How to improve color accuracy?

[hpritchett801]

Hi all,

I have been experimenting with the X-NUCLEO-53L1A1 and the NUCLEO-F401RE and here are my results.

Actual distance White Box Black Box Difference

20cm 20.4cm 19.4cm 1cm

45.5cm 46cm 56.7cm 10.7cm

64.1cm 65.6cm 80.8cm 15.2cm

Timing budget:100ms

Distance Mode: Short

I did try to calibrate, but did not seem to make a huge difference.

I did try medium mode but did not seem to change much.

Everything else is left to default in the GUI.

Is there something that I might be missing?

Thanks,

[John E KVAM]

I don't understand what your numbers mean, but I'm going to guess at your problem.

​The best way to understand this is to compare the ranging distance to the actual value.

Different surfaces will return different amounts of photons - giving a different signal rate. If life were perfect it wouldn't matter. But to get the 27 degree Field of View, a multi-mode laser was used. And unfortunately the non-primary lasers turn on slightly after the main beam.

On a highly reflective target, the photons will come mostly from the main beam.

On a less reflective target the photons will come from all the modes. It does throw off the numbers a bit.

To get the best possible answer you could get the signal rate and number of SPADs from the sensor. Signal Rate*Number of Spads = total signal.

Use the total signal to influence your range.

You can keep a couple of look-up tables and decide how much to add or subtract from your range number based on your signal strength.

You can 'know' if the box is black or white and adjust accordingly.

It might get you the accuracy you desire, but it would take a lot of testing.

[hpritchett801]

Thank you for your response. I will try your recommendations.