This is documentation of an archived release.
For documentation on the current version, please check Knowledge Base.
For documentation on the current version, please check Knowledge Base.
Calibration of camera
Camera calibration is the process of estimating the internal parameters of the camera that deal with the camera's internal characteristics, such as, its focal length, sensor size, pixel resolution, and image center. Knowing intrinsic parameters is an essential first step for postprocessing your aerial data to high-accurate 3D maps. It removes lens distortion, which degraces accuracy.
Calibration parameters of camera
| Field | Declaration | Units | Necessity |
|---|---|---|---|
| Camera name | Any unique name linked to the position and orientation file | string | required |
| Focal Length | The focal distance of the lens expressed in mm Wikipedia Focal Length | mm decimal (point) notation | required |
| Sensor longtrack | The exact value equal to the number of pixels in the longtrack multiplied by the pixel size | mm decimal (point) notation | required |
| Sensor crosstrack | The exact value equal to the number of pixels in the crosstrack multiplied by the pixel size | required | |
| Sensor pixel size | Physical size of the pixels in CCD camera sensor. | micron decimal (point) notation | required |
| PP Longtrack (Focal Point X) | The principal point value along the sensor longtrack expressed in mm (origin in center of image). Wikipedia Focal Point | mm decimal (point) notation | optional |
| PP Crosstrack (Focal Point Y) | The principal point value along the sensor crosstrack expressed in mm (origin in center of image). | mm decimal (point) notation | optional |
| K1, K2, K3 | Radial distortion parameters, regarding the formulas of D.C. Brown Wikipedia Distortion | decimal (point) notation | optional |
| P1, P2 |
Steps to be taken
- Select a pattern, download, and print
- Mount the pattern onto a rigid flat surface
- Take many pictures of the target at different orientations and distances
- Download pictures to compute and select ones that are in focus
- Use provided examples to automatically detect calibration target and compute parameters
- Move calibration file to a safe location
Tips for a good calibration
- The target needs to be mounted on a flat surface and any warping will decrease calibration accuracy. An ideal surface will be rigid and smooth, for example a table, glass, or marble tile. Cardboard or foam will still work well, but has a tendency to warp over time. Having said that, don't go overboard trying to make a perfectly smooth surface.
- Before you start taking pictures, make sure the camera has a fixed focal length on infinite. The calibration procedure assumes that every picture taken has the same focal length. Using a tripod to stabilize the camera is also a good idea. Motion blur and being out of focus will throw off calibration.
- A quick note on cheap consumer cameras which can apparently change focus without changing their focal length. A zoom lens will always change the focal length when at a different zoom. The easiest way to know if autofocus will change your focal length is to put it in manual mode and calibrate at different focus values.
- Taking a diverse set of in focus image is essential to calibration. Images should be taken at several different orientations, distances, and locations in the image, while filling up as much of the image as possible. Be sure that the calibration target appears along the image edge and the center. If all the pictures are taken in one region then the results will be biased, even if the residual error is low. The whole target needs to be visible in the image and in some cases the target's border also needs to be visible. Also avoid extreme angles or distances when taking photos.
Please contact Orbit GT, if you need a usefull pattern for calibration.
Source & more information can be found at Tutorial Camera Calibration.
Last modified:: 2019/03/25 11:36