| Japanese |

Camera Vector Image

Camera Vector Image

CVImage

Camera Vector Image

The feature of our technology is about the acquisition of the camera position from the automatic, highly accurate all-surrounding video image. We call all surrounding video image as 4π video image, since the possession of not only the horizontal direction of 360 degrees but also the elevation and depression angle in the image. Moreover, from Image Processing of this 4π video image we can acquire 3 dimensional information, and we call this 3D video image the CV video image.

CV stands for camera vector. The operation by which this camera vector is calculated is called CV operation. Moreover, the CV values means the six free degrees of the shooting camera that form the camera vector, the three dimensional positions of each frame of the image refer to the three dimensional posture (CV value).

Though CV value can be acquired from GPS and IMU, our company mathematically analyzes the position and the posture of the camera using 4π video image, and obtain the CV value with more accuracy. 4π video image indicates the direction from the center of the camera to each pixel coordinates of the image. It has an excellent feature that not only can process within the range of the linear algebra of a classic triangulation, but also can acquire a characteristic necessary for the operation for all directions that improves operation accuracy.

In addition, it is unnecessary to use many cameras, only a single camera is enough to obtain CV image. The CV operation has an excellent feature that the device is extremely simplified so that it is possible to process fundamentally with a single camera.

Three dimensional data of camera position (X,Y,Z) and posture (θx, θy, θz) corresponding to each frame of the image are requested by this method through out all frames of the image. It is added to the image data. And the new concept of CV image, that is all surrounding image with CV value is acquired. When each frame of the image had the CV value, the image can become three dimensional at any time. Therefore the CV image can be referred to as 2.5 dimension image. Since the camera position and the posture in all frames become already known, it is possible to read three dimensional coordinates and three dimensional shape of the object which is caught from a wide ranging viewpoint by the camera. Moreover, the image can be treated just like CG. So various processes of image become possible which is an excellent feature.

: Click to zoom

: Click to zoom

Go to top
Shooting Process

Shooting Process

Shooting Process

Shooting Process

The unique feature of our company is that it is possible to take video using only one camera and one PC. The expensive devices like GPS or IMU is not required to use our three dimensional technology.

This all surrounding camera has six CCD lens.

After setting the camera up on the car, Shooting are started.

PC's setting up is very easy.

Shooting scenery.

While confirming the screen image, the setting is adjusted in realtime.

The shooting staff is taking video along with confirming all situations such as the weathers, road slope, etc.

Go to top
All Surrounding Image

All Surrounding Image

Complete All Surrounding Image

All Surrounding Image

All surrounding image are joined 6 image precisely after adjusting lens distortion developed by us for jointing precisely. And we are providing complete all surrounding image which don't have any dead area, since we can process automatically the image of the area from the other video frames.


: Click to zoom

: Click to zoom

Go to top
Stop Vibration

Stop Vibration

Stop Vibration (Image Stabilization)

Stop Vibration

All surrounding image are joined 6 image precisely after adjusting lens distortion developed by us for jointing precisely. And we are providing complete all surrounding image which don't have any dead area, since we can process automatically the image of the area from the other video frames.

In the CV image, three dimensional position (X, Y, Z) of each frame or more and three dimensional posture (θx, θy, θz) are already known in the CV value. Then, the image display without the rotation element became possible because each shake correction element (-Δθx, -Δθy, -Δθz) of each axis of coordinate was reversely rotated by using three dimensional posture (θx, θy, θz) value in the CV value and it displayed the perspective.

: Click to zoom

Go to top