What is the meaning of the vision result from Mech-Vision, and what transformation has been done?
Example: When the coordinates are on the origin pose, the coordinates should be [0,0,0,0,0,0,0], if an object is placed 1000mm directly below the camera field of view, then its coordinates should be changed to [0,0,1000,0,0,0,0]
So what transformation has been done internally or how to get the coordinates before this transformation?
I need to fix the camera above without any calibration. If I define a coordinate system on the ground, all the coordinates of the objects captured by the camera should be converted to the coordinate system on the ground. What should I do?
1.The coordinate composition given by the camera is [translation value + quaternion], which is used to represent the attitude of the object in the camera coordinate system, and the translation unit is m.
The origin of the camera coordinate system is [0,0,0,1,0,0,0], and [1,0,0,0] represents the rotation unit matrix. If an object is placed 1000mm directly below the camera, when only the position is considered, it can be represented as the coordinates [0,0,1,1,0,0,0] in the camera’s field of view. The calculation of the relative pose uses rigid body transformation, but the visual algorithm used to obtain this rigid body pose is not necessarily the same, and varies according to the scene.
2. If the coordinates of the object captured by the camera is A and the ground coordinate system is B in the camera field of view, then the object’s coordinates in the camera field of view is B.inverse()*A. This transformation can be calculated by pose transformation and pose inversion in Mech-Vision, and the transformation method in the software is as follows
for the use of pose transformation, please refer to:
Pose Transformation—Software Suite Documentation
3. If the coordinates of the object captured by the camera is A, and the ground coordinate system is the robot base coordinate system, hand-eye calibration is required. It is necessary to calibrate the relationship between the camera and the coordinate system (extrinsic parameters), and then use the pose transformation for processing. For the part of the hand-eye calibration principle, please refer to:
Calibration Principles - Software Suite Documentation
