Reading the Amplitude of an Ultrasound Echo

Hi guys, I am working on a project that requires a device that could discern between a human and a non-human object.

My idea was to use the [Ultrasound Sensor] and read the magnitude of the return signal. This is going on the idea that different materials would return the magnitude of the signal in different ratios.

I bought [Maxbotix LV-EZ1] and connected the Echo pin to an analogue input pin to get a reading, but it yielded only the times that it receives the ping but not its magnitude.

Should I change my sensor or is there a way to read the amplitude of the echo?

Most ultrasonic sensors don't allow you to do this. The magnitude of the signal is not useful information for most purposes. For example, it will be pretty much impossible to tell the difference between a human wearing clothes and a piece of cloth.

You could try to "hack" an HC-SR04 to read the amplitude of the signal. Reading the following should give you some ideas, but it's definitely not a hand-holding "tutorial" that will lead you down a rosy path: HC-SR04 | David Pilling

And even if you can read the amplitude of the signal, like MorganS, I'm skeptical that doing so will help you achieve your goal in anything other than a laboratory setting....if that.

I'd go for IR.
Humans tend to be warmer than their environment, even when wearing clothes you can normally detect this (or look for the face - it normally shows up really bright in an IR image). Note that an Arduino may not be able to do this due to lack of processing power.

But a simple IR thermometer pointing about the same direction as the ultrasonic will give you the temperature of the object it's pointed at.

Skin temperature usually falls in a pretty small range, so it should not be too hard to distinguish a live person from a cup of coffee, except at that moment in time when the coffee has cooled to skin temperature.

For more fun, check out the Panasonic Grid-Eye sensor.

MorganS:
Most ultrasonic sensors don't allow you to do this. The magnitude of the signal is not useful information for most purposes. For example, it will be pretty much impossible to tell the difference between a human wearing clothes and a piece of cloth.

Oh I see, when I was talking to two of my Teacher's Assistant about this project, they told me any ultrasonic sensors could fulfill this objective.

I see where you're getting at. However, since my target "object" is any article that could let the device know if the object being pinged is a potential human or not. So in this case, if the device could detect cloth, it will be fantastic as I couldn't really think of a situation where there would be cloths hanging around freely.

What I was really trying to distinct is hard surface and soft surface, since I am assuming there wouldn't be any humans walking around wearing an armor.

Thank you for mentioning a specific IR sensor, I will look into that model, should I fail to hack the ultrasound sensor.

DaveEvans:
You could try to "hack" an HC-SR04 to read the amplitude of the signal. Reading the following should give you some ideas, but it's definitely not a hand-holding "tutorial" that will lead you down a rosy path: HC-SR04 | David Pilling

And even if you can read the amplitude of the signal, like MorganS, I'm skeptical that doing so will help you achieve your goal in anything other than a laboratory setting....if that.

Oh thank you for the link! I will visit it right after this. Going down the engineering path, I'd never dare to hope that it will ever be rosy. I really appreciate you finding this link for me.

wvmarle:
I'd go for IR.
Humans tend to be warmer than their environment, even when wearing clothes you can normally detect this (or look for the face - it normally shows up really bright in an IR image). Note that an Arduino may not be able to do this due to lack of processing power.

Yeah, I was considering this option, but I was also not sure if an Arduino would have the necessary processing power to do so. I will look into this if my endeavour at hacking the ultrasound sensor fails. Thank you for the suggestion, I really appreciate it!

jadi001:
the object being pinged is a potential human or not. So in this case, if the device could detect cloth, it will be fantastic as I couldn't really think of a situation where there would be cloths hanging around freely.

You guys don't use curtains and table cloths?

What I was really trying to distinct is hard surface and soft surface, since I am assuming there wouldn't be any humans walking around wearing an armor.

Lycra leggings, or the bare legs of people wearing shorts/miniskirts would probably read as hard object. A hard object with flat surfaces reflecting most of the sound away from it rather than back, may well read as a soft object.

An Arduino can work with the Grid Eye infrared sensor. Overview | Adafruit AMG8833 8x8 Thermal Camera Sensor | Adafruit Learning System

The Panasonic documentation for the Grid Eye includes a human-detection algorithm that could probably be implemented on an Arduino.