LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

image subtraction on intensity graph

Hi i'm a new user of Labview,

i have a problem with my 3d sensor viewing of IFM company.

I wanto to doing a background subtraction,on the image. The output format of my sensor is a 2d array displaying on intensity graph.

I would to take the image and doing a background suppression:

i have solved this type of problem saving an image of background and doing a comparison between z axis min of the sensor and the real image.

For giving more complete discussion i upload my file..

 

The case cycle saving the image of the background when is true.

When is false,compare the image roi displaying by the sensor and the background. This type of problem is solved if the sensor work in plane and is parallel with the plane of the object..

However i have to rotate the sensor because there is a problem of reflection with some object...but i don't know how to solve.

I'm a student

0 Kudos
Message 1 of 21
(3,947 Views)

You did not attach your datafile or the background data, but you are doing things way too complicated.

 

  • In range and coerce can accept a 2D array in the middle and simple scalars for the limits. No need to these stacked FOR loops.
  • To subtract the background, simply subtract the two 2D arrays.
  • Never use "path to string... concatenate string... string to path" to build a path. Use "build path" instead.
  • ...
0 Kudos
Message 2 of 21
(3,936 Views)

ok sorry... this is a vi crated in 2 minutes,so i have not introduce NI RULES of program.

The background suppression in my case is not a simple subtract.

I have just test this type of way and is not good.

the VI that i have created going well when the sensor work in plane with axis orthogonal with the plane of the object..

When we start to use the sensor,we have to start the teach background function. Teach background function acquisit a intensity graph calculating the minimal z quote and assume that this is the cutting plane when we start to considering the object.

So if there are some object far to background,is not considered by the system of acquisition.Instead,if the object is near to sensor there are considered.

I used coerce for determing the real object that i want to measure. The object that i want to measure is get between two limits:

the minimal z quote of the background and the minimal z quote of the object.

 

 

The problem is subtract the background when i'm using the sensor not orthogonal with the plane of the object...For this aspect i have to learn a background that introduce a shift plane trasverse but i have not idea to doing this...

 

example when the sensor is rotated of 45 degrees...

 

THank a lot for the reply

0 Kudos
Message 3 of 21
(3,926 Views)

Initially we teach background and there isn't the object in the roi area..

After the teach program we put the object and we begin the acquisition of the object 

0 Kudos
Message 4 of 21
(3,924 Views)

i put two images for giving more explaination

Download All
0 Kudos
Message 5 of 21
(3,919 Views)

sorry but the request is not solved...

0 Kudos
Message 6 of 21
(3,912 Views)

Can you show us background pictures and pictures with an object for the case that works and for the inclined view.

 

I simply don't understand your terminology. For example, what is the "minimal z quote".

 

Do you simply want to subtract all regions where the image with and without background is very similar.

Message 7 of 21
(3,890 Views)

@altenbach wrote:

Do you simply want to subtract all regions where the image with and without background is very similar.


Here's a quick draft, LabVIEW 2010.

 

 

In your case (image is DBL) you would subtract the two images and check if the difference is within a narrow range close to zero and use that instead of the not equal comparison I am using in the example.

Download All
Message 8 of 21
(3,887 Views)

ok ..my sensor acquisit a matrix of z quote points and the minimal z quote distance is the minimal distance between sensor and object

0 Kudos
Message 9 of 21
(3,879 Views)

i post my data of acquisition if you want to understand what i mean...

i have tested your vi but is not the result that i want

0 Kudos
Message 10 of 21
(3,876 Views)