Go Down

Topic: Detection of flying tiny Objects (Read 468 times) previous topic - next topic

MooShaMee

Dear All, I am working on a Project which is related to Monitoring of small Garden approx.  2 - 5 Meters. I  wants to detects small flying objects like butterflies / Moths and other small flying pests with the help of Sensor.

Kindly anyone help me out to find out the appropriate sensor.

pylon

Quote
Kindly anyone help me out to find out the appropriate sensor.
A camera is probably best for this. To use it in the night you need infrared illumination to still see the insects.

MooShaMee


Dear , thanks for your reply but I have to use Sensor for this purpose ....  let me know any type of sensor which can detect such objects...

mikb55

If you are just counting insects, then a laser beam break sensor should work.

Do you need to know the insect type? If so then that is a much more complex problem that will require image processing on a fast computer.

MooShaMee

Dear ,

Thank you for your email. Yes I just wanted to count the insect around a plant which we want to monitor not Identification ... kindly refer which Laser Sensor is suitable for this purpose ...

jremington

A camera and a PC with image processing software is your only option.

pylon

Quote
Dear , thanks for your reply but I have to use Sensor for this purpose ....  let me know any type of sensor which can detect such objects...
A camera is a sensor, an optical sensor, it's even a multi-sensor as it has many sensors (thousands to millions) in one case.

A laser beam sensor might help if you build a case around your plant and open only a small hole. You can install the laser beam there and detect the insects that are lucky or intelligent enough to find the hole.

You may also try a lidar sensor (also laser based) but a lidar with such a resolution is quite expensive. As you didn't mention your budget I don't know if this is an option.

In any case: the solution is probably not Arduino based as you need much more processing capabilities and memory than the Arduino is able to offer.

Paul_KD7HB

A laser beam may chase the insects away.

Paul

MarkT

To my knowledge insect eyes won't detect IR, they only see blue/UV, which is why flowers are often yellow (which is
very black to insect eyes).  So IR laser detector shouldn't affect the insects, but then again most insects will
miss the beam.

CV looking at plants will have a massive problem telling plant motion from insect motion on a breezy
day, you'd need to find a CV expert to give hints and tips for this.

Larger insects will be detectable by ultrasound (bats found this out long ago), so short-range ultrasound
is a possible approach (you want to range limit to mask out the background).

Another thing to consider is speckle-pattern sensing (as in the Kinect).
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

Grumpy_Mike

#9
Jan 20, 2019, 12:02 pm Last Edit: Jan 20, 2019, 12:02 pm by Grumpy_Mike
but I have to use Sensor for this purpose
That normally means this is some sort of assignment you have been given. Is that correct?
If so is your assignment to actually build such a system? It seems to me as an Ex accademic that the assignment is more of a research project into why this can not be done simply. Or that the person setting the assignment has no idea if it is feasible.

wvmarle

To my knowledge insect eyes won't detect IR, they only see blue/UV, which is why flowers are often yellow (which is very black to insect eyes).
It's more like: insects can ALSO see (part of the) UV spectrum, on top of the regular (for humans) visible spectrum. Making a flower look black is not really attractive... and flowers are usually meant to attract insects. Of course insect eyes work rather different than our eyes, being facet eyes and all, so it's hard to make direct comparisons.
Quality of answers is related to the quality of questions. Good questions will get good answers. Useless answers are a sign of a poor question.

MooShaMee

That normally means this is some sort of assignment you have been given. Is that correct?
If so is your assignment to actually build such a system? It seems to me as an Ex accademic that the assignment is more of a research project into why this can not be done simply. Or that the person setting the assignment has no idea if it is feasible.
Yes!! this is project ..... need help to be done even not simple ... you mean it is not feasible idea ???

mattlogue

Only a very mascochistic instructor would assign such a project. It is no way a simple affair to detect dainty flying things like butterflies outside AI-based image recognition, and I feel you were looking more for a sensor than camera setup. One idea that is more feasible at night is to use a certain pulse of omnidirectional IR light, whose pattern is detected by a IR sensisitive photodiode mounted behind it. It would be blocked from direct exposure to the beacon LED, and white and light-colored to IR bugs would reflect to some degree to the photodiode. You would have to amplify the output from photodiode before the arduiono regardless.
Just because I live in the states don't mean I care

wvmarle

#13
Jan 23, 2019, 07:43 am Last Edit: Jan 23, 2019, 07:45 am by wvmarle
There's so far no indication this project is not possible.
Feasibility, OTOH, depends also on factors such as your skills (the fact you're asking here is a red flag), budget (it will need some high end equipment and you said you don't have much to spend) and time (it's not easy, so unless there's an off the shelf solution you're likely going to need to do a lot of work).
Quality of answers is related to the quality of questions. Good questions will get good answers. Useless answers are a sign of a poor question.

arduin_ologist

#14
Jan 23, 2019, 07:59 am Last Edit: Jan 23, 2019, 08:05 am by arduin_ologist
You started with this:

I  wants to detects small flying objects like butterflies
... which escalated to this:

I just wanted to count the insect around a plant
... which is a whole different kettle of fish butterflies.

This is not like counting people entering and leaving a room through a door where you can log them in and out. The issue here surely (even given a suitable detection sensor) is to keep track of the fact that any distinct insect has been counted once and once only.

That's where some kind of vision would be required I'd say, to tag a counted insect on screen as it zooms around so it doesn't get counted again.

And what happens if it leaves the vicinity of interest and comes back....

You need to fit each insect with a transponder ;)





Go Up