M
M
mIka012021-02-17 21:39:52
data mining
mIka01, 2021-02-17 21:39:52

What filter should be used for the XYZ plot?

Hello, I have a question about filtering a 3d graph with noise and emissions.
There is an image where you need to know the coordinates of the red dots, but they can be very close.
602d60d905654618753768.jpeg
Or
602d60fbe4231158884931.jpeg

I decided to start with a double array of values ​​for how much each point (pixel) is close to red.
And I got a 3d graph that looks like this in a section.
As an example.
602d616c2357f073448402.jpeg

What filter do I need to apply so that as a result the graph looks like smoothed peaks. So that I could determine the coordinates of the heights from the vertices.
Like something like this.
602d623883a16906824962.png

Thanks in advance for your constructive response.

Answer the question

In order to leave comments, you need to log in

2 answer(s)
G
Griboks, 2021-02-17
@mIka01

moving average

G
Grigory Boev, 2021-02-18
@ProgrammerForever

The simplest thing is to process the neighborhoods of each of the points, and then take either the average, or the average with the extreme values ​​\u200b\u200bthrown out, or something else

Didn't find what you were looking for?

Ask your question

Ask a Question

731 491 924 answers to any question