Measuring the intensity of fog using a light source, pixy cam, Arduino

Hi guys, this is my first time in this forum. I'm currently working on a project to dissipate fog; in this project, I'm looking for a method on how I can "measure" how thick or dense the fog is. For that, the idea I came up with that fits my setup, is to use a camera and focus it on a single light source and measure its intensity. So for example, if there is no fog at all, the intensity would bee 100%, and as fog forms, the intensity of the light (read by the camera) would decrease further and further as the fog thickens.
I need to be able to do this even if I have other light sources around, the camera must focus on one and only one light source (LED maybe), and give me the intensity in %. Any idea on how I can accomplish this? I'm currently looking into the Pixycam, but I'm unsure how it can be done. Can someone guide me on this please. Thanks

Pixy is optimised to recognise objects; I can't think why you would consider it for this application.

I've never tried such a thing, but I imagine you'd be looking for a sensor with a much larger dynamic range than a simple camera, say in the range 12 to 16 bits.

You probably would also want a beam splitter and reflector arrangement, so that you can continuously measure the source intensity, so that you can null-out source fluctuations

Most people use light scattering to measure the density of small particles in air.

A light beam (from a laser diode) passes through the sample in an otherwise light tight chamber, and the intensity of light scattered at right angles to the beam is measured with a photodiode.

Hi, thanks for the reply.

I'm not intending to specifically use the pixy cam, I just need to make a setup that works, especially for long ranges. I have no idea how can I measure the intensity of a specific light source with something other than a camera. My project actually involves drones, and I want the drone to be able to measure visibility through fog. So, I was thinking if I could use the pixy cam to detect a light through fog to measure the visibility. I don't know if there are better approaches, if there are please guide me. Thanks

jremington:
Most people use light scattering to measure the density of small particles in air.

A light beam (from a laser diode) passes through the sample in an otherwise light tight chamber, and the intensity of light scattered at right angles to the beam is measured with a photodiode.

Hi, thanks for the insight.

The issue with the photodiode or LDR is that it measures light from all sources. I need to measure the intensity of a specific light to translate the intensity into electric signals to give a scale of visibility. Also, since my project involves drones, the setup cannot be designed for a light tight chamber; it should work in the open space, under the sun or the moon.
Any ideas? thanks

The issue with the photodiode or LDR is that it measures light from all sources.

So does a camera.

With a little thought and effort, you can force fog through a scattering chamber containing a light source and photodiode, as everyone else does.

"I need to be able to do this even if I have other light sources around, the camera must focus on one and only one light source (LED maybe), "

Have the cam sight thru a paper towel tube so other light sources are not in its field of view.

What does this have to do with drones, and how do you imagine making measurements with the setup?

zoomkat:
"I need to be able to do this even if I have other light sources around, the camera must focus on one and only one light source (LED maybe), "

Have the cam sight thru a paper towel tube so other light sources are not in its field of view.

I can't do that, I need the camera to be able to focus on one light source only and ignoring the others. The lights can be colored to make this easier.

I suspect you're going to be disappointed and frustrated if you persist with the Pixy route.

jremington:
So does a camera.

With a little thought and effort, you can force fog through a scattering chamber containing a light source and photodiode, as everyone else does.

TheMemberFormerlyKnownAsAWOL:
I suspect you're going to be disappointed and frustrated if you persist with the Pixy route.

jremington:
What does this have to do with drones, and how do you imagine making measurements with the setup?

To clarify things, my project is mainly a drone that needs to be able to "measure" visibility in fog, and based on that measurement It would spray a solution that dissipates fog. At the time being, I'm trying to simulate this system in a closed chamber, where I have a fog machine to generate fog. When I used my eye to comment on visibility, my supervisor was not happy because this is not an engineering way of doing things. So, I decided I would design a system that would measure fog by its self and give the order to the pump unit to spray the solution to dissipate fog. The reason I can't use a photodiode is because I'm designing this system for the drone, and the drone flies in the open spaces, and under sun light. Hence, a photodiode wouldn't be able to give me correct measurements because it would collect light from all sources it can. Thus, I need to have a camera, that can focus on a single spot (LED perhaps), and from a distance. It measures the intensity of that light source and translates it into a percentage of visibility, and when the visibility drops under a certain level, the drone would start spraying the solution to dissipate fog. I'm targeting airport runways if that would give you a better idea.

Hence, a photodiode wouldn't be able to give me correct measurements because it would collect light from all sources it can. Thus, I need to have a camera, that can focus on a single spot (LED perhaps), and from a distance

It looks to me like you're thinking a photodiode can't be used with optics.

This is not an engineering way of thinking.

TheMemberFormerlyKnownAsAWOL:
It looks to me like you're thinking a photodiode can't be used with optics.

This is not an engineering way of thinking.

Dear sir, I apologize for my lack of knowledge, but I don't know much about a photodiode. I'm still a student and that's why I'm asking for guidance here. If you have an idea of how can a photodiode accomplish the task I have at hands, please guide me. Thank you.

Fresnel lens over the emitter is a good place to start. A 'cover' over the receiver to lesson the reception of scattered light or, perhaps, even a Fresnel lens over the receiver as well.
Or:


You can use a TFMini and the signal strength reading to detect light thingy. No need for a remote detection/light source.

Please realize that to someone who DOES understand photodiodes and how they are used, this argument is complete nonsense:

The reason I can't use a photodiode is because I'm designing this system for the drone, and the drone flies in the open spaces, and under sun light. Hence, a photodiode wouldn't be able to give me correct measurements because it would collect light from all sources it can.

When asking for advice on a technical forum, do some research and thinking before responding, otherwise people will assume that you are an idiot and and a waste of time.

Note also that use of the phrase "I'm designing" implies that you have relevant training and skills in the appropriate design area, which is obviously not the case.

jremington:
Please realize that to someone who DOES understand photodiodes and how they are used, this argument is complete nonsense:

When asking for advice on a technical forum, do some research and thinking before responding, otherwise people will assume that you are an idiot and and a waste of time.

I don't know if I'm not asking my question properly or not. For the case of a photodiode, I need to have a beam of light directed at my diode, which means the light and the diode should be stationary. My drone flies over a number of light sources, and they are not directed at the drone, like the lights available on a runway! That is why I'm asking how to do it with a camera, not a photodiode. Because, I know how does a photodiode work, and it would not do the job I need it to do!

For the case of a photodiode, I need to have a beam of light directed at my diode, which means the light and the diode should be stationary

And how is this different for a camera?

I don't know if I'm not asking my question properly or not.

You haven't thought through the project completely, which makes it impossible to ask good questions.

It sounds like your advisor is not much help.

"Thus, I need to have a camera, that can focus on a single spot (LED perhaps), and from a distance. It measures the intensity of that light source and translates it into a percentage of visibility, and when the visibility drops under a certain level, the drone would start spraying the solution to dissipate fog. I'm targeting airport runways if that would give you a better idea."

Well, most large airports already have installed instruments that measure "cloud ceiling" and "visibility". You might take a look at how these work. I would assume you would have to set your remote light source as a beacon that your drone could identify, and then do its monitoring.

TheMemberFormerlyKnownAsAWOL:
And how is this different for a camera?

The main difference I believe exists, the photodiode needs a light beam to be concentrated on it. But for the camera, the light does not need to be directed at the camera, whereas the camera will identify it as a colorful bright spot through the fog; and I need to measure the intensity of this bright spot. So to emphasize, I don't want a light like a laser directed at my camera or photodiode, I just want to measure the intensity of a (colored) bright spot, whether it's directed at the camera or not. Hope that makes things clearer

The key idea here Is that the camera should focus on the light, not the light focusing on the camera/photodiode.