How to find distance between each person in a large population(any component)?

Hello Everyone,
I am new to this forum :slight_smile: . I am working upon a project where I need the help of finding the distance between people in a large population. Also this is going to be attached to a camera.

I know that for a ultrasonic sensor we can measure distance but for my project I want measure distance between each person in a large population. I was hoping you can help which component is suitable for the job that is Arduino compatible so I can send the data back.

Hope you can help me.

What is Your experience of coding and electronics?
To me it looks like a vision project that needs a powerful pc and a fat wallet.

Railroader:
What is Your experience of coding and electronics?
To me it looks like a vision project that needs a powerful pc and a fat wallet.

I am familiar with using Arduino and have used many components like LCD, RFID, IR Sensors, LED's, Motors and Motor Drivers, Speakers, PIR Sensors, NodeMCU, etc. For this project, I am trying to make it a prototype where one part is we get to know the distance between each person in a large population and this data in real-time is send back to the Arduino where it will visualize it. I was hoping that someone could help me find a component suitable for the job. Also I am in a bit of a hurry.

Being in a hurry is no help for You and not for Your topic here. Overoptimistic or lack of project planning.
How do You intend to know the position of each person? Don't ask for "a sensor". There is no such item.
As I wrote, my suggestion is picture recognizing, lots of memory and calculation power. Nothing that any Arduino would manage.

So what is the solution for this? What do you think has the capability to do this?

To calculate the distance between people, you need to know their positions. Solve that problem first.

Image processing is your only option, and an Arduino is not suitable.

Can't we use image recognition or thermal cameras to get the position of the person? If we were to use this, then what microprocesser has the capability to do this. Also , I am making a prototype so it would be some toothpicks or something else or such so we can't such thermal sensors on it

See replies #1, #3 and #5

Stop the hand-waving.

Put numbers and dimensions on the problem specification.

what microprocesser has the capability to do this? Also I am making a prototype as in such it need not be real people but some toothpicks or something else as such.

The distance between people will be true for only a very short instance of time. What part of people do you measure to determine the distance apart? How do you differentiate a single large person from a parent holding a child? etc, etc,,,,,

Paul

I saw a tutorial on just this subject recently.

It uses a Raspberry Pi4 and OpenCV.

Please let us know if this works for your needs.

what microprocesser has the capability to do this?

First, figure out what you are actually going to do.

Second, figure out how you are going to do it.

Then pick a computer.

Hi,
If you are looking at a scattered population, you will need to be above the population and looking down to see the separation.
Then you would need to identify the location of each member of the population to calculate their INDIVIDUAL spacing between adjacent subjects.

Can you please tell us your electronics, programming, arduino, hardware experience?

Tom... :slight_smile:

tejas_78910:
I am working upon a project where I need the help of finding the distance between people in a large population.

What do you mean by "large population" - 10, 100 or 1000 people?

What do you mean by "distance between people"? Do you mean the distances between every single person and his/her nearest neighbour? That would be a lot of measurements even in a population of 10 people.

OR do you just want to know the average number of people per square metre?

OR what?

It would make it much easier to make useful suggestions if you describe the project you want to create, including why you need to know the distance between people.

...R

An interesting problem which I feel is wildly beyond the scope of an arduino. (OR A PI)

Lets make it easy and suppose you are assessing the mean population density in a sensibly closed environment. (ie with few entrance/exit points)

Then you can evaluate the number inside by sensing who is going in and out. Seperate the exits from the entrances and you can count them with a sensor.

The mean pop density is then number inside / floor area.

Of course that doesnt tell you how they are clumping once inside.

Perhaps you could use a thermal imaging camera and software (on a PC) to assess that.

So.. looking for population density of say a large room?

I'd use temperature. Look at ambient temp and then use a temp to image scan of the room looking for hot spots over a calculated limit.

Calculated would be a function of ambient. Tighter packed clumps will be hotter clumps.

-jim lee