression device decompresses the panoramic image compressed by the image compression device, and sends the decompressed panoramic image to the virtual reality device for texture display. The image decompression device can be arranged in the virtual reality device.

Optionally, the image stitching device uses homography transformation between plane projections to convert images under different projection planes sent by multiple photographing devices into the same plane, and performs a process on each image after the projection planes are unified. Feature point monitoring and extraction, matching and screening of feature points for two images with overlapping areas, establishing matching point pairs between two images with overlapping areas, and performing image registration, fusion, and exposure on images after feature point matching Compensation and deghosting processing, stitching images captured by multiple camera devices into a panorama.

Optionally, the virtual reality device acquires the user's position and viewing angle information when the user circles around the center of the hemisphere, and performs a three-dimensional image on the image on the hemispherical model under the corresponding position and viewing angle according to the user's position and viewing angle information. Display; when the viewpoint height adjustment command is triggered, the corresponding viewpoint height position is raised, and the image on the hemispherical model under the corresponding viewpoint height, horizontal positioning and viewing angle direction is displayed as a three-dimensional image; when the user moves toward the center of the hemisphere When approaching, the resolution of the displayed 3D image is increased by a preset multiple to ensure that the displayed 3D image is clear and undistorted.

The present invention also provides a method for establishing a three-dimensional panoramic image, including:

Obtain images taken simultaneously by a plurality of photographing devices towards the center of the hemisphere, wherein the plurality of photographing devices are arranged symmetrically and evenly on the hemispherical surface in a spherical position relationship, and the images taken by the photographing devices adjacent in spatial position are part of each other overlapping;

Stitching images taken by a plurality of photographing devices into a panorama, the panorama being a hemispherical central panorama;

The stitched panorama is mapped onto the surface of the corresponding virtual hemispherical model to display the three-dimensional panorama, and the virtual hemispherical model corresponds to the hemispherical surface.

Optionally, the hemispherical surface is divided into multiple parts by spherical dimension lines and spherical longitude lines, and each part is intersected by adjacent longitude lines and dimension lines to obtain at least 3 intersection points, and the plurality of photographing devices include Wide-angle camera at the intersection.

Optionally, the method also includes:

Using the homography transformation between plane projections, the images in different projection planes sent by multiple camera devices are converted to the same plane;

Perform feature point monitoring and extraction for each image after the projection plane is unified, perform matching screening of feature points for two images with overlapping areas, and establish matching point pairs between the two images with overlapping areas;

Image registration and fusion, exposure compensation and deghosting are performed on the images after feature point matching, and multiple images are stitched into a panorama.

Optionally, the method also includes:

When the user circles around the central direction of the hemisphere, obtain the user's position and viewing angle information, and perform three-dimensional image display on the image on the hemispherical surface model under the corresponding position and viewing angle according to the user's position and viewing angle information;

When the viewpoint height adjustment command is triggered, the corresponding viewpoint height position is raised, and the image on the hemispherical model under the corresponding viewpoint height, horizontal positioning, and viewing angle direction is displayed as a three-dimensional image;

When the user approaches the center of the hemisphere, the resolution of the displayed 3D image is increased by a preset multiple to ensure that the displayed 3D image is clear and undistorted.

Optionally, the method also includes:

When it is detected that the button on the virtual reality device that triggers the height adjustment command of the viewpoint is triggered, the height adjustment command of the viewpoint is triggered; or

When it is detected that the user wearing the virtual reality device jumps continuously and triggers the preset times of the viewpoint height adjustment instruction, the viewpoint height adjustment instruction is triggered.

In the embodiment of the present invention, a plurality of photographing devices arranged symmetrically and evenly on the hemispherical surface in a spherical position relationship with each other simultaneously take pictures of the center direction of the hemisphere, splicing the images captured by the multiple photographing devices into a panorama and sending it to the virtual reality The device maps the panorama to the surface of the corresponding virtual hemispherical model to display the three-dimensional panorama. Since the images captured by each camera device partially overlap with each other (high degree of local overlap), the density of image coverage is high, and full-angle panoramic image coverage can be achieved, so that when partially overlapping images are spliced into a 3D panoramic image, the effect is smoother , which is closer to the real scene. Further, since each photographing device simultaneously shoots the central direction (scene) of the hemisphere from different angles from the outside to the inside, it is possible to display the three-dimensional panoramic view corresponding to the user's viewing angle in real time when the user circles around the central direction of the hemisphere , so that the user can change the observation position at will, which greatly increases the scope of the user's observation activity area and improves user experience.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description These are some embodiments of the present invention, and those skilled in the art can also obtain other drawings based on these drawings without creative effort.

FIG. 1 is a system architecture diagram for establishing a three-dimensional panoramic image provided by an embodiment of the present invention;

Fig. 2 is a schematic layout diagram of a photographing device provided by an embodiment of the present invention;

FIG. 3 is a schematic flowchart of a method for creating a three-dimensional panoramic image provided by an embodiment of the present invention.

detailed description

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

Terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates otherwise, "multiple" Generally, at least two kinds are included, but the case of including at least one kind is not excluded.

It should be understood that the term "and/or" used herein is only an association relationship describing associated objects, which means that there may be three relationships, for example, A and/or B, which may mean that A exists alone, and A and B exist simultaneously. B, there are three situations of B alone. In addition, the character "/" in this article generally indicates that the contextual objects are an "or" relationship.

It should be understood that although the terms first, second, third, etc. may be used to describe XXX in the embodiments of the present invention, these XXX should not be limited to these terms. These terms are only used to distinguish XXX from each other. For example, without departing from the scope of the embodiments of the present invention, the first XXX may also be called the second XXX, and similarly, the second XXX may also be called the first XXX.

It should also be noted that the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a good or system comprising a set of elements includes not only those elements but also includes items not expressly listed. other elements of the product, or elements inherent in the commodity or system. Without further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the article or system comprising said element.

The present invention simultaneously captures the same scene from different angles through a plurality of photographing devices, obtains multiple images overlapping and covering the entire scene in real time, and obtains a denser image coverage, so that the corresponding image can be provided to the user according to the real-time user orientation requirements, This enables the user to obtain a real-time stereoscopic visual effect experience.

FIG. 1 is a system architecture diagram for establishing a three-dimensional panoramic image provided by an embodiment of the present invention. As shown in FIG. 1, the system provided by the embodiment of the present invention includes:

N photographing devices are arranged symmetrically and evenly on the hemispherical surface in a spherical position relationship with each other, and simultaneously photograph the central direction of the hemisphere, wherein the images taken by the spatially adjacent photographing devices partially overlap each other; where N is greater than 3 .

In this embodiment, the hemispherical surface is divided into multiple parts by spherical dimension lines and spherical longitude lines, and each part is intersected by adjacent longitude lines and dimension lines to obtain at least 3 intersection points. wide-angle camera at the point of intersection. In an optional implementation, the hemisphere can be divided into two parts along the center point of the radius of the sphere by the dimension line of the sphere, and divided into four parts by the longitude line of the sphere. There are two adjacent longitude lines and two adjacent Two dimension lines intersect to obtain eight intersection points, and wide-angle cameras are arranged at each of the eight intersection points, and wide-angle cameras are arranged at the intersection points where two adjacent longitude lines intersect on the spherical north pole.

Fig. 2 is a schematic layout diagram of a photographing device provided by an embodiment of the present invention. As shown in Fig. 2, 9 wide-angle cameras are symmetrically and evenly arranged on the hemispherical surface, and the 9 wide-angle cameras are all directed toward the center of the hemisphere to shoot, and also That is to say, each wide-angle camera shoots the central direction of the hemisphere from different angles from the outside to the inside at the same time, and the images taken by adjacent wide-angle cameras partially overlap with each other (high local overlap), and the density of image coverage is high. Realize full-angle panoramic image coverage, so that when partially overlapping images are spliced into a 3D panoramic image, the effect is smoother and closer to the real scene.

The system of this embodiment also includes: an image splicing device, which is respectively connected to a plurality of photographing devices, receives images sent by a plurality of photographing devices, and splices the images sent by a plurality of photographing devices into a panorama, and the panorama is the spherical center a panorama of the object at the location;

When the image stitching device implements image stitching, it can use the homography transformation between plane projections to convert the images under different projection planes sent by multiple camera devices into the same plane, and each image after the projection plane is consistent To monitor and extract feature points from each image, perform feature point matching screening for two images with overlapping areas, establish matching point pairs between the two images with overlapping areas, and perform image registration and alignment on the images after feature point matching. Fusion, exposure compensation, and deghosting to stitch together images from multiple camera rigs into a panorama.

The system of this embodiment also includes: a virtual reality device, connected to the image stitching device, receiving the panorama after stitching by the image stitching device, mapping the panorama to the appearance of the corresponding virtual hemispherical model to display the three-dimensional panorama, The virtual hemispherical model corresponds to a hemispherical surface.

Specifically, when the virtual reality device is performing a three-dimensional panorama, when it detects that the user is circling around the central direction of the hemisphere, it acquires the user's position and viewing angle information, and according to the user's position and viewing angle information, the hemispherical surface under the corresponding position and viewing angle The image on the model is displayed as a three-dimensional image; when the viewpoint height adjustment command is triggered, the corresponding viewpoint height position is raised, and the image on the hemispherical model under the corresponding viewpoint height, horizontal positioning and viewing angle direction is displayed as a three-dimensional image; when When it is detected that the user approaches the center of the hemisphere, the resolution of the displayed 3D image is increased by a preset multiple to ensure that the displayed 3D image is clear and undistorted.

For this reason, the virtual reality device is provided with sensors, which can monitor the user's position and the transformation of the viewing angle.

In an optional implementation manner, the system of this embodiment further includes:

The image compression device compresses the panorama after the image splicing device splicing process, and the image compression device is arranged in the image splicing device, or is externally placed in the image splicing device and connected with the image splicing device;

The image decompression device decompresses the panoramic image compressed by the image compression device, and sends the decompressed panoramic image to the virtual reality device for texture display. The image decompression device can be arranged in the virtual reality device.

In the embodiment of the present invention, a plurality of photographing devices arranged symmetrically and evenly on the hemispherical surface in a spherical position relationship with each other simultaneously take pictures of the center direction of the hemisphere, splicing the images captured by the multiple photographing devices into a panorama and sending it to the virtual reality The device maps the panorama to the surface of the corresponding virtual hemispherical model to display the three-dimensional panorama. Since the images captured by each camera device partially overlap with each other (high degree of local overlap), the density of image coverage is high, and full-angle panoramic image coverage can be achieved, so that when partially overlapping images are spliced into a 3D panoramic image, the effect is smoother , which is closer to the real scene. Further, since each photographing device simultaneously takes pictures of the central direction of the hemisphere from different angles from the outside to the inside, it is possible to display the three-dimensional panoramic view corresponding to the user's viewing angle in real time when the user circles around the central direction of the hemisphere, so that the user The observation position can be changed at will, which greatly increases the scope of the user's observation activity area and improves user experience.

Based on the system architecture diagram shown in FIG. 1, FIG. 3 is a schematic flowchart of a method for establishing a three-dimensional panoramic image provided by an embodiment of the present invention, as shown in FIG. 3, including:

101. Acquiring images simultaneously captured by a plurality of photographing devices symmetrically and evenly arranged on the hemispherical surface in a spherical positional relationship with respect to the central direction of the hemispherical shape.

Wherein, the plurality of photographing devices are arranged symmetrically and evenly on the hemispherical surface in a spherical position relationship with each other, and the images taken by the respective photographing devices adjacent to each other in spatial position partially overlap each other;

Wherein, the specific arrangement of multiple photographing devices on the hemispherical surface includes: the hemispherical surface is divided into a plurality of parts by spherical dimension lines and spherical longitude lines, and each part is intersected by adjacent longitude lines and dimension lines to obtain at least 3 intersection points, so The plurality of photographing devices include a wide-angle camera arranged at each intersection. As shown in Figure 2, 9 wide-angle cameras are symmetrically and evenly arranged on the hemispherical surface, and the 9 wide-angle cameras are all directed towards the center of the hemisphere, that is to say, each wide-angle camera simultaneously views the hemisphere from different angles from the outside to the inside. In this way, the images captured by adjacent wide-angle cameras partially overlap with each other (high local overlap), and the density of image coverage is relatively high. Full-angle panoramic image coverage can be achieved, so that partially overlapping images can be spliced together. When it is a 3D panoramic image, the effect is smoother and closer to the real scene.

In the embodiment of the present invention, each photographing device is connected to the image stitching device, and can send the captured pictures to the image stitching device.

102. Stitch images captured by multiple photographing devices into a panorama.

Specifically, after the image stitching device receives the image from each camera device, it needs to perform the following processing on the image: projection plane unification, feature extraction, feature matching, image registration and fusion, exposure compensation, deghosting, and finally Gets a panorama texture that covers the outer surface of a hemispherical model.

Wherein, the consistent processing of projection planes includes: using homography transformation between plane projections to convert images under different projection planes captured by each photographing device into the same plane.

Wherein, the feature extraction includes: performing feature point detection and extraction on each image, for example, Scale-invariant feature transform (SIFT) method can be used to extract Sift feature points of each image, and each image is performed several times Continuously filter to obtain the first scale group image, reduce the width and height of the original image by half, and then perform several times of continuous filtering to obtain the second scale group image, and repeat this process until the image width and height are less than or equal to the set threshold; The Gaussian images in each scale are differentiated to form a Gaussian difference scale group image; the local extreme points of these Gaussian difference scale group images are solved and calculated, which are the Sift feature points. Among them, the Sift feature point is the local feature of the image, which has good invariance to translation, rotation, scaling, brightness change, occlusion and noise, and also maintains a certain degree of stability to visual changes and affine transformations.

Feature matching includes: Since the images captured by adjacent wide-angle cameras partially overlap with each other (high local overlap), for two images with overlapping areas, first use RANSAC (an acronym for RANdom Sample Consensus, which is based on a set of abnormal The sample data set of the data, calculate the mathematical model parameters of the data, and obtain the effective sample data algorithm) algorithm to filter the feature points, and then use Sift/KD-Tree (abbreviation of k-dimensional tree, which is a kind of segmentation k-dimensional data space data structure) to establish the matching feature point pairs between two images.

Image registration and fusion include: image registration is to obtain image space coordinate transformation parameters through matching feature point pairs; finally image registration is performed by coordinate transformation parameters; image fusion can use non-multi-resolution technology or multi-resolution technology to improve Utilization of image information, improving the accuracy and reliability of computer interpretation, and improving the spatial resolution and spectral resolution of images.

Exposure compensation and deghosting include: According to the geometric principle of optical perspective, there will always be a certain amount of parallax in the images taken by two different camera devices, such as ghosting and ghosting, which will eventually lead to visually unacceptable spliced images. Therefore, the image needs to be deghosted; in addition, because the cameras have different shooting angles, for example, some are located on the light-facing side, and some are on the back-light side, resulting in inconsistencies in the exposure of the captured images. Therefore, in the embodiment of the present invention, exposure compensation and deghosting processing need to be performed on each image, wherein the exposure compensation can adjust the exposure of the image according to the target histogram curve.

The image stitching device repeats the above steps to stitch the images captured by each shooting device into a panorama.

103. Map the spliced panoramic image onto the surface of the corresponding virtual hemispherical surface model to display the three-dimensional panoramic image.

The virtual hemispherical model in the embodiment of the present invention is a model that the virtual reality device simulates the hemispherical surface in reality through scientific and technological simulation and then superimposed on the virtual reality world to be perceived by the user. way of seeing the world. Therefore, there is a position mapping relationship between the virtual hemispherical model and the hemispherical surface.

In the embodiment of the present invention, in order to improve transmission efficiency and reduce network transmission resources, after the image stitching device stitches the images captured by each shooting device into a panorama, it can perform lightweight lossless compression on the stitched panorama and then pass it through the network to a virtual reality device.

The virtual reality device first decompresses the received panorama, and then maps the decompressed panorama to the outer surface of the virtual hemispherical model to form a three-dimensional panorama based on virtual reality.

In the embodiment of the present invention, since each photographing device simultaneously shoots the center direction of the hemisphere from different angles from the outside to the inside, it can realize real-time display of the three-dimensional panorama corresponding to the user's viewing angle when the user circles around the center direction of the hemisphere view, so that the user can change the observation position at will, which greatly increases the scope of the user's observation activity area and improves the user experience.

Specifically, when the user circles around the central direction of the hemisphere, the user's position and viewing angle information are obtained, and the images on the hemispherical model at the corresponding position and viewing angle are displayed as three-dimensional images according to the user's position and viewing angle information.

When the viewpoint height adjustment instruction is triggered, the corresponding viewpoint height position is raised, and the image on the hemispherical model under the corresponding viewpoint height, horizontal positioning, and viewing angle direction is displayed as a three-dimensional image; wherein, the triggering conditions of the viewpoint height adjustment instruction include But not limited to the following examples: when it is detected that the button on the virtual reality device that triggers the height adjustment instruction of the viewpoint is triggered, the height adjustment instruction of the viewpoint is triggered; When the number of times is set, trigger the viewpoint height adjustment command.

When the user approaches the center of the hemisphere, the resolution of the displayed 3D image is increased by a preset multiple to ensure that the displayed 3D image is clear and undistorted. As shown in Figure 2, when the user approaches the center point of the scene along the radius, in order to allow the virtual reality device to display the three-dimensional image of the virtual reality with high resolution and detail under different depths of field, the center of the scene at the user's point of view When the point distance is shortened, the resolution of 1080P will be increased to a higher resolution. For example, the resolution can be increased by a specified multiple through a neural network. The specified multiple is generally set to 2-4 times. The closer the viewpoint is to the center of the scene, the higher the multiple. The higher the value, the clearer and undistorted images are guaranteed.

It should be noted that, in the embodiment of the present invention, the virtual reality device is provided with a sensor for tracking the user's head to determine the position and posture of the user's head in the real world, so as to determine the viewing angle of the virtual world.

In the embodiment of the present invention, a plurality of photographing devices arranged symmetrically and evenly on the hemispherical surface in a spherical position relationship with each other simultaneously take pictures of the center direction of the hemisphere, splicing the images captured by the multiple photographing devices into a panorama and sending it to the virtual reality The device maps the panorama to the surface of the corresponding virtual hemispherical model to display the three-dimensional panorama. Since the images captured by each camera device partially overlap with each other (high degree of local overlap), the density of image coverage is high, and full-angle panoramic image coverage can be achieved, so that when partially overlapping images are spliced into a 3D panoramic image, the effect is smoother , which is closer to the real scene. Further, since each photographing device simultaneously takes pictures of the central direction of the hemisphere from different angles from the outside to the inside, it is possible to display the three-dimensional panoramic view corresponding to the user's viewing angle in real time when the user circles around the central direction of the hemisphere, so that the user The observation position can be changed at will, which greatly increases the scope of the user's observation activity area and improves user experience.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it can still be Modifications are made to the technical solutions described in the foregoing embodiments, or equivalent replacements are made to some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims (10)

1. a kind of three-dimensional panoramic image sets up system, it is characterised in that including：

Multiple camera arrangements, are symmetrically evenly arranged on hemisphere face each other in spherical surface position relation, in hemispheric Heart direction is taken pictures, wherein, the image that each adjacent camera arrangement of locus is shot is partially overlapped by each other；

Image splicing device, is connected with multiple camera arrangements respectively, receives the image that multiple camera arrangements are sent, and clap multiple The image mosaic sent according to device is into panorama sketch, and the panorama sketch is the panorama sketch of hemispheric center position；

Virtual reality device, is connected with image splicing device, the panorama sketch after image splicing device splicing is received, by panorama Figure be mapped to corresponding virtual semi-sphere surface model in appearance carry out three-dimensional panorama figure display, the virtual semi-sphere surface model with Hemisphere face correspondence.

2. system according to claim 1, it is characterised in that：

Hemisphere face is divided into some by sphere dimension line and sphere meridian, and each part is by adjacent meridian and dimension Line is intersecting to obtain at least three intersection point, and the multiple camera arrangement includes the wide angle cameras located at each point of intersection.

3. system according to claim 1, it is characterised in that also include：

Image compressing device, processing, described image compression dress are compressed by the panorama sketch after image splicing device splicing Put and be arranged in image splicing device, or be placed on image splicing device and be connected with image splicing device；

Image decompression device, the panorama sketch after handling image compressing device compression carries out decompression processing, and by after decompression processing Panorama sketch be sent to virtual reality device carry out textures show that described image decompression device can be arranged in virtual reality device In.

4. system according to claim 1, it is characterised in that：

Image splicing device, using the homography conversion between plane projection, the different projections that multiple camera arrangements are sent are flat Image under face is transformed under approximately the same plane, and characteristic point monitoring is carried out to each image after projection plane unification and is extracted, For there are the two of overlapping region images to carry out the matching screening of characteristic point, having between two images of overlapping region is set up With point pair, image registration and fusion, exposure compensating are carried out to the image after Feature Points Matching and goes ghost to handle, is taken pictures multiple The image mosaic that device is shot is into panorama sketch.

5. system according to claim 1, it is characterised in that：

Virtual reality device, when user is gone in ring around hemispheric center position, obtains position and the Viewing-angle information of user, root The image on the hemisphere surface model under correspondence position and visual angle is carried out into 3-D view according to the position and Viewing-angle information of user to show； When triggering viewpoint height adjust instruction, raise corresponding viewpoint height position, by corresponding viewpoint height, horizontal location and Image on hemisphere surface model under view directions carries out 3-D view and shown；When user is close to hemispheric center position When, by the default multiple of the increase resolution of the 3-D view of display, it is ensured that the 3-D view of display is clearly undistorted.

6.A kind of method for building up of three-dimensional panoramic image, it is characterised in that including：

The image that multiple camera arrangements are shot simultaneously towards hemispheric center position is obtained, wherein, the multiple camera arrangement Symmetrically it is evenly arranged in each other in spherical surface position relation on hemisphere face, the image that each adjacent camera arrangement of locus is shot It is partially overlapped by each other；

The image mosaic that multiple camera arrangements are shot is into panorama sketch, and the panorama sketch is the panorama of hemispheric center position Figure；

The aobvious of three-dimensional panorama figure is carried out in appearance by what the panorama sketch after splicing was mapped to corresponding virtual semi-sphere surface model Show, the virtual semi-sphere surface model is corresponding with hemisphere face.

7. method according to claim 6, it is characterised in that：

Hemisphere face is divided into some by sphere dimension line and sphere meridian, and each part is by adjacent meridian and dimension Line is intersecting to obtain at least three intersection point, and the multiple camera arrangement includes the wide angle cameras located at each point of intersection.

8. method according to claim 6, it is characterised in that also include：

Using the homography conversion between plane projection, the image conversion under the different projection planes that multiple camera arrangements are sent To under approximately the same plane；

Carry out characteristic point monitoring to each image after projection plane unification to extract, for there are the two of overlapping region images to enter The matching screening of row characteristic point, sets up the matching double points having between two images of overlapping region；

Image registration and fusion, exposure compensating are carried out to the image after Feature Points Matching and goes ghost to handle, multiple images are spelled It is connected into panorama sketch.

9. method according to claim 6, it is characterised in that also include：

When user is gone in ring around hemispheric center position, position and the Viewing-angle information of user is obtained, according to the position of user The image on the hemisphere surface model under correspondence position and visual angle is carried out into 3-D view with Viewing-angle information to show；

When triggering viewpoint height adjust instruction, corresponding viewpoint height position is raised, by corresponding viewpoint height, horizontal location And the image on the hemisphere surface model under view directions carries out 3-D view and shown；

When user is close to hemispheric center position, the default multiple of the increase resolution of the 3-D view of display is protected The 3-D view for demonstrate,proving display is clearly undistorted.

10. method according to claim 9, it is characterised in that also include：

When the button for monitoring to be located at triggering viewpoint height adjust instruction on virtual reality device is triggered, viewpoint height is triggered Adjust instruction；Or

When monitor the user for wearing virtual reality device continuously jump triggering viewpoint height adjust instruction preset times when, touch Send out viewpoint height adjust instruction.