Benefits of a calibration chart
A calibration chart helps to identify and correct these errors and deviations. It performs several functions at the same time and ensures that image quality and evaluation accuracy are improved.
More details on (re)calibration:
Before starting to develop a software application, the initial image must be as PERFECT as possible.
Creating an undistorted image
A suitable encoder configuration and appropriate multiplier/divider settings on the camera's line trigger input must first be used to produce a distortion-free image.
Do not rely on your eye, but measure the shape of circles or squares in the X and Y directions, for example using a calibration grid. This is the only way to be sure that you are capturing distortion-free images of good quality.
Wrong encoder settings
Vertical stripes and light gradient effects of the lighting
In any case, the first step should be to ensure that the lighting accurately covers the image area to be captured. Tips on lighting alignment can be found here.
Once the image composition in the X and Y directions and the image brightness are approximately correct, the next step is fine tuning. In our image above, you can still see large variations in brightness in the X direction.
To compensate for these inhomogeneities, every major line scan camera manufacturer supplies a ready-made software interface for calibrating the camera.
Calibration is done in two or three steps:
- Dark image calibration: Simply place the cover on the lens - done. The different dark currents of each pixel are determined and calibrated. Each unexposed pixel should therefore produce the same black value after calibration.
- Sensors show a non-uniform brightness distribution in the two-dimensional grey image. In line scan cameras, this can be seen along the line and results in image stripes. This can also be calibrated out. (Of course, dirt on the sensor can also cause the same streaks).
- Bright image calibration is used to correct for variations in illumination and the different light sensitivity of the pixels: To do this, set the image brightness to a value of 50-75% on a homogeneous reference surface (paper), defocus the optics and then calibrate.
(The calibration steps are called PRNU, FPU, DNU: Dark Signal Non Uniformity, Fixed Pattern Noise, Photo Response Non Uniformity).
Missing camera sensor calibration
Any vertical streaks in the image should have disappeared and the image should appear smooth. Make any final fine adjustments to the camera's exposure time, gain and lens aperture and refocus the lens.
Any vertical streaks in the image should have disappeared and the image should appear smooth. Make any final fine adjustments to the camera's exposure time, gain and lens aperture, and refocus the lens.
Horizontal stripes
Horizontal streaks are also rare. These can have several causes:
- Over-triggering of the camera due to excessively fast encoder signals (see Encoder for details).
- Fluctuating brightness due to the effects of ceiling lighting (50 Hz flicker). This can be remedied by increasing line lighting or suppressing the effects of extraneous light by shielding, etc.
- Flicker caused by unstabilised, poorly filtered or undersized power supplies for lighting. This can only be remedied by better hardware.
External influence
Calibrating a colour line scan camera
Any colour camera, including line scan cameras, needs to be white balanced to achieve neutral colours. Due to the colour temperature of the white light used (spectra of different white LEDs), the distribution of red, green and blue intensities in the camera image is inhomogeneous. The different spectral sensitivity of the sensor to different wavelengths further exacerbates this effect.
The image and distance intensity diagram of a colour line scan camera shown here shows two typical errors that need to be calibrated:
- Light fall-off at the left and right edges of the image. As with a monochrome camera, this can be compensated for by calibrating the light image.
- The image is not neutral grey, but has a blue and green cast (cyan). The gain of the RBG channels must be set so that the three brightness curves in the diagram above match. With digital colour cameras, this can sometimes be achieved by using different exposure times for each sensor.
- Again, set the image brightness to a value of 50-75% on a homogeneous white reference surface for calibration, defocus the optics, then calibrate and refocus.
Light source and white balance
Need to buy the right camera?
Vision-Doctor.com is a private, independent, non-commercial homepage project and not a technology provider or system integrator. Suitable technologies and further professional support can be obtained from the companies & partners listed below.
If necessary, I will be happy to provide a quick recommendation, contacts and brief information.
