0
$\begingroup$

I'm doing a study on surfaces and I would like to describe all of the surfaces that I expect to encounter in the field and determine which ones will be the most difficult to detect. To do this, I need to determine what attributes in a surface I need to consider. I have no experience in optics so I'm asking this question here in hopes that someone could save me from overlooking important factors.

$\endgroup$
1
  • $\begingroup$ I think you need to add some of what you wrote in the comment to Carl's answer in your question, to give your question more detail. From that comment, the light sensor you use will not be the main issue. Your problem will be analyzing/interpreting the data from the sensor. Some things you may need to consider will be contrasts of light, shadows & edge effects produced by 3D surfaces, like jagged cliff faces or buildings. There will also be the issue of dealing with glare - you may may need to use a polarization filter. $\endgroup$ – Fred May 3 '18 at 3:33
2
$\begingroup$

Surface roughness, absorptivity, transmittivity, absorption spectrum, etc might be some attributes that may help you.

$\endgroup$
1
$\begingroup$

Your question lacks important parameters: in addition to the qualities of the surface or material that Pranav mentions, you need to be able to define what "detect" means, what sort of light sensors you think you want to use, what the interfering noise sources are, and more.
If your target surface is not active (like a lightbulb, generating its own output) then you need to know as a minimum what the illuminating sources are. Then you can calculate, based on surface qualities such as BRDF the expected signal magnitude at your sensor location. But there's a huge difference between saying "Look out! you're going to hit that thing!" and saying "Hey, there's a chunk of rock with 20% silicates and 1% manganese at about 240 Kelvins over there"

$\endgroup$
2
  • $\begingroup$ I'm unsure yet about what types of light sensors I will use--that is one of the main purposes of this study. I would say that the surfaces I am looking at are not active, but there may be a case when there are active surfaces--I am aiming to detect cliffs within safe stopping distance for mobile robots. My application is therefore closer to the former, "Look out! You're going to hit that thing!". At a higher level, my goal is to improve safety. $\endgroup$ – Klik May 2 '18 at 19:38
  • $\begingroup$ @Klik If you're using it for collision detection you could perhaps augment your system with sonar as well to at least help detect outliers missed by your primary system $\endgroup$ – ChP May 4 '18 at 12:24

Not the answer you're looking for? Browse other questions tagged or ask your own question.